git/builtin/submodule--helper.c

3450 строки
96 KiB
C
Исходник Обычный вид История

#define USE_THE_INDEX_COMPATIBILITY_MACROS
#include "builtin.h"
#include "repository.h"
#include "cache.h"
#include "config.h"
#include "parse-options.h"
#include "quote.h"
#include "pathspec.h"
#include "dir.h"
#include "submodule.h"
#include "submodule-config.h"
#include "string-list.h"
#include "run-command.h"
#include "remote.h"
#include "refs.h"
#include "refspec.h"
#include "connect.h"
#include "revision.h"
#include "diffcore.h"
#include "diff.h"
#include "object-store.h"
#include "advice.h"
branch: add --recurse-submodules option for branch creation To improve the submodules UX, we would like to teach Git to handle branches in submodules. Start this process by teaching "git branch" the --recurse-submodules option so that "git branch --recurse-submodules topic" will create the `topic` branch in the superproject and its submodules. Although this commit does not introduce breaking changes, it does not work well with existing --recurse-submodules commands because "git branch --recurse-submodules" writes to the submodule ref store, but most commands only consider the superproject gitlink and ignore the submodule ref store. For example, "git checkout --recurse-submodules" will check out the commits in the superproject gitlinks (and put the submodules in detached HEAD) instead of checking out the submodule branches. Because of this, this commit introduces a new configuration value, `submodule.propagateBranches`. The plan is for Git commands to prioritize submodule ref store information over superproject gitlinks if this value is true. Because "git branch --recurse-submodules" writes to submodule ref stores, for the sake of clarity, it will not function unless this configuration value is set. This commit also includes changes that support working with submodules from a superproject commit because "branch --recurse-submodules" (and future commands) need to read .gitmodules and gitlinks from the superproject commit, but submodules are typically read from the filesystem's .gitmodules and the index's gitlinks. These changes are: * add a submodules_of_tree() helper that gives the relevant information of an in-tree submodule (e.g. path and oid) and initializes the repository * add is_tree_submodule_active() by adding a treeish_name parameter to is_submodule_active() * add the "submoduleNotUpdated" advice to advise users to update the submodules in their trees Incidentally, fix an incorrect usage string that combined the 'list' usage of git branch (-l) with the 'create' usage; this string has been incorrect since its inception, a8dfd5eac4 (Make builtin-branch.c use parse_options., 2007-10-07). Helped-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Glen Choo <chooglen@google.com> Reviewed-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-01-29 03:04:45 +03:00
#include "branch.h"
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
#include "list-objects-filter-options.h"
#define OPT_QUIET (1 << 0)
#define OPT_CACHED (1 << 1)
#define OPT_RECURSIVE (1 << 2)
#define OPT_FORCE (1 << 3)
typedef void (*each_submodule_fn)(const struct cache_entry *list_item,
void *cb_data);
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
static char *repo_get_default_remote(struct repository *repo)
{
char *dest = NULL, *ret;
struct strbuf sb = STRBUF_INIT;
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
struct ref_store *store = get_main_ref_store(repo);
const char *refname = refs_resolve_ref_unsafe(store, "HEAD", 0, NULL,
NULL);
if (!refname)
die(_("No such ref: %s"), "HEAD");
/* detached HEAD */
if (!strcmp(refname, "HEAD"))
return xstrdup("origin");
if (!skip_prefix(refname, "refs/heads/", &refname))
die(_("Expecting a full ref name, got %s"), refname);
strbuf_addf(&sb, "branch.%s.remote", refname);
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
if (repo_config_get_string(repo, sb.buf, &dest))
ret = xstrdup("origin");
else
ret = dest;
strbuf_release(&sb);
return ret;
}
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
static char *get_default_remote_submodule(const char *module_path)
{
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
struct repository subrepo;
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
repo_submodule_init(&subrepo, the_repository, module_path, null_oid());
return repo_get_default_remote(&subrepo);
}
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
static char *get_default_remote(void)
{
return repo_get_default_remote(the_repository);
}
static char *resolve_relative_url(const char *rel_url, const char *up_path, int quiet)
{
char *remoteurl, *resolved_url;
char *remote = get_default_remote();
struct strbuf remotesb = STRBUF_INIT;
strbuf_addf(&remotesb, "remote.%s.url", remote);
if (git_config_get_string(remotesb.buf, &remoteurl)) {
if (!quiet)
warning(_("could not look up configuration '%s'. "
"Assuming this repository is its own "
"authoritative upstream."),
remotesb.buf);
remoteurl = xgetcwd();
}
resolved_url = relative_url(remoteurl, rel_url, up_path);
free(remote);
free(remoteurl);
strbuf_release(&remotesb);
return resolved_url;
}
static int resolve_relative_url_test(int argc, const char **argv, const char *prefix)
{
char *remoteurl, *res;
const char *up_path, *url;
if (argc != 4)
die("resolve-relative-url-test only accepts three arguments: <up_path> <remoteurl> <url>");
up_path = argv[1];
remoteurl = xstrdup(argv[2]);
url = argv[3];
if (!strcmp(up_path, "(null)"))
up_path = NULL;
res = relative_url(remoteurl, url, up_path);
puts(res);
free(res);
free(remoteurl);
return 0;
}
static char *do_get_submodule_displaypath(const char *path,
const char *prefix,
const char *super_prefix)
{
if (prefix && super_prefix) {
BUG("cannot have prefix '%s' and superprefix '%s'",
prefix, super_prefix);
} else if (prefix) {
struct strbuf sb = STRBUF_INIT;
char *displaypath = xstrdup(relative_path(path, prefix, &sb));
strbuf_release(&sb);
return displaypath;
} else if (super_prefix) {
return xstrfmt("%s%s", super_prefix, path);
} else {
return xstrdup(path);
}
}
/* the result should be freed by the caller. */
static char *get_submodule_displaypath(const char *path, const char *prefix)
{
const char *super_prefix = get_super_prefix();
return do_get_submodule_displaypath(path, prefix, super_prefix);
}
static char *compute_rev_name(const char *sub_path, const char* object_id)
{
struct strbuf sb = STRBUF_INIT;
const char ***d;
static const char *describe_bare[] = { NULL };
static const char *describe_tags[] = { "--tags", NULL };
static const char *describe_contains[] = { "--contains", NULL };
static const char *describe_all_always[] = { "--all", "--always", NULL };
static const char **describe_argv[] = { describe_bare, describe_tags,
describe_contains,
describe_all_always, NULL };
for (d = describe_argv; *d; d++) {
struct child_process cp = CHILD_PROCESS_INIT;
prepare_submodule_repo_env(&cp.env);
cp.dir = sub_path;
cp.git_cmd = 1;
cp.no_stderr = 1;
strvec_push(&cp.args, "describe");
strvec_pushv(&cp.args, *d);
strvec_push(&cp.args, object_id);
if (!capture_command(&cp, &sb, 0)) {
strbuf_strip_suffix(&sb, "\n");
return strbuf_detach(&sb, NULL);
}
}
strbuf_release(&sb);
return NULL;
}
struct module_list {
const struct cache_entry **entries;
int alloc, nr;
};
#define MODULE_LIST_INIT { 0 }
static int module_list_compute(int argc, const char **argv,
const char *prefix,
struct pathspec *pathspec,
struct module_list *list)
{
int i, result = 0;
char *ps_matched = NULL;
parse_pathspec(pathspec, 0,
PATHSPEC_PREFER_FULL,
prefix, argv);
if (pathspec->nr)
ps_matched = xcalloc(pathspec->nr, 1);
if (read_cache() < 0)
die(_("index file corrupt"));
for (i = 0; i < active_nr; i++) {
const struct cache_entry *ce = active_cache[i];
if (!match_pathspec(&the_index, pathspec, ce->name, ce_namelen(ce),
0, ps_matched, 1) ||
!S_ISGITLINK(ce->ce_mode))
continue;
ALLOC_GROW(list->entries, list->nr + 1, list->alloc);
list->entries[list->nr++] = ce;
while (i + 1 < active_nr &&
!strcmp(ce->name, active_cache[i + 1]->name))
/*
* Skip entries with the same name in different stages
* to make sure an entry is returned only once.
*/
i++;
}
if (ps_matched && report_path_error(ps_matched, pathspec))
result = -1;
free(ps_matched);
return result;
}
static void module_list_active(struct module_list *list)
{
int i;
struct module_list active_modules = MODULE_LIST_INIT;
for (i = 0; i < list->nr; i++) {
const struct cache_entry *ce = list->entries[i];
if (!is_submodule_active(the_repository, ce->name))
continue;
ALLOC_GROW(active_modules.entries,
active_modules.nr + 1,
active_modules.alloc);
active_modules.entries[active_modules.nr++] = ce;
}
free(list->entries);
*list = active_modules;
}
static char *get_up_path(const char *path)
{
int i;
struct strbuf sb = STRBUF_INIT;
for (i = count_slashes(path); i; i--)
strbuf_addstr(&sb, "../");
/*
* Check if 'path' ends with slash or not
* for having the same output for dir/sub_dir
* and dir/sub_dir/
*/
if (!is_dir_sep(path[strlen(path) - 1]))
strbuf_addstr(&sb, "../");
return strbuf_detach(&sb, NULL);
}
static int module_list(int argc, const char **argv, const char *prefix)
{
int i;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
struct option module_list_options[] = {
OPT_STRING(0, "prefix", &prefix,
N_("path"),
N_("alternative anchor for relative paths")),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule--helper list [--prefix=<path>] [<path>...]"),
NULL
};
argc = parse_options(argc, argv, prefix, module_list_options,
git_submodule_helper_usage, 0);
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
for (i = 0; i < list.nr; i++) {
const struct cache_entry *ce = list.entries[i];
if (ce_stage(ce))
printf("%06o %s U\t", ce->ce_mode,
oid_to_hex(null_oid()));
else
printf("%06o %s %d\t", ce->ce_mode,
oid_to_hex(&ce->oid), ce_stage(ce));
fprintf(stdout, "%s\n", ce->name);
}
return 0;
}
static void for_each_listed_submodule(const struct module_list *list,
each_submodule_fn fn, void *cb_data)
{
int i;
for (i = 0; i < list->nr; i++)
fn(list->entries[i], cb_data);
}
struct foreach_cb {
int argc;
const char **argv;
const char *prefix;
int quiet;
int recursive;
};
#define FOREACH_CB_INIT { 0 }
static void runcommand_in_submodule_cb(const struct cache_entry *list_item,
void *cb_data)
{
struct foreach_cb *info = cb_data;
const char *path = list_item->name;
const struct object_id *ce_oid = &list_item->oid;
const struct submodule *sub;
struct child_process cp = CHILD_PROCESS_INIT;
char *displaypath;
displaypath = get_submodule_displaypath(path, info->prefix);
sub = submodule_from_path(the_repository, null_oid(), path);
if (!sub)
die(_("No url found for submodule path '%s' in .gitmodules"),
displaypath);
if (!is_submodule_populated_gently(path, NULL))
goto cleanup;
prepare_submodule_repo_env(&cp.env);
/*
* For the purpose of executing <command> in the submodule,
* separate shell is used for the purpose of running the
* child process.
*/
cp.use_shell = 1;
cp.dir = path;
/*
* NEEDSWORK: the command currently has access to the variables $name,
* $sm_path, $displaypath, $sha1 and $toplevel only when the command
* contains a single argument. This is done for maintaining a faithful
* translation from shell script.
*/
if (info->argc == 1) {
char *toplevel = xgetcwd();
struct strbuf sb = STRBUF_INIT;
strvec_pushf(&cp.env, "name=%s", sub->name);
strvec_pushf(&cp.env, "sm_path=%s", path);
strvec_pushf(&cp.env, "displaypath=%s", displaypath);
strvec_pushf(&cp.env, "sha1=%s",
oid_to_hex(ce_oid));
strvec_pushf(&cp.env, "toplevel=%s", toplevel);
/*
* Since the path variable was accessible from the script
* before porting, it is also made available after porting.
* The environment variable "PATH" has a very special purpose
* on windows. And since environment variables are
* case-insensitive in windows, it interferes with the
* existing PATH variable. Hence, to avoid that, we expose
* path via the args strvec and not via env.
*/
sq_quote_buf(&sb, path);
strvec_pushf(&cp.args, "path=%s; %s",
sb.buf, info->argv[0]);
strbuf_release(&sb);
free(toplevel);
} else {
strvec_pushv(&cp.args, info->argv);
}
if (!info->quiet)
printf(_("Entering '%s'\n"), displaypath);
if (info->argv[0] && run_command(&cp))
die(_("run_command returned non-zero status for %s\n."),
displaypath);
if (info->recursive) {
struct child_process cpr = CHILD_PROCESS_INIT;
cpr.git_cmd = 1;
cpr.dir = path;
prepare_submodule_repo_env(&cpr.env);
strvec_pushl(&cpr.args, "--super-prefix", NULL);
strvec_pushf(&cpr.args, "%s/", displaypath);
strvec_pushl(&cpr.args, "submodule--helper", "foreach", "--recursive",
NULL);
if (info->quiet)
strvec_push(&cpr.args, "--quiet");
strvec_push(&cpr.args, "--");
strvec_pushv(&cpr.args, info->argv);
if (run_command(&cpr))
die(_("run_command returned non-zero status while "
"recursing in the nested submodules of %s\n."),
displaypath);
}
cleanup:
free(displaypath);
}
static int module_foreach(int argc, const char **argv, const char *prefix)
{
struct foreach_cb info = FOREACH_CB_INIT;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
struct option module_foreach_options[] = {
OPT__QUIET(&info.quiet, N_("suppress output of entering each submodule command")),
OPT_BOOL(0, "recursive", &info.recursive,
N_("recurse into nested submodules")),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule foreach [--quiet] [--recursive] [--] <command>"),
NULL
};
argc = parse_options(argc, argv, prefix, module_foreach_options,
submodule foreach: fix "<command> --quiet" not being respected Robin reported that git submodule foreach --quiet git pull --quiet origin is not really quiet anymore [1]. "git pull" behaves as if --quiet is not given. This happens because parseopt in submodule--helper will try to parse both --quiet options as if they are foreach's options, not git-pull's. The parsed options are removed from the command line. So when we do pull later, we execute just this git pull origin When calling submodule helper, adding "--" in front of "git pull" will stop parseopt for parsing options that do not really belong to submodule--helper foreach. PARSE_OPT_KEEP_UNKNOWN is removed as a safety measure. parseopt should never see unknown options or something has gone wrong. There are also a couple usage string update while I'm looking at them. While at it, I also add "--" to other subcommands that pass "$@" to submodule--helper. "$@" in these cases are paths and less likely to be --something-like-this. But the point still stands, git-submodule has parsed and classified what are options, what are paths. submodule--helper should never consider paths passed by git-submodule to be options even if they look like one. The test case is also contributed by Robin. [1] it should be quiet before fc1b9243cd (submodule: port submodule subcommand 'foreach' from shell to C, 2018-05-10) because parseopt can't accidentally eat options then. Reported-by: Robin H. Johnson <robbat2@gentoo.org> Tested-by: Robin H. Johnson <robbat2@gentoo.org> Signed-off-by: Robin H. Johnson <robbat2@gentoo.org> Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-04-12 13:08:19 +03:00
git_submodule_helper_usage, 0);
if (module_list_compute(0, NULL, prefix, &pathspec, &list) < 0)
return 1;
info.argc = argc;
info.argv = argv;
info.prefix = prefix;
for_each_listed_submodule(&list, runcommand_in_submodule_cb, &info);
return 0;
}
static int starts_with_dot_slash(const char *const path)
{
return path_match_flags(path, PATH_MATCH_STARTS_WITH_DOT_SLASH |
PATH_MATCH_XPLATFORM);
}
static int starts_with_dot_dot_slash(const char *const path)
{
return path_match_flags(path, PATH_MATCH_STARTS_WITH_DOT_DOT_SLASH |
PATH_MATCH_XPLATFORM);
}
struct init_cb {
const char *prefix;
const char *superprefix;
unsigned int flags;
};
#define INIT_CB_INIT { 0 }
static void init_submodule(const char *path, const char *prefix,
const char *superprefix, unsigned int flags)
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
{
const struct submodule *sub;
struct strbuf sb = STRBUF_INIT;
char *upd = NULL, *url = NULL, *displaypath;
/* try superprefix from the environment, if it is not passed explicitly */
if (!superprefix)
superprefix = get_super_prefix();
displaypath = do_get_submodule_displaypath(path, prefix, superprefix);
sub = submodule_from_path(the_repository, null_oid(), path);
if (!sub)
die(_("No url found for submodule path '%s' in .gitmodules"),
displaypath);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
/*
* NEEDSWORK: In a multi-working-tree world, this needs to be
* set in the per-worktree config.
*
* Set active flag for the submodule being initialized
*/
if (!is_submodule_active(the_repository, path)) {
strbuf_addf(&sb, "submodule.%s.active", sub->name);
git_config_set_gently(sb.buf, "true");
strbuf_reset(&sb);
}
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
/*
* Copy url setting when it is not set yet.
* To look up the url in .git/config, we must not fall back to
* .gitmodules, so look it up directly.
*/
strbuf_addf(&sb, "submodule.%s.url", sub->name);
if (git_config_get_string(sb.buf, &url)) {
if (!sub->url)
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
die(_("No url found for submodule path '%s' in .gitmodules"),
displaypath);
url = xstrdup(sub->url);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
/* Possibly a url relative to parent */
if (starts_with_dot_dot_slash(url) ||
starts_with_dot_slash(url)) {
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
char *oldurl = url;
url = resolve_relative_url(oldurl, NULL, 0);
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
free(oldurl);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
}
if (git_config_set_gently(sb.buf, url))
die(_("Failed to register url for submodule path '%s'"),
displaypath);
if (!(flags & OPT_QUIET))
fprintf(stderr,
_("Submodule '%s' (%s) registered for path '%s'\n"),
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
sub->name, url, displaypath);
}
strbuf_reset(&sb);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
/* Copy "update" setting when it is not set yet */
strbuf_addf(&sb, "submodule.%s.update", sub->name);
if (git_config_get_string(sb.buf, &upd) &&
sub->update_strategy.type != SM_UPDATE_UNSPECIFIED) {
if (sub->update_strategy.type == SM_UPDATE_COMMAND) {
fprintf(stderr, _("warning: command update mode suggested for submodule '%s'\n"),
sub->name);
upd = xstrdup("none");
} else
upd = xstrdup(submodule_strategy_to_string(&sub->update_strategy));
if (git_config_set_gently(sb.buf, upd))
die(_("Failed to register update mode for submodule path '%s'"), displaypath);
}
strbuf_release(&sb);
free(displaypath);
free(url);
free(upd);
}
static void init_submodule_cb(const struct cache_entry *list_item, void *cb_data)
{
struct init_cb *info = cb_data;
init_submodule(list_item->name, info->prefix, info->superprefix, info->flags);
}
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
static int module_init(int argc, const char **argv, const char *prefix)
{
struct init_cb info = INIT_CB_INIT;
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
int quiet = 0;
struct option module_init_options[] = {
OPT__QUIET(&quiet, N_("suppress output for initializing a submodule")),
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule init [<options>] [<path>]"),
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
NULL
};
argc = parse_options(argc, argv, prefix, module_init_options,
git_submodule_helper_usage, 0);
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
/*
* If there are no path args and submodule.active is set then,
* by default, only initialize 'active' modules.
*/
if (!argc && git_config_get_value_multi("submodule.active"))
module_list_active(&list);
info.prefix = prefix;
if (quiet)
info.flags |= OPT_QUIET;
for_each_listed_submodule(&list, init_submodule_cb, &info);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
return 0;
}
struct status_cb {
const char *prefix;
unsigned int flags;
};
#define STATUS_CB_INIT { 0 }
static void print_status(unsigned int flags, char state, const char *path,
const struct object_id *oid, const char *displaypath)
{
if (flags & OPT_QUIET)
return;
printf("%c%s %s", state, oid_to_hex(oid), displaypath);
if (state == ' ' || state == '+') {
const char *name = compute_rev_name(path, oid_to_hex(oid));
if (name)
printf(" (%s)", name);
}
printf("\n");
}
static int handle_submodule_head_ref(const char *refname,
const struct object_id *oid, int flags,
void *cb_data)
{
struct object_id *output = cb_data;
if (oid)
oidcpy(output, oid);
return 0;
}
static void status_submodule(const char *path, const struct object_id *ce_oid,
unsigned int ce_flags, const char *prefix,
unsigned int flags)
{
char *displaypath;
struct strvec diff_files_args = STRVEC_INIT;
struct rev_info rev = REV_INFO_INIT;
int diff_files_result;
struct strbuf buf = STRBUF_INIT;
const char *git_dir;
if (!submodule_from_path(the_repository, null_oid(), path))
die(_("no submodule mapping found in .gitmodules for path '%s'"),
path);
displaypath = get_submodule_displaypath(path, prefix);
if ((CE_STAGEMASK & ce_flags) >> CE_STAGESHIFT) {
print_status(flags, 'U', path, null_oid(), displaypath);
goto cleanup;
}
strbuf_addf(&buf, "%s/.git", path);
git_dir = read_gitfile(buf.buf);
if (!git_dir)
git_dir = buf.buf;
if (!is_submodule_active(the_repository, path) ||
!is_git_directory(git_dir)) {
print_status(flags, '-', path, ce_oid, displaypath);
strbuf_release(&buf);
goto cleanup;
}
strbuf_release(&buf);
strvec_pushl(&diff_files_args, "diff-files",
"--ignore-submodules=dirty", "--quiet", "--",
path, NULL);
git_config(git_diff_basic_config, NULL);
repo_init_revisions(the_repository, &rev, NULL);
rev.abbrev = 0;
diff_files_args.nr = setup_revisions(diff_files_args.nr,
diff_files_args.v,
&rev, NULL);
diff_files_result = run_diff_files(&rev, 0);
if (!diff_result_code(&rev.diffopt, diff_files_result)) {
print_status(flags, ' ', path, ce_oid,
displaypath);
} else if (!(flags & OPT_CACHED)) {
struct object_id oid;
struct ref_store *refs = get_submodule_ref_store(path);
if (!refs) {
print_status(flags, '-', path, ce_oid, displaypath);
goto cleanup;
}
if (refs_head_ref(refs, handle_submodule_head_ref, &oid))
die(_("could not resolve HEAD ref inside the "
"submodule '%s'"), path);
print_status(flags, '+', path, &oid, displaypath);
} else {
print_status(flags, '+', path, ce_oid, displaypath);
}
if (flags & OPT_RECURSIVE) {
struct child_process cpr = CHILD_PROCESS_INIT;
cpr.git_cmd = 1;
cpr.dir = path;
prepare_submodule_repo_env(&cpr.env);
strvec_push(&cpr.args, "--super-prefix");
strvec_pushf(&cpr.args, "%s/", displaypath);
strvec_pushl(&cpr.args, "submodule--helper", "status",
"--recursive", NULL);
if (flags & OPT_CACHED)
strvec_push(&cpr.args, "--cached");
if (flags & OPT_QUIET)
strvec_push(&cpr.args, "--quiet");
if (run_command(&cpr))
die(_("failed to recurse into submodule '%s'"), path);
}
cleanup:
strvec_clear(&diff_files_args);
free(displaypath);
release_revisions(&rev);
}
static void status_submodule_cb(const struct cache_entry *list_item,
void *cb_data)
{
struct status_cb *info = cb_data;
status_submodule(list_item->name, &list_item->oid, list_item->ce_flags,
info->prefix, info->flags);
}
static int module_status(int argc, const char **argv, const char *prefix)
{
struct status_cb info = STATUS_CB_INIT;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
int quiet = 0;
struct option module_status_options[] = {
OPT__QUIET(&quiet, N_("suppress submodule status output")),
OPT_BIT(0, "cached", &info.flags, N_("use commit stored in the index instead of the one stored in the submodule HEAD"), OPT_CACHED),
OPT_BIT(0, "recursive", &info.flags, N_("recurse into nested submodules"), OPT_RECURSIVE),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule status [--quiet] [--cached] [--recursive] [<path>...]"),
NULL
};
argc = parse_options(argc, argv, prefix, module_status_options,
git_submodule_helper_usage, 0);
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
info.prefix = prefix;
if (quiet)
info.flags |= OPT_QUIET;
for_each_listed_submodule(&list, status_submodule_cb, &info);
submodule: port init from shell to C By having the `submodule init` functionality in C, we can reference it easier from other parts in the code in later patches. The code is split up to have one function to initialize one submodule and a calling function that takes care of the rest, such as argument handling and translating the arguments to the paths of the submodules. This is the first submodule subcommand that is fully converted to C except for the usage string, so this is actually removing a call to the `submodule--helper list` function, which is supposed to be used in this transition. Instead we'll make a direct call to `module_list_compute`. An explanation why we need to edit the prefixes in cmd_update in git-submodule.sh in this patch: By having no processing in the shell part, we need to convey the notion of wt_prefix and prefix to the C parts, which former patches punted on and did the processing of displaying path in the shell. `wt_prefix` used to hold the path from the repository root to the current directory, e.g. wt_prefix would be t/ if the user invoked the `git submodule` command in ~/repo/t and ~repo is the GIT_DIR. `prefix` used to hold the relative path from the repository root to the operation, e.g. if you have recursive submodules, the shell script would modify the `prefix` in each recursive step by adding the submodule path. We will pass `wt_prefix` into the C helper via `git -C <dir>` as that will setup git in the directory the user actually called git-submodule.sh from. The `prefix` will be passed in via the `--prefix` option. Having `prefix` and `wt_prefix` relative to the GIT_DIR of the calling superproject is unfortunate with this patch as the C code doesn't know about a possible recursion from a superproject via `submodule update --init --recursive`. To fix this, we change the meaning of `wt_prefix` to point to the current project instead of the superproject and `prefix` to include any relative paths issues in the superproject. That way `prefix` will become the leading part for displaying paths and `wt_prefix` will be empty in recursive calls for now. The new notion of `wt_prefix` and `prefix` still allows us to reconstruct the calling directory in the superproject by just traveling reverse of `prefix`. Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-04-16 03:50:13 +03:00
return 0;
}
static int module_name(int argc, const char **argv, const char *prefix)
{
const struct submodule *sub;
if (argc != 2)
usage(_("git submodule--helper name <path>"));
sub = submodule_from_path(the_repository, null_oid(), argv[1]);
if (!sub)
die(_("no submodule mapping found in .gitmodules for path '%s'"),
argv[1]);
printf("%s\n", sub->name);
return 0;
}
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
struct module_cb {
unsigned int mod_src;
unsigned int mod_dst;
struct object_id oid_src;
struct object_id oid_dst;
char status;
const char *sm_path;
};
#define MODULE_CB_INIT { 0 }
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
struct module_cb_list {
struct module_cb **entries;
int alloc, nr;
};
#define MODULE_CB_LIST_INIT { 0 }
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
struct summary_cb {
int argc;
const char **argv;
const char *prefix;
unsigned int cached: 1;
unsigned int for_status: 1;
unsigned int files: 1;
int summary_limit;
};
#define SUMMARY_CB_INIT { 0 }
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
enum diff_cmd {
DIFF_INDEX,
DIFF_FILES
};
static char *verify_submodule_committish(const char *sm_path,
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
const char *committish)
{
struct child_process cp_rev_parse = CHILD_PROCESS_INIT;
struct strbuf result = STRBUF_INIT;
cp_rev_parse.git_cmd = 1;
cp_rev_parse.dir = sm_path;
prepare_submodule_repo_env(&cp_rev_parse.env);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
strvec_pushl(&cp_rev_parse.args, "rev-parse", "-q", "--short", NULL);
strvec_pushf(&cp_rev_parse.args, "%s^0", committish);
strvec_push(&cp_rev_parse.args, "--");
if (capture_command(&cp_rev_parse, &result, 0))
return NULL;
strbuf_trim_trailing_newline(&result);
return strbuf_detach(&result, NULL);
}
static void print_submodule_summary(struct summary_cb *info, char *errmsg,
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
int total_commits, const char *displaypath,
const char *src_abbrev, const char *dst_abbrev,
struct module_cb *p)
{
if (p->status == 'T') {
if (S_ISGITLINK(p->mod_dst))
printf(_("* %s %s(blob)->%s(submodule)"),
displaypath, src_abbrev, dst_abbrev);
else
printf(_("* %s %s(submodule)->%s(blob)"),
displaypath, src_abbrev, dst_abbrev);
} else {
printf("* %s %s...%s",
displaypath, src_abbrev, dst_abbrev);
}
if (total_commits < 0)
printf(":\n");
else
printf(" (%d):\n", total_commits);
if (errmsg) {
printf(_("%s"), errmsg);
} else if (total_commits > 0) {
struct child_process cp_log = CHILD_PROCESS_INIT;
cp_log.git_cmd = 1;
cp_log.dir = p->sm_path;
prepare_submodule_repo_env(&cp_log.env);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
strvec_pushl(&cp_log.args, "log", NULL);
if (S_ISGITLINK(p->mod_src) && S_ISGITLINK(p->mod_dst)) {
if (info->summary_limit > 0)
strvec_pushf(&cp_log.args, "-%d",
info->summary_limit);
strvec_pushl(&cp_log.args, "--pretty= %m %s",
"--first-parent", NULL);
strvec_pushf(&cp_log.args, "%s...%s",
src_abbrev, dst_abbrev);
} else if (S_ISGITLINK(p->mod_dst)) {
strvec_pushl(&cp_log.args, "--pretty= > %s",
"-1", dst_abbrev, NULL);
} else {
strvec_pushl(&cp_log.args, "--pretty= < %s",
"-1", src_abbrev, NULL);
}
run_command(&cp_log);
}
printf("\n");
}
static void generate_submodule_summary(struct summary_cb *info,
struct module_cb *p)
{
char *displaypath, *src_abbrev = NULL, *dst_abbrev;
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
int missing_src = 0, missing_dst = 0;
char *errmsg = NULL;
int total_commits = -1;
if (!info->cached && oideq(&p->oid_dst, null_oid())) {
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
if (S_ISGITLINK(p->mod_dst)) {
struct ref_store *refs = get_submodule_ref_store(p->sm_path);
if (refs)
refs_head_ref(refs, handle_submodule_head_ref, &p->oid_dst);
} else if (S_ISLNK(p->mod_dst) || S_ISREG(p->mod_dst)) {
struct stat st;
int fd = open(p->sm_path, O_RDONLY);
if (fd < 0 || fstat(fd, &st) < 0 ||
index_fd(&the_index, &p->oid_dst, fd, &st, OBJ_BLOB,
p->sm_path, 0))
error(_("couldn't hash object from '%s'"), p->sm_path);
} else {
/* for a submodule removal (mode:0000000), don't warn */
if (p->mod_dst)
warning(_("unexpected mode %o\n"), p->mod_dst);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
}
}
if (S_ISGITLINK(p->mod_src)) {
if (p->status != 'D')
src_abbrev = verify_submodule_committish(p->sm_path,
oid_to_hex(&p->oid_src));
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
if (!src_abbrev) {
missing_src = 1;
/*
* As `rev-parse` failed, we fallback to getting
* the abbreviated hash using oid_src. We do
* this as we might still need the abbreviated
* hash in cases like a submodule type change, etc.
*/
src_abbrev = xstrndup(oid_to_hex(&p->oid_src), 7);
}
} else {
/*
* The source does not point to a submodule.
* So, we fallback to getting the abbreviation using
* oid_src as we might still need the abbreviated
* hash in cases like submodule add, etc.
*/
src_abbrev = xstrndup(oid_to_hex(&p->oid_src), 7);
}
if (S_ISGITLINK(p->mod_dst)) {
dst_abbrev = verify_submodule_committish(p->sm_path,
oid_to_hex(&p->oid_dst));
if (!dst_abbrev) {
missing_dst = 1;
/*
* As `rev-parse` failed, we fallback to getting
* the abbreviated hash using oid_dst. We do
* this as we might still need the abbreviated
* hash in cases like a submodule type change, etc.
*/
dst_abbrev = xstrndup(oid_to_hex(&p->oid_dst), 7);
}
} else {
/*
* The destination does not point to a submodule.
* So, we fallback to getting the abbreviation using
* oid_dst as we might still need the abbreviated
* hash in cases like a submodule removal, etc.
*/
dst_abbrev = xstrndup(oid_to_hex(&p->oid_dst), 7);
}
displaypath = get_submodule_displaypath(p->sm_path, info->prefix);
if (!missing_src && !missing_dst) {
struct child_process cp_rev_list = CHILD_PROCESS_INIT;
struct strbuf sb_rev_list = STRBUF_INIT;
strvec_pushl(&cp_rev_list.args, "rev-list",
"--first-parent", "--count", NULL);
if (S_ISGITLINK(p->mod_src) && S_ISGITLINK(p->mod_dst))
strvec_pushf(&cp_rev_list.args, "%s...%s",
src_abbrev, dst_abbrev);
else
strvec_push(&cp_rev_list.args, S_ISGITLINK(p->mod_src) ?
src_abbrev : dst_abbrev);
strvec_push(&cp_rev_list.args, "--");
cp_rev_list.git_cmd = 1;
cp_rev_list.dir = p->sm_path;
prepare_submodule_repo_env(&cp_rev_list.env);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
if (!capture_command(&cp_rev_list, &sb_rev_list, 0))
total_commits = atoi(sb_rev_list.buf);
strbuf_release(&sb_rev_list);
} else {
/*
* Don't give error msg for modification whose dst is not
* submodule, i.e., deleted or changed to blob
*/
if (S_ISGITLINK(p->mod_dst)) {
struct strbuf errmsg_str = STRBUF_INIT;
if (missing_src && missing_dst) {
strbuf_addf(&errmsg_str, " Warn: %s doesn't contain commits %s and %s\n",
displaypath, oid_to_hex(&p->oid_src),
oid_to_hex(&p->oid_dst));
} else {
strbuf_addf(&errmsg_str, " Warn: %s doesn't contain commit %s\n",
displaypath, missing_src ?
oid_to_hex(&p->oid_src) :
oid_to_hex(&p->oid_dst));
}
errmsg = strbuf_detach(&errmsg_str, NULL);
}
}
print_submodule_summary(info, errmsg, total_commits,
displaypath, src_abbrev,
dst_abbrev, p);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
free(displaypath);
free(src_abbrev);
free(dst_abbrev);
}
static void prepare_submodule_summary(struct summary_cb *info,
struct module_cb_list *list)
{
int i;
for (i = 0; i < list->nr; i++) {
const struct submodule *sub;
struct module_cb *p = list->entries[i];
struct strbuf sm_gitdir = STRBUF_INIT;
if (p->status == 'D' || p->status == 'T') {
generate_submodule_summary(info, p);
continue;
}
if (info->for_status && p->status != 'A' &&
(sub = submodule_from_path(the_repository,
null_oid(), p->sm_path))) {
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
char *config_key = NULL;
const char *value;
int ignore_all = 0;
config_key = xstrfmt("submodule.%s.ignore",
sub->name);
if (!git_config_get_string_tmp(config_key, &value))
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
ignore_all = !strcmp(value, "all");
else if (sub->ignore)
ignore_all = !strcmp(sub->ignore, "all");
free(config_key);
if (ignore_all)
continue;
}
/* Also show added or modified modules which are checked out */
strbuf_addstr(&sm_gitdir, p->sm_path);
if (is_nonbare_repository_dir(&sm_gitdir))
generate_submodule_summary(info, p);
strbuf_release(&sm_gitdir);
}
}
static void submodule_summary_callback(struct diff_queue_struct *q,
struct diff_options *options,
void *data)
{
int i;
struct module_cb_list *list = data;
for (i = 0; i < q->nr; i++) {
struct diff_filepair *p = q->queue[i];
struct module_cb *temp;
if (!S_ISGITLINK(p->one->mode) && !S_ISGITLINK(p->two->mode))
continue;
temp = (struct module_cb*)malloc(sizeof(struct module_cb));
temp->mod_src = p->one->mode;
temp->mod_dst = p->two->mode;
temp->oid_src = p->one->oid;
temp->oid_dst = p->two->oid;
temp->status = p->status;
temp->sm_path = xstrdup(p->one->path);
ALLOC_GROW(list->entries, list->nr + 1, list->alloc);
list->entries[list->nr++] = temp;
}
}
static const char *get_diff_cmd(enum diff_cmd diff_cmd)
{
switch (diff_cmd) {
case DIFF_INDEX: return "diff-index";
case DIFF_FILES: return "diff-files";
default: BUG("bad diff_cmd value %d", diff_cmd);
}
}
static int compute_summary_module_list(struct object_id *head_oid,
struct summary_cb *info,
enum diff_cmd diff_cmd)
{
struct strvec diff_args = STRVEC_INIT;
struct rev_info rev;
struct module_cb_list list = MODULE_CB_LIST_INIT;
int ret = 0;
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
strvec_push(&diff_args, get_diff_cmd(diff_cmd));
if (info->cached)
strvec_push(&diff_args, "--cached");
strvec_pushl(&diff_args, "--ignore-submodules=dirty", "--raw", NULL);
if (head_oid)
strvec_push(&diff_args, oid_to_hex(head_oid));
strvec_push(&diff_args, "--");
if (info->argc)
strvec_pushv(&diff_args, info->argv);
git_config(git_diff_basic_config, NULL);
init_revisions(&rev, info->prefix);
rev.abbrev = 0;
MacOS: precompose_argv_prefix() The following sequence leads to a "BUG" assertion running under MacOS: DIR=git-test-restore-p Adiarnfd=$(printf 'A\314\210') DIRNAME=xx${Adiarnfd}yy mkdir $DIR && cd $DIR && git init && mkdir $DIRNAME && cd $DIRNAME && echo "Initial" >file && git add file && echo "One more line" >>file && echo y | git restore -p . Initialized empty Git repository in /tmp/git-test-restore-p/.git/ BUG: pathspec.c:495: error initializing pathspec_item Cannot close git diff-index --cached --numstat [snip] The command `git restore` is run from a directory inside a Git repo. Git needs to split the $CWD into 2 parts: The path to the repo and "the rest", if any. "The rest" becomes a "prefix" later used inside the pathspec code. As an example, "/path/to/repo/dir-inside-repå" would determine "/path/to/repo" as the root of the repo, the place where the configuration file .git/config is found. The rest becomes the prefix ("dir-inside-repå"), from where the pathspec machinery expands the ".", more about this later. If there is a decomposed form, (making the decomposing visible like this), "dir-inside-rep°a" doesn't match "dir-inside-repå". Git commands need to: (a) read the configuration variable "core.precomposeunicode" (b) precocompose argv[] (c) precompose the prefix, if there was any The first commit, 76759c7dff53 "git on Mac OS and precomposed unicode" addressed (a) and (b). The call to precompose_argv() was added into parse-options.c, because that seemed to be a good place when the patch was written. Commands that don't use parse-options need to do (a) and (b) themselfs. The commands `diff-files`, `diff-index`, `diff-tree` and `diff` learned (a) and (b) in commit 90a78b83e0b8 "diff: run arguments through precompose_argv" Branch names (or refs in general) using decomposed code points resulting in decomposed file names had been fixed in commit 8e712ef6fc97 "Honor core.precomposeUnicode in more places" The bug report from above shows 2 things: - more commands need to handle precomposed unicode - (c) should be implemented for all commands using pathspecs Solution: precompose_argv() now handles the prefix (if needed), and is renamed into precompose_argv_prefix(). Inside this function the config variable core.precomposeunicode is read into the global variable precomposed_unicode, as before. This reading is skipped if precomposed_unicode had been read before. The original patch for preocomposed unicode, 76759c7dff53, placed precompose_argv() into parse-options.c Now add it into git.c::run_builtin() as well. Existing precompose calls in diff-files.c and others may become redundant, and if we audit the callflows that reach these places to make sure that they can never be reached without going through the new call added to run_builtin(), we might be able to remove these existing ones. But in this commit, we do not bother to do so and leave these precompose callsites as they are. Because precompose() is idempotent and can be called on an already precomposed string safely, this is safer than removing existing calls without fully vetting the callflows. There is certainly room for cleanups - this change intends to be a bug fix. Cleanups needs more tests in e.g. t/t3910-mac-os-precompose.sh, and should be done in future commits. [1] git-bugreport-2021-01-06-1209.txt (git can't deal with special characters) [2] https://lore.kernel.org/git/A102844A-9501-4A86-854D-E3B387D378AA@icloud.com/ Reported-by: Daniel Troger <random_n0body@icloud.com> Helped-By: Philippe Blain <levraiphilippeblain@gmail.com> Signed-off-by: Torsten Bögershausen <tboegi@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2021-02-03 19:28:23 +03:00
precompose_argv_prefix(diff_args.nr, diff_args.v, NULL);
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
setup_revisions(diff_args.nr, diff_args.v, &rev, NULL);
rev.diffopt.output_format = DIFF_FORMAT_NO_OUTPUT | DIFF_FORMAT_CALLBACK;
rev.diffopt.format_callback = submodule_summary_callback;
rev.diffopt.format_callback_data = &list;
if (!info->cached) {
if (diff_cmd == DIFF_INDEX)
setup_work_tree();
if (read_cache_preload(&rev.diffopt.pathspec) < 0) {
perror("read_cache_preload");
ret = -1;
goto cleanup;
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
}
} else if (read_cache() < 0) {
perror("read_cache");
ret = -1;
goto cleanup;
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
}
if (diff_cmd == DIFF_INDEX)
run_diff_index(&rev, info->cached);
else
run_diff_files(&rev, 0);
prepare_submodule_summary(info, &list);
cleanup:
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
strvec_clear(&diff_args);
release_revisions(&rev);
return ret;
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
}
static int module_summary(int argc, const char **argv, const char *prefix)
{
struct summary_cb info = SUMMARY_CB_INIT;
int cached = 0;
int for_status = 0;
int files = 0;
int summary_limit = -1;
enum diff_cmd diff_cmd = DIFF_INDEX;
struct object_id head_oid;
int ret;
struct option module_summary_options[] = {
OPT_BOOL(0, "cached", &cached,
N_("use the commit stored in the index instead of the submodule HEAD")),
OPT_BOOL(0, "files", &files,
N_("compare the commit in the index with that in the submodule HEAD")),
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
OPT_BOOL(0, "for-status", &for_status,
N_("skip submodules with 'ignore_config' value set to 'all'")),
OPT_INTEGER('n', "summary-limit", &summary_limit,
N_("limit the summary size")),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule summary [<options>] [<commit>] [--] [<path>]"),
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
NULL
};
argc = parse_options(argc, argv, prefix, module_summary_options,
git_submodule_helper_usage, 0);
if (!summary_limit)
return 0;
if (!get_oid(argc ? argv[0] : "HEAD", &head_oid)) {
if (argc) {
argv++;
argc--;
}
} else if (!argc || !strcmp(argv[0], "HEAD")) {
/* before the first commit: compare with an empty tree */
oidcpy(&head_oid, the_hash_algo->empty_tree);
if (argc) {
argv++;
argc--;
}
} else {
if (get_oid("HEAD", &head_oid))
die(_("could not fetch a revision for HEAD"));
}
if (files) {
if (cached)
die(_("options '%s' and '%s' cannot be used together"), "--cached", "--files");
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
diff_cmd = DIFF_FILES;
}
info.argc = argc;
info.argv = argv;
info.prefix = prefix;
info.cached = !!cached;
info.files = !!files;
info.for_status = !!for_status;
info.summary_limit = summary_limit;
ret = compute_summary_module_list((diff_cmd == DIFF_INDEX) ? &head_oid : NULL,
&info, diff_cmd);
return ret;
}
struct sync_cb {
const char *prefix;
unsigned int flags;
};
#define SYNC_CB_INIT { 0 }
static void sync_submodule(const char *path, const char *prefix,
unsigned int flags)
{
const struct submodule *sub;
char *remote_key = NULL;
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
char *sub_origin_url, *super_config_url, *displaypath, *default_remote;
struct strbuf sb = STRBUF_INIT;
char *sub_config_path = NULL;
if (!is_submodule_active(the_repository, path))
return;
sub = submodule_from_path(the_repository, null_oid(), path);
if (sub && sub->url) {
if (starts_with_dot_dot_slash(sub->url) ||
starts_with_dot_slash(sub->url)) {
char *up_path = get_up_path(path);
sub_origin_url = resolve_relative_url(sub->url, up_path, 1);
super_config_url = resolve_relative_url(sub->url, NULL, 1);
free(up_path);
} else {
sub_origin_url = xstrdup(sub->url);
super_config_url = xstrdup(sub->url);
}
} else {
sub_origin_url = xstrdup("");
super_config_url = xstrdup("");
}
displaypath = get_submodule_displaypath(path, prefix);
if (!(flags & OPT_QUIET))
printf(_("Synchronizing submodule url for '%s'\n"),
displaypath);
strbuf_reset(&sb);
strbuf_addf(&sb, "submodule.%s.url", sub->name);
if (git_config_set_gently(sb.buf, super_config_url))
die(_("failed to register url for submodule path '%s'"),
displaypath);
if (!is_submodule_populated_gently(path, NULL))
goto cleanup;
strbuf_reset(&sb);
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
default_remote = get_default_remote_submodule(path);
if (!default_remote)
die(_("failed to get the default remote for submodule '%s'"),
path);
submodule--helper: get remote names from any repository `get_default_remote()` retrieves the name of a remote by resolving the refs from of the current repository's ref store. Thus in order to use it for retrieving the remote name of a submodule, we have to start a new subprocess which runs from the submodule directory. Let's instead introduce a function called `repo_get_default_remote()` which takes any repository object and retrieves the remote accordingly. `get_default_remote()` is then defined as a call to `repo_get_default_remote()` with 'the_repository' passed to it. Now that we have `repo_get_default_remote()`, we no longer have to start a subprocess that called `submodule--helper get-default-remote` from within the submodule directory. So let's make a function called `get_default_remote_submodule()` which takes a submodule path, and returns the default remote for that submodule, all within the same process. We can now use this function to save an unnecessary subprocess spawn in `sync_submodule()`, and also in a subsequent patch, which will require this functionality. Mentored-by: Christian Couder <christian.couder@gmail.com> Mentored-by: Shourya Shukla <periperidip@gmail.com> Helped-by: Glen Choo <chooglen@google.com> Signed-off-by: Atharva Raykar <raykar.ath@gmail.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Glen Choo <chooglen@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-05 03:13:53 +03:00
remote_key = xstrfmt("remote.%s.url", default_remote);
free(default_remote);
submodule_to_gitdir(&sb, path);
strbuf_addstr(&sb, "/config");
if (git_config_set_in_file_gently(sb.buf, remote_key, sub_origin_url))
die(_("failed to update remote for submodule '%s'"),
path);
if (flags & OPT_RECURSIVE) {
struct child_process cpr = CHILD_PROCESS_INIT;
cpr.git_cmd = 1;
cpr.dir = path;
prepare_submodule_repo_env(&cpr.env);
strvec_push(&cpr.args, "--super-prefix");
strvec_pushf(&cpr.args, "%s/", displaypath);
strvec_pushl(&cpr.args, "submodule--helper", "sync",
"--recursive", NULL);
if (flags & OPT_QUIET)
strvec_push(&cpr.args, "--quiet");
if (run_command(&cpr))
die(_("failed to recurse into submodule '%s'"),
path);
}
cleanup:
free(super_config_url);
free(sub_origin_url);
strbuf_release(&sb);
free(remote_key);
free(displaypath);
free(sub_config_path);
}
static void sync_submodule_cb(const struct cache_entry *list_item, void *cb_data)
{
struct sync_cb *info = cb_data;
sync_submodule(list_item->name, info->prefix, info->flags);
}
static int module_sync(int argc, const char **argv, const char *prefix)
{
struct sync_cb info = SYNC_CB_INIT;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
int quiet = 0;
int recursive = 0;
struct option module_sync_options[] = {
OPT__QUIET(&quiet, N_("suppress output of synchronizing submodule url")),
OPT_BOOL(0, "recursive", &recursive,
N_("recurse into nested submodules")),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule sync [--quiet] [--recursive] [<path>]"),
NULL
};
argc = parse_options(argc, argv, prefix, module_sync_options,
git_submodule_helper_usage, 0);
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
info.prefix = prefix;
if (quiet)
info.flags |= OPT_QUIET;
if (recursive)
info.flags |= OPT_RECURSIVE;
for_each_listed_submodule(&list, sync_submodule_cb, &info);
return 0;
}
struct deinit_cb {
const char *prefix;
unsigned int flags;
};
#define DEINIT_CB_INIT { 0 }
static void deinit_submodule(const char *path, const char *prefix,
unsigned int flags)
{
const struct submodule *sub;
char *displaypath = NULL;
struct child_process cp_config = CHILD_PROCESS_INIT;
struct strbuf sb_config = STRBUF_INIT;
char *sub_git_dir = xstrfmt("%s/.git", path);
sub = submodule_from_path(the_repository, null_oid(), path);
if (!sub || !sub->name)
goto cleanup;
displaypath = get_submodule_displaypath(path, prefix);
/* remove the submodule work tree (unless the user already did it) */
if (is_directory(path)) {
struct strbuf sb_rm = STRBUF_INIT;
const char *format;
if (is_directory(sub_git_dir)) {
if (!(flags & OPT_QUIET))
warning(_("Submodule work tree '%s' contains a .git "
"directory. This will be replaced with a "
".git file by using absorbgitdirs."),
displaypath);
absorb_git_dir_into_superproject(path,
ABSORB_GITDIR_RECURSE_SUBMODULES);
}
if (!(flags & OPT_FORCE)) {
struct child_process cp_rm = CHILD_PROCESS_INIT;
cp_rm.git_cmd = 1;
strvec_pushl(&cp_rm.args, "rm", "-qn",
path, NULL);
if (run_command(&cp_rm))
die(_("Submodule work tree '%s' contains local "
"modifications; use '-f' to discard them"),
displaypath);
}
strbuf_addstr(&sb_rm, path);
if (!remove_dir_recursively(&sb_rm, 0))
format = _("Cleared directory '%s'\n");
else
format = _("Could not remove submodule work tree '%s'\n");
if (!(flags & OPT_QUIET))
printf(format, displaypath);
submodule_unset_core_worktree(sub);
strbuf_release(&sb_rm);
}
if (mkdir(path, 0777))
printf(_("could not create empty submodule directory %s"),
displaypath);
cp_config.git_cmd = 1;
strvec_pushl(&cp_config.args, "config", "--get-regexp", NULL);
strvec_pushf(&cp_config.args, "submodule.%s\\.", sub->name);
/* remove the .git/config entries (unless the user already did it) */
if (!capture_command(&cp_config, &sb_config, 0) && sb_config.len) {
char *sub_key = xstrfmt("submodule.%s", sub->name);
/*
* remove the whole section so we have a clean state when
* the user later decides to init this submodule again
*/
git_config_rename_section_in_file(NULL, sub_key, NULL);
if (!(flags & OPT_QUIET))
printf(_("Submodule '%s' (%s) unregistered for path '%s'\n"),
sub->name, sub->url, displaypath);
free(sub_key);
}
cleanup:
free(displaypath);
free(sub_git_dir);
strbuf_release(&sb_config);
}
static void deinit_submodule_cb(const struct cache_entry *list_item,
void *cb_data)
{
struct deinit_cb *info = cb_data;
deinit_submodule(list_item->name, info->prefix, info->flags);
}
static int module_deinit(int argc, const char **argv, const char *prefix)
{
struct deinit_cb info = DEINIT_CB_INIT;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
int quiet = 0;
int force = 0;
int all = 0;
struct option module_deinit_options[] = {
OPT__QUIET(&quiet, N_("suppress submodule status output")),
OPT__FORCE(&force, N_("remove submodule working trees even if they contain local changes"), 0),
OPT_BOOL(0, "all", &all, N_("unregister all submodules")),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule deinit [--quiet] [-f | --force] [--all | [--] [<path>...]]"),
NULL
};
argc = parse_options(argc, argv, prefix, module_deinit_options,
git_submodule_helper_usage, 0);
if (all && argc) {
error("pathspec and --all are incompatible");
usage_with_options(git_submodule_helper_usage,
module_deinit_options);
}
if (!argc && !all)
die(_("Use '--all' if you really want to deinitialize all submodules"));
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
info.prefix = prefix;
if (quiet)
info.flags |= OPT_QUIET;
if (force)
info.flags |= OPT_FORCE;
for_each_listed_submodule(&list, deinit_submodule_cb, &info);
return 0;
}
struct module_clone_data {
const char *prefix;
const char *path;
const char *name;
const char *url;
const char *depth;
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
struct list_objects_filter_options *filter_options;
struct string_list reference;
unsigned int quiet: 1;
unsigned int progress: 1;
unsigned int dissociate: 1;
unsigned int require_init: 1;
int single_branch;
};
#define MODULE_CLONE_DATA_INIT { \
.reference = STRING_LIST_INIT_NODUP, \
.single_branch = -1, \
}
struct submodule_alternate_setup {
const char *submodule_name;
enum SUBMODULE_ALTERNATE_ERROR_MODE {
SUBMODULE_ALTERNATE_ERROR_DIE,
SUBMODULE_ALTERNATE_ERROR_INFO,
SUBMODULE_ALTERNATE_ERROR_IGNORE
} error_mode;
struct string_list *reference;
};
#define SUBMODULE_ALTERNATE_SETUP_INIT { \
.error_mode = SUBMODULE_ALTERNATE_ERROR_IGNORE, \
}
static const char alternate_error_advice[] = N_(
"An alternate computed from a superproject's alternate is invalid.\n"
"To allow Git to clone without an alternate in such a case, set\n"
"submodule.alternateErrorStrategy to 'info' or, equivalently, clone with\n"
"'--reference-if-able' instead of '--reference'."
);
static int add_possible_reference_from_superproject(
struct object_directory *odb, void *sas_cb)
{
struct submodule_alternate_setup *sas = sas_cb;
size_t len;
/*
* If the alternate object store is another repository, try the
* standard layout with .git/(modules/<name>)+/objects
*/
if (strip_suffix(odb->path, "/objects", &len)) {
struct repository alternate;
char *sm_alternate;
struct strbuf sb = STRBUF_INIT;
struct strbuf err = STRBUF_INIT;
strbuf_add(&sb, odb->path, len);
repo_init(&alternate, sb.buf, NULL);
/*
* We need to end the new path with '/' to mark it as a dir,
* otherwise a submodule name containing '/' will be broken
* as the last part of a missing submodule reference would
* be taken as a file name.
*/
strbuf_reset(&sb);
submodule_name_to_gitdir(&sb, &alternate, sas->submodule_name);
strbuf_addch(&sb, '/');
repo_clear(&alternate);
sm_alternate = compute_alternate_path(sb.buf, &err);
if (sm_alternate) {
string_list_append(sas->reference, xstrdup(sb.buf));
free(sm_alternate);
} else {
switch (sas->error_mode) {
case SUBMODULE_ALTERNATE_ERROR_DIE:
if (advice_enabled(ADVICE_SUBMODULE_ALTERNATE_ERROR_STRATEGY_DIE))
advise(_(alternate_error_advice));
die(_("submodule '%s' cannot add alternate: %s"),
sas->submodule_name, err.buf);
case SUBMODULE_ALTERNATE_ERROR_INFO:
fprintf_ln(stderr, _("submodule '%s' cannot add alternate: %s"),
sas->submodule_name, err.buf);
case SUBMODULE_ALTERNATE_ERROR_IGNORE:
; /* nothing */
}
}
strbuf_release(&sb);
}
return 0;
}
static void prepare_possible_alternates(const char *sm_name,
struct string_list *reference)
{
char *sm_alternate = NULL, *error_strategy = NULL;
struct submodule_alternate_setup sas = SUBMODULE_ALTERNATE_SETUP_INIT;
git_config_get_string("submodule.alternateLocation", &sm_alternate);
if (!sm_alternate)
return;
git_config_get_string("submodule.alternateErrorStrategy", &error_strategy);
if (!error_strategy)
error_strategy = xstrdup("die");
sas.submodule_name = sm_name;
sas.reference = reference;
if (!strcmp(error_strategy, "die"))
sas.error_mode = SUBMODULE_ALTERNATE_ERROR_DIE;
else if (!strcmp(error_strategy, "info"))
sas.error_mode = SUBMODULE_ALTERNATE_ERROR_INFO;
else if (!strcmp(error_strategy, "ignore"))
sas.error_mode = SUBMODULE_ALTERNATE_ERROR_IGNORE;
else
die(_("Value '%s' for submodule.alternateErrorStrategy is not recognized"), error_strategy);
if (!strcmp(sm_alternate, "superproject"))
foreach_alt_odb(add_possible_reference_from_superproject, &sas);
else if (!strcmp(sm_alternate, "no"))
; /* do nothing */
else
die(_("Value '%s' for submodule.alternateLocation is not recognized"), sm_alternate);
free(sm_alternate);
free(error_strategy);
}
static int clone_submodule(struct module_clone_data *clone_data)
{
char *p, *sm_gitdir;
char *sm_alternate = NULL, *error_strategy = NULL;
struct strbuf sb = STRBUF_INIT;
struct child_process cp = CHILD_PROCESS_INIT;
submodule_name_to_gitdir(&sb, the_repository, clone_data->name);
sm_gitdir = absolute_pathdup(sb.buf);
strbuf_reset(&sb);
if (!is_absolute_path(clone_data->path)) {
strbuf_addf(&sb, "%s/%s", get_git_work_tree(), clone_data->path);
clone_data->path = strbuf_detach(&sb, NULL);
} else {
clone_data->path = xstrdup(clone_data->path);
}
if (validate_submodule_git_dir(sm_gitdir, clone_data->name) < 0)
die(_("refusing to create/use '%s' in another submodule's "
"git dir"), sm_gitdir);
if (!file_exists(sm_gitdir)) {
if (safe_create_leading_directories_const(sm_gitdir) < 0)
die(_("could not create directory '%s'"), sm_gitdir);
prepare_possible_alternates(clone_data->name, &clone_data->reference);
strvec_push(&cp.args, "clone");
strvec_push(&cp.args, "--no-checkout");
if (clone_data->quiet)
strvec_push(&cp.args, "--quiet");
if (clone_data->progress)
strvec_push(&cp.args, "--progress");
if (clone_data->depth && *(clone_data->depth))
strvec_pushl(&cp.args, "--depth", clone_data->depth, NULL);
if (clone_data->reference.nr) {
struct string_list_item *item;
for_each_string_list_item(item, &clone_data->reference)
strvec_pushl(&cp.args, "--reference",
item->string, NULL);
}
if (clone_data->dissociate)
strvec_push(&cp.args, "--dissociate");
if (sm_gitdir && *sm_gitdir)
strvec_pushl(&cp.args, "--separate-git-dir", sm_gitdir, NULL);
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
if (clone_data->filter_options && clone_data->filter_options->choice)
strvec_pushf(&cp.args, "--filter=%s",
expand_list_objects_filter_spec(
clone_data->filter_options));
if (clone_data->single_branch >= 0)
strvec_push(&cp.args, clone_data->single_branch ?
"--single-branch" :
"--no-single-branch");
strvec_push(&cp.args, "--");
strvec_push(&cp.args, clone_data->url);
strvec_push(&cp.args, clone_data->path);
cp.git_cmd = 1;
prepare_submodule_repo_env(&cp.env);
cp.no_stdin = 1;
if(run_command(&cp))
die(_("clone of '%s' into submodule path '%s' failed"),
clone_data->url, clone_data->path);
} else {
if (clone_data->require_init && !access(clone_data->path, X_OK) &&
!is_empty_dir(clone_data->path))
die(_("directory not empty: '%s'"), clone_data->path);
if (safe_create_leading_directories_const(clone_data->path) < 0)
die(_("could not create directory '%s'"), clone_data->path);
strbuf_addf(&sb, "%s/index", sm_gitdir);
unlink_or_warn(sb.buf);
strbuf_reset(&sb);
}
connect_work_tree_and_git_dir(clone_data->path, sm_gitdir, 0);
p = git_pathdup_submodule(clone_data->path, "config");
if (!p)
die(_("could not get submodule directory for '%s'"), clone_data->path);
/* setup alternateLocation and alternateErrorStrategy in the cloned submodule if needed */
git_config_get_string("submodule.alternateLocation", &sm_alternate);
if (sm_alternate)
git_config_set_in_file(p, "submodule.alternateLocation",
sm_alternate);
git_config_get_string("submodule.alternateErrorStrategy", &error_strategy);
if (error_strategy)
git_config_set_in_file(p, "submodule.alternateErrorStrategy",
error_strategy);
free(sm_alternate);
free(error_strategy);
strbuf_release(&sb);
free(sm_gitdir);
free(p);
return 0;
}
static int module_clone(int argc, const char **argv, const char *prefix)
{
int dissociate = 0, quiet = 0, progress = 0, require_init = 0;
struct module_clone_data clone_data = MODULE_CLONE_DATA_INIT;
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
struct list_objects_filter_options filter_options;
struct option module_clone_options[] = {
OPT_STRING(0, "prefix", &clone_data.prefix,
N_("path"),
N_("alternative anchor for relative paths")),
OPT_STRING(0, "path", &clone_data.path,
N_("path"),
N_("where the new submodule will be cloned to")),
OPT_STRING(0, "name", &clone_data.name,
N_("string"),
N_("name of the new submodule")),
OPT_STRING(0, "url", &clone_data.url,
N_("string"),
N_("url where to clone the submodule from")),
OPT_STRING_LIST(0, "reference", &clone_data.reference,
N_("repo"),
N_("reference repository")),
OPT_BOOL(0, "dissociate", &dissociate,
N_("use --reference only while cloning")),
OPT_STRING(0, "depth", &clone_data.depth,
N_("string"),
N_("depth for shallow clones")),
OPT__QUIET(&quiet, "suppress output for cloning a submodule"),
clone: pass --progress decision to recursive submodules When cloning with "--recursive", we'd generally expect submodules to show progress reports if the main clone did, too. In older versions of git, this mostly worked out of the box. Since we show progress by default when stderr is a tty, and since the child clones inherit the parent stderr, then both processes would come to the same decision by default. If the parent clone was asked for "--quiet", we passed down "--quiet" to the child. However, if stderr was not a tty and the user specified "--progress", we did not propagate this to the child. That's a minor bug, but things got much worse when we switched recently to submodule--helper's update_clone command. With that change, the stderr of the child clones are always connected to a pipe, and we never output progress at all. This patch teaches git-submodule and git-submodule--helper how to pass down an explicit "--progress" flag when cloning. The clone command then decides to propagate that flag based on the cloning decision made earlier (which takes into account isatty(2) of the parent process, existing --progress or --quiet flags, etc). Since the child processes always run without a tty on stderr, we don't have to worry about passing an explicit "--no-progress"; it's the default for them. This fixes the recent loss of progress during recursive clones. And as a bonus, it makes: git clone --recursive --progress ... 2>&1 | cat work by triggering progress explicitly in the children. Signed-off-by: Jeff King <peff@peff.net> Acked-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-09-22 08:24:46 +03:00
OPT_BOOL(0, "progress", &progress,
N_("force cloning progress")),
clone --recurse-submodules: prevent name squatting on Windows In addition to preventing `.git` from being tracked by Git, on Windows we also have to prevent `git~1` from being tracked, as the default NTFS short name (also known as the "8.3 filename") for the file name `.git` is `git~1`, otherwise it would be possible for malicious repositories to write directly into the `.git/` directory, e.g. a `post-checkout` hook that would then be executed _during_ a recursive clone. When we implemented appropriate protections in 2b4c6efc821 (read-cache: optionally disallow NTFS .git variants, 2014-12-16), we had analyzed carefully that the `.git` directory or file would be guaranteed to be the first directory entry to be written. Otherwise it would be possible e.g. for a file named `..git` to be assigned the short name `git~1` and subsequently, the short name generated for `.git` would be `git~2`. Or `git~3`. Or even `~9999999` (for a detailed explanation of the lengths we have to go to protect `.gitmodules`, see the commit message of e7cb0b4455c (is_ntfs_dotgit: match other .git files, 2018-05-11)). However, by exploiting two issues (that will be addressed in a related patch series close by), it is currently possible to clone a submodule into a non-empty directory: - On Windows, file names cannot end in a space or a period (for historical reasons: the period separating the base name from the file extension was not actually written to disk, and the base name/file extension was space-padded to the full 8/3 characters, respectively). Helpfully, when creating a directory under the name, say, `sub.`, that trailing period is trimmed automatically and the actual name on disk is `sub`. This means that while Git thinks that the submodule names `sub` and `sub.` are different, they both access `.git/modules/sub/`. - While the backslash character is a valid file name character on Linux, it is not so on Windows. As Git tries to be cross-platform, it therefore allows backslash characters in the file names stored in tree objects. Which means that it is totally possible that a submodule `c` sits next to a file `c\..git`, and on Windows, during recursive clone a file called `..git` will be written into `c/`, of course _before_ the submodule is cloned. Note that the actual exploit is not quite as simple as having a submodule `c` next to a file `c\..git`, as we have to make sure that the directory `.git/modules/b` already exists when the submodule is checked out, otherwise a different code path is taken in `module_clone()` that does _not_ allow a non-empty submodule directory to exist already. Even if we will address both issues nearby (the next commit will disallow backslash characters in tree entries' file names on Windows, and another patch will disallow creating directories/files with trailing spaces or periods), it is a wise idea to defend in depth against this sort of attack vector: when submodules are cloned recursively, we now _require_ the directory to be empty, addressing CVE-2019-1349. Note: the code path we patch is shared with the code path of `git submodule update --init`, which must not expect, in general, that the directory is empty. Hence we have to introduce the new option `--force-init` and hand it all the way down from `git submodule` to the actual `git submodule--helper` process that performs the initial clone. Reported-by: Nicolas Joly <Nicolas.Joly@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2019-09-12 15:20:39 +03:00
OPT_BOOL(0, "require-init", &require_init,
N_("disallow cloning into non-empty directory")),
OPT_BOOL(0, "single-branch", &clone_data.single_branch,
N_("clone only one branch, HEAD or --branch")),
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
OPT_PARSE_LIST_OBJECTS_FILTER(&filter_options),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule--helper clone [--prefix=<path>] [--quiet] "
"[--reference <repository>] [--name <name>] [--depth <depth>] "
"[--single-branch] [--filter <filter-spec>] "
"--url <url> --path <path>"),
NULL
};
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
memset(&filter_options, 0, sizeof(filter_options));
argc = parse_options(argc, argv, prefix, module_clone_options,
git_submodule_helper_usage, 0);
clone_data.dissociate = !!dissociate;
clone_data.quiet = !!quiet;
clone_data.progress = !!progress;
clone_data.require_init = !!require_init;
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
clone_data.filter_options = &filter_options;
if (argc || !clone_data.url || !clone_data.path || !*(clone_data.path))
usage_with_options(git_submodule_helper_usage,
module_clone_options);
clone_submodule(&clone_data);
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
list_objects_filter_release(&filter_options);
return 0;
}
static void determine_submodule_update_strategy(struct repository *r,
int just_cloned,
const char *path,
const char *update,
struct submodule_update_strategy *out)
{
const struct submodule *sub = submodule_from_path(r, null_oid(), path);
char *key;
const char *val;
key = xstrfmt("submodule.%s.update", sub->name);
if (update) {
if (parse_submodule_update_strategy(update, out) < 0)
die(_("Invalid update mode '%s' for submodule path '%s'"),
update, path);
config: fix leaks from git_config_get_string_const() There are two functions to get a single config string: - git_config_get_string() - git_config_get_string_const() One might naively think that the first one allocates a new string and the second one just points us to the internal configset storage. But in fact they both allocate a new copy; the second one exists only to avoid having to cast when using it with a const global which we never intend to free. The documentation for the function explains that clearly, but it seems I'm not alone in being surprised by this. Of 17 calls to the function, 13 of them leak the resulting value. We could obviously fix these by adding the appropriate free(). But it would be simpler still if we actually had a non-allocating way to get the string. There's git_config_get_value() but that doesn't quite do what we want. If the config key is present but is a boolean with no value (e.g., "[foo]bar" in the file), then we'll get NULL (whereas the string versions will print an error and die). So let's introduce a new variant, git_config_get_string_tmp(), that behaves as these callers expect. We need a new name because we have new semantics but the same function signature (so even if we converted the four remaining callers, topics in flight might be surprised). The "tmp" is because this value should only be held onto for a short time. In practice it's rare for us to clear and refresh the configset, invalidating the pointer, but hopefully the "tmp" makes callers think about the lifetime. In each of the converted cases here the value only needs to last within the local function or its immediate caller. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-14 19:17:36 +03:00
} else if (!repo_config_get_string_tmp(r, key, &val)) {
if (parse_submodule_update_strategy(val, out) < 0)
die(_("Invalid update mode '%s' configured for submodule path '%s'"),
val, path);
} else if (sub->update_strategy.type != SM_UPDATE_UNSPECIFIED) {
if (sub->update_strategy.type == SM_UPDATE_COMMAND)
BUG("how did we read update = !command from .gitmodules?");
out->type = sub->update_strategy.type;
out->command = sub->update_strategy.command;
} else
out->type = SM_UPDATE_CHECKOUT;
if (just_cloned &&
(out->type == SM_UPDATE_MERGE ||
out->type == SM_UPDATE_REBASE ||
out->type == SM_UPDATE_NONE))
out->type = SM_UPDATE_CHECKOUT;
free(key);
}
struct update_clone_data {
const struct submodule *sub;
struct object_id oid;
unsigned just_cloned;
};
struct submodule_update_clone {
/* index into 'update_data.list', the list of submodules to look into for cloning */
int current;
/* configuration parameters which are passed on to the children */
struct update_data *update_data;
/* to be consumed by update_submodule() */
struct update_clone_data *update_clone;
int update_clone_nr; int update_clone_alloc;
/* If we want to stop as fast as possible and return an error */
unsigned quickstop : 1;
/* failed clones to be retried again */
const struct cache_entry **failed_clones;
int failed_clones_nr, failed_clones_alloc;
};
#define SUBMODULE_UPDATE_CLONE_INIT { 0 }
struct update_data {
const char *prefix;
const char *recursive_prefix;
const char *displaypath;
const char *update_default;
struct object_id suboid;
struct string_list references;
struct submodule_update_strategy update_strategy;
struct list_objects_filter_options *filter_options;
struct module_list list;
int depth;
int max_jobs;
int single_branch;
int recommend_shallow;
unsigned int require_init;
unsigned int force;
unsigned int quiet;
unsigned int nofetch;
unsigned int remote;
unsigned int progress;
unsigned int dissociate;
unsigned int init;
unsigned int warn_if_uninitialized;
unsigned int recursive;
/* copied over from update_clone_data */
struct object_id oid;
unsigned int just_cloned;
const char *sm_path;
};
#define UPDATE_DATA_INIT { \
.update_strategy = SUBMODULE_UPDATE_STRATEGY_INIT, \
.list = MODULE_LIST_INIT, \
.recommend_shallow = -1, \
.references = STRING_LIST_INIT_DUP, \
.single_branch = -1, \
.max_jobs = 1, \
}
static void next_submodule_warn_missing(struct submodule_update_clone *suc,
struct strbuf *out, const char *displaypath)
{
/*
* Only mention uninitialized submodules when their
* paths have been specified.
*/
if (suc->update_data->warn_if_uninitialized) {
strbuf_addf(out,
_("Submodule path '%s' not initialized"),
displaypath);
strbuf_addch(out, '\n');
strbuf_addstr(out,
_("Maybe you want to use 'update --init'?"));
strbuf_addch(out, '\n');
}
}
/**
* Determine whether 'ce' needs to be cloned. If so, prepare the 'child' to
* run the clone. Returns 1 if 'ce' needs to be cloned, 0 otherwise.
*/
static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
struct child_process *child,
struct submodule_update_clone *suc,
struct strbuf *out)
{
const struct submodule *sub = NULL;
const char *url = NULL;
const char *update_string;
enum submodule_update_type update_type;
char *key;
struct strbuf displaypath_sb = STRBUF_INIT;
struct strbuf sb = STRBUF_INIT;
const char *displaypath = NULL;
int needs_cloning = 0;
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
int need_free_url = 0;
if (ce_stage(ce)) {
if (suc->update_data->recursive_prefix)
strbuf_addf(&sb, "%s/%s", suc->update_data->recursive_prefix, ce->name);
else
strbuf_addstr(&sb, ce->name);
strbuf_addf(out, _("Skipping unmerged submodule %s"), sb.buf);
strbuf_addch(out, '\n');
goto cleanup;
}
sub = submodule_from_path(the_repository, null_oid(), ce->name);
if (suc->update_data->recursive_prefix)
displaypath = relative_path(suc->update_data->recursive_prefix,
ce->name, &displaypath_sb);
else
displaypath = ce->name;
if (!sub) {
next_submodule_warn_missing(suc, out, displaypath);
goto cleanup;
}
key = xstrfmt("submodule.%s.update", sub->name);
config: fix leaks from git_config_get_string_const() There are two functions to get a single config string: - git_config_get_string() - git_config_get_string_const() One might naively think that the first one allocates a new string and the second one just points us to the internal configset storage. But in fact they both allocate a new copy; the second one exists only to avoid having to cast when using it with a const global which we never intend to free. The documentation for the function explains that clearly, but it seems I'm not alone in being surprised by this. Of 17 calls to the function, 13 of them leak the resulting value. We could obviously fix these by adding the appropriate free(). But it would be simpler still if we actually had a non-allocating way to get the string. There's git_config_get_value() but that doesn't quite do what we want. If the config key is present but is a boolean with no value (e.g., "[foo]bar" in the file), then we'll get NULL (whereas the string versions will print an error and die). So let's introduce a new variant, git_config_get_string_tmp(), that behaves as these callers expect. We need a new name because we have new semantics but the same function signature (so even if we converted the four remaining callers, topics in flight might be surprised). The "tmp" is because this value should only be held onto for a short time. In practice it's rare for us to clear and refresh the configset, invalidating the pointer, but hopefully the "tmp" makes callers think about the lifetime. In each of the converted cases here the value only needs to last within the local function or its immediate caller. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-14 19:17:36 +03:00
if (!repo_config_get_string_tmp(the_repository, key, &update_string)) {
update_type = parse_submodule_update_type(update_string);
} else {
update_type = sub->update_strategy.type;
}
free(key);
if (suc->update_data->update_strategy.type == SM_UPDATE_NONE
|| (suc->update_data->update_strategy.type == SM_UPDATE_UNSPECIFIED
&& update_type == SM_UPDATE_NONE)) {
strbuf_addf(out, _("Skipping submodule '%s'"), displaypath);
strbuf_addch(out, '\n');
goto cleanup;
}
/* Check if the submodule has been initialized. */
if (!is_submodule_active(the_repository, ce->name)) {
next_submodule_warn_missing(suc, out, displaypath);
goto cleanup;
}
strbuf_reset(&sb);
strbuf_addf(&sb, "submodule.%s.url", sub->name);
config: fix leaks from git_config_get_string_const() There are two functions to get a single config string: - git_config_get_string() - git_config_get_string_const() One might naively think that the first one allocates a new string and the second one just points us to the internal configset storage. But in fact they both allocate a new copy; the second one exists only to avoid having to cast when using it with a const global which we never intend to free. The documentation for the function explains that clearly, but it seems I'm not alone in being surprised by this. Of 17 calls to the function, 13 of them leak the resulting value. We could obviously fix these by adding the appropriate free(). But it would be simpler still if we actually had a non-allocating way to get the string. There's git_config_get_value() but that doesn't quite do what we want. If the config key is present but is a boolean with no value (e.g., "[foo]bar" in the file), then we'll get NULL (whereas the string versions will print an error and die). So let's introduce a new variant, git_config_get_string_tmp(), that behaves as these callers expect. We need a new name because we have new semantics but the same function signature (so even if we converted the four remaining callers, topics in flight might be surprised). The "tmp" is because this value should only be held onto for a short time. In practice it's rare for us to clear and refresh the configset, invalidating the pointer, but hopefully the "tmp" makes callers think about the lifetime. In each of the converted cases here the value only needs to last within the local function or its immediate caller. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-14 19:17:36 +03:00
if (repo_config_get_string_tmp(the_repository, sb.buf, &url)) {
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
if (starts_with_dot_slash(sub->url) ||
starts_with_dot_dot_slash(sub->url)) {
url = resolve_relative_url(sub->url, NULL, 0);
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
need_free_url = 1;
} else
url = sub->url;
}
strbuf_reset(&sb);
strbuf_addf(&sb, "%s/.git", ce->name);
needs_cloning = !file_exists(sb.buf);
ALLOC_GROW(suc->update_clone, suc->update_clone_nr + 1,
suc->update_clone_alloc);
oidcpy(&suc->update_clone[suc->update_clone_nr].oid, &ce->oid);
suc->update_clone[suc->update_clone_nr].just_cloned = needs_cloning;
suc->update_clone[suc->update_clone_nr].sub = sub;
suc->update_clone_nr++;
if (!needs_cloning)
goto cleanup;
child->git_cmd = 1;
child->no_stdin = 1;
child->stdout_to_stderr = 1;
child->err = -1;
strvec_push(&child->args, "submodule--helper");
strvec_push(&child->args, "clone");
if (suc->update_data->progress)
strvec_push(&child->args, "--progress");
if (suc->update_data->quiet)
strvec_push(&child->args, "--quiet");
if (suc->update_data->prefix)
strvec_pushl(&child->args, "--prefix", suc->update_data->prefix, NULL);
if (suc->update_data->recommend_shallow && sub->recommend_shallow == 1)
strvec_push(&child->args, "--depth=1");
else if (suc->update_data->depth)
strvec_pushf(&child->args, "--depth=%d", suc->update_data->depth);
if (suc->update_data->filter_options && suc->update_data->filter_options->choice)
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
strvec_pushf(&child->args, "--filter=%s",
expand_list_objects_filter_spec(suc->update_data->filter_options));
if (suc->update_data->require_init)
strvec_push(&child->args, "--require-init");
strvec_pushl(&child->args, "--path", sub->path, NULL);
strvec_pushl(&child->args, "--name", sub->name, NULL);
strvec_pushl(&child->args, "--url", url, NULL);
if (suc->update_data->references.nr) {
struct string_list_item *item;
for_each_string_list_item(item, &suc->update_data->references)
strvec_pushl(&child->args, "--reference", item->string, NULL);
}
if (suc->update_data->dissociate)
strvec_push(&child->args, "--dissociate");
if (suc->update_data->single_branch >= 0)
strvec_push(&child->args, suc->update_data->single_branch ?
"--single-branch" :
"--no-single-branch");
cleanup:
strbuf_release(&displaypath_sb);
strbuf_release(&sb);
submodule helper: convert relative URL to absolute URL if needed The submodule helper update_clone called by "git submodule update", clones submodules if needed. As submodules used to have the URL indicating if they were active, the step to resolve relative URLs was done in the "submodule init" step. Nowadays submodules can be configured active without calling an explicit init, e.g. via configuring submodule.active. When trying to obtain submodules that are set active this way, we'll fallback to the URL found in the .gitmodules, which may be relative to the superproject, but we do not resolve it, yet: git clone https://gerrit.googlesource.com/gerrit cd gerrit && grep url .gitmodules url = ../plugins/codemirror-editor ... git config submodule.active . git submodule update fatal: repository '../plugins/codemirror-editor' does not exist fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor'. Retry scheduled [...] fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed Failed to clone 'plugins/codemirror-editor' a second time, aborting [...] To resolve the issue, factor out the function that resolves the relative URLs in "git submodule init" (in the submodule helper in the init_submodule function) and call it at the appropriate place in the update_clone helper. Reported-by: Jaewoong Jung <jungjw@google.com> Signed-off-by: Stefan Beller <sbeller@google.com> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-16 20:27:03 +03:00
if (need_free_url)
free((void*)url);
return needs_cloning;
}
static int update_clone_get_next_task(struct child_process *child,
struct strbuf *err,
void *suc_cb,
void **idx_task_cb)
{
struct submodule_update_clone *suc = suc_cb;
const struct cache_entry *ce;
int index;
for (; suc->current < suc->update_data->list.nr; suc->current++) {
ce = suc->update_data->list.entries[suc->current];
if (prepare_to_clone_next_submodule(ce, child, suc, err)) {
int *p = xmalloc(sizeof(*p));
*p = suc->current;
*idx_task_cb = p;
suc->current++;
return 1;
}
}
/*
* The loop above tried cloning each submodule once, now try the
* stragglers again, which we can imagine as an extension of the
* entry list.
*/
index = suc->current - suc->update_data->list.nr;
if (index < suc->failed_clones_nr) {
int *p;
ce = suc->failed_clones[index];
2016-08-10 00:29:13 +03:00
if (!prepare_to_clone_next_submodule(ce, child, suc, err)) {
suc->current ++;
strbuf_addstr(err, "BUG: submodule considered for "
"cloning, doesn't need cloning "
"any more?\n");
2016-08-10 00:29:13 +03:00
return 0;
}
p = xmalloc(sizeof(*p));
*p = suc->current;
*idx_task_cb = p;
suc->current ++;
return 1;
}
return 0;
}
static int update_clone_start_failure(struct strbuf *err,
void *suc_cb,
void *idx_task_cb)
{
struct submodule_update_clone *suc = suc_cb;
suc->quickstop = 1;
return 1;
}
static int update_clone_task_finished(int result,
struct strbuf *err,
void *suc_cb,
void *idx_task_cb)
{
const struct cache_entry *ce;
struct submodule_update_clone *suc = suc_cb;
int *idxP = idx_task_cb;
int idx = *idxP;
free(idxP);
if (!result)
return 0;
if (idx < suc->update_data->list.nr) {
ce = suc->update_data->list.entries[idx];
strbuf_addf(err, _("Failed to clone '%s'. Retry scheduled"),
ce->name);
strbuf_addch(err, '\n');
ALLOC_GROW(suc->failed_clones,
suc->failed_clones_nr + 1,
suc->failed_clones_alloc);
suc->failed_clones[suc->failed_clones_nr++] = ce;
return 0;
} else {
idx -= suc->update_data->list.nr;
ce = suc->failed_clones[idx];
strbuf_addf(err, _("Failed to clone '%s' a second time, aborting"),
ce->name);
strbuf_addch(err, '\n');
suc->quickstop = 1;
return 1;
}
return 0;
}
static int git_update_clone_config(const char *var, const char *value,
void *cb)
{
int *max_jobs = cb;
if (!strcmp(var, "submodule.fetchjobs"))
*max_jobs = parse_submodule_fetchjobs(var, value);
return 0;
}
static int is_tip_reachable(const char *path, struct object_id *oid)
{
struct child_process cp = CHILD_PROCESS_INIT;
struct strbuf rev = STRBUF_INIT;
char *hex = oid_to_hex(oid);
cp.git_cmd = 1;
cp.dir = xstrdup(path);
cp.no_stderr = 1;
strvec_pushl(&cp.args, "rev-list", "-n", "1", hex, "--not", "--all", NULL);
prepare_submodule_repo_env(&cp.env);
if (capture_command(&cp, &rev, GIT_MAX_HEXSZ + 1) || rev.len)
return 0;
return 1;
}
static int fetch_in_submodule(const char *module_path, int depth, int quiet, struct object_id *oid)
{
struct child_process cp = CHILD_PROCESS_INIT;
prepare_submodule_repo_env(&cp.env);
cp.git_cmd = 1;
cp.dir = xstrdup(module_path);
strvec_push(&cp.args, "fetch");
if (quiet)
strvec_push(&cp.args, "--quiet");
if (depth)
strvec_pushf(&cp.args, "--depth=%d", depth);
if (oid) {
char *hex = oid_to_hex(oid);
char *remote = get_default_remote();
strvec_pushl(&cp.args, remote, hex, NULL);
}
return run_command(&cp);
}
static int run_update_command(struct update_data *ud, int subforce)
{
struct child_process cp = CHILD_PROCESS_INIT;
char *oid = oid_to_hex(&ud->oid);
int must_die_on_failure = 0;
switch (ud->update_strategy.type) {
case SM_UPDATE_CHECKOUT:
cp.git_cmd = 1;
strvec_pushl(&cp.args, "checkout", "-q", NULL);
if (subforce)
strvec_push(&cp.args, "-f");
break;
case SM_UPDATE_REBASE:
cp.git_cmd = 1;
strvec_push(&cp.args, "rebase");
if (ud->quiet)
strvec_push(&cp.args, "--quiet");
must_die_on_failure = 1;
break;
case SM_UPDATE_MERGE:
cp.git_cmd = 1;
strvec_push(&cp.args, "merge");
if (ud->quiet)
strvec_push(&cp.args, "--quiet");
must_die_on_failure = 1;
break;
case SM_UPDATE_COMMAND:
cp.use_shell = 1;
strvec_push(&cp.args, ud->update_strategy.command);
must_die_on_failure = 1;
break;
default:
BUG("unexpected update strategy type: %s",
submodule_strategy_to_string(&ud->update_strategy));
}
strvec_push(&cp.args, oid);
cp.dir = xstrdup(ud->sm_path);
prepare_submodule_repo_env(&cp.env);
if (run_command(&cp)) {
switch (ud->update_strategy.type) {
case SM_UPDATE_CHECKOUT:
die_message(_("Unable to checkout '%s' in submodule path '%s'"),
oid, ud->displaypath);
break;
case SM_UPDATE_REBASE:
die_message(_("Unable to rebase '%s' in submodule path '%s'"),
oid, ud->displaypath);
break;
case SM_UPDATE_MERGE:
die_message(_("Unable to merge '%s' in submodule path '%s'"),
oid, ud->displaypath);
break;
case SM_UPDATE_COMMAND:
die_message(_("Execution of '%s %s' failed in submodule path '%s'"),
ud->update_strategy.command, oid, ud->displaypath);
break;
default:
BUG("unexpected update strategy type: %s",
submodule_strategy_to_string(&ud->update_strategy));
}
if (must_die_on_failure)
exit(128);
/* the command failed, but update must continue */
return 1;
}
if (ud->quiet)
return 0;
switch (ud->update_strategy.type) {
case SM_UPDATE_CHECKOUT:
printf(_("Submodule path '%s': checked out '%s'\n"),
ud->displaypath, oid);
break;
case SM_UPDATE_REBASE:
printf(_("Submodule path '%s': rebased into '%s'\n"),
ud->displaypath, oid);
break;
case SM_UPDATE_MERGE:
printf(_("Submodule path '%s': merged in '%s'\n"),
ud->displaypath, oid);
break;
case SM_UPDATE_COMMAND:
printf(_("Submodule path '%s': '%s %s'\n"),
ud->displaypath, ud->update_strategy.command, oid);
break;
default:
BUG("unexpected update strategy type: %s",
submodule_strategy_to_string(&ud->update_strategy));
}
return 0;
}
static int run_update_procedure(struct update_data *ud)
{
int subforce = is_null_oid(&ud->suboid) || ud->force;
if (!ud->nofetch) {
/*
* Run fetch only if `oid` isn't present or it
* is not reachable from a ref.
*/
if (!is_tip_reachable(ud->sm_path, &ud->oid) &&
fetch_in_submodule(ud->sm_path, ud->depth, ud->quiet, NULL) &&
!ud->quiet)
fprintf_ln(stderr,
_("Unable to fetch in submodule path '%s'; "
"trying to directly fetch %s:"),
ud->displaypath, oid_to_hex(&ud->oid));
/*
* Now we tried the usual fetch, but `oid` may
* not be reachable from any of the refs.
*/
if (!is_tip_reachable(ud->sm_path, &ud->oid) &&
fetch_in_submodule(ud->sm_path, ud->depth, ud->quiet, &ud->oid))
die(_("Fetched in submodule path '%s', but it did not "
"contain %s. Direct fetching of that commit failed."),
ud->displaypath, oid_to_hex(&ud->oid));
}
return run_update_command(ud, subforce);
}
static const char *remote_submodule_branch(const char *path)
{
const struct submodule *sub;
const char *branch = NULL;
char *key;
sub = submodule_from_path(the_repository, null_oid(), path);
if (!sub)
return NULL;
key = xstrfmt("submodule.%s.branch", sub->name);
if (repo_config_get_string_tmp(the_repository, key, &branch))
branch = sub->branch;
free(key);
if (!branch)
return "HEAD";
if (!strcmp(branch, ".")) {
const char *refname = resolve_ref_unsafe("HEAD", 0, NULL, NULL);
if (!refname)
die(_("No such ref: %s"), "HEAD");
/* detached HEAD */
if (!strcmp(refname, "HEAD"))
die(_("Submodule (%s) branch configured to inherit "
"branch from superproject, but the superproject "
"is not on any branch"), sub->name);
if (!skip_prefix(refname, "refs/heads/", &refname))
die(_("Expecting a full ref name, got %s"), refname);
return refname;
}
return branch;
}
static void ensure_core_worktree(const char *path)
{
const char *cw;
struct repository subrepo;
if (repo_submodule_init(&subrepo, the_repository, path, null_oid()))
die(_("could not get a repository handle for submodule '%s'"), path);
if (!repo_config_get_string_tmp(&subrepo, "core.worktree", &cw)) {
char *cfg_file, *abs_path;
const char *rel_path;
struct strbuf sb = STRBUF_INIT;
cfg_file = repo_git_path(&subrepo, "config");
abs_path = absolute_pathdup(path);
rel_path = relative_path(abs_path, subrepo.gitdir, &sb);
git_config_set_in_file(cfg_file, "core.worktree", rel_path);
free(cfg_file);
free(abs_path);
strbuf_release(&sb);
}
}
static const char *submodule_update_type_to_label(enum submodule_update_type type)
{
switch (type) {
case SM_UPDATE_CHECKOUT:
return "checkout";
case SM_UPDATE_MERGE:
return "merge";
case SM_UPDATE_REBASE:
return "rebase";
case SM_UPDATE_UNSPECIFIED:
case SM_UPDATE_NONE:
case SM_UPDATE_COMMAND:
break;
}
BUG("unreachable with type %d", type);
}
static void update_data_to_args(struct update_data *update_data, struct strvec *args)
{
strvec_pushl(args, "submodule--helper", "update", "--recursive", NULL);
strvec_pushf(args, "--jobs=%d", update_data->max_jobs);
if (update_data->recursive_prefix)
strvec_pushl(args, "--recursive-prefix",
update_data->recursive_prefix, NULL);
if (update_data->quiet)
strvec_push(args, "--quiet");
if (update_data->force)
strvec_push(args, "--force");
if (update_data->init)
strvec_push(args, "--init");
if (update_data->remote)
strvec_push(args, "--remote");
if (update_data->nofetch)
strvec_push(args, "--no-fetch");
if (update_data->dissociate)
strvec_push(args, "--dissociate");
if (update_data->progress)
strvec_push(args, "--progress");
if (update_data->require_init)
strvec_push(args, "--require-init");
if (update_data->depth)
strvec_pushf(args, "--depth=%d", update_data->depth);
if (update_data->update_default)
strvec_pushl(args, "--update", update_data->update_default, NULL);
if (update_data->references.nr) {
struct string_list_item *item;
for_each_string_list_item(item, &update_data->references)
strvec_pushl(args, "--reference", item->string, NULL);
}
if (update_data->filter_options && update_data->filter_options->choice)
strvec_pushf(args, "--filter=%s",
expand_list_objects_filter_spec(
update_data->filter_options));
if (update_data->recommend_shallow == 0)
strvec_push(args, "--no-recommend-shallow");
else if (update_data->recommend_shallow == 1)
strvec_push(args, "--recommend-shallow");
if (update_data->single_branch >= 0)
strvec_push(args, update_data->single_branch ?
"--single-branch" :
"--no-single-branch");
}
static int update_submodule(struct update_data *update_data)
{
char *prefixed_path;
ensure_core_worktree(update_data->sm_path);
if (update_data->recursive_prefix)
prefixed_path = xstrfmt("%s%s", update_data->recursive_prefix,
update_data->sm_path);
else
prefixed_path = xstrdup(update_data->sm_path);
update_data->displaypath = get_submodule_displaypath(prefixed_path,
update_data->prefix);
free(prefixed_path);
determine_submodule_update_strategy(the_repository, update_data->just_cloned,
update_data->sm_path, update_data->update_default,
&update_data->update_strategy);
if (update_data->just_cloned)
oidcpy(&update_data->suboid, null_oid());
else if (resolve_gitlink_ref(update_data->sm_path, "HEAD", &update_data->suboid))
die(_("Unable to find current revision in submodule path '%s'"),
update_data->displaypath);
if (update_data->remote) {
char *remote_name = get_default_remote_submodule(update_data->sm_path);
const char *branch = remote_submodule_branch(update_data->sm_path);
char *remote_ref = xstrfmt("refs/remotes/%s/%s", remote_name, branch);
if (!update_data->nofetch) {
if (fetch_in_submodule(update_data->sm_path, update_data->depth,
0, NULL))
die(_("Unable to fetch in submodule path '%s'"),
update_data->sm_path);
}
if (resolve_gitlink_ref(update_data->sm_path, remote_ref, &update_data->oid))
die(_("Unable to find %s revision in submodule path '%s'"),
remote_ref, update_data->sm_path);
free(remote_ref);
}
if (!oideq(&update_data->oid, &update_data->suboid) || update_data->force)
if (run_update_procedure(update_data))
return 1;
if (update_data->recursive) {
struct child_process cp = CHILD_PROCESS_INIT;
struct update_data next = *update_data;
int res;
if (update_data->recursive_prefix)
prefixed_path = xstrfmt("%s%s/", update_data->recursive_prefix,
update_data->sm_path);
else
prefixed_path = xstrfmt("%s/", update_data->sm_path);
next.recursive_prefix = get_submodule_displaypath(prefixed_path,
update_data->prefix);
next.prefix = NULL;
oidcpy(&next.oid, null_oid());
oidcpy(&next.suboid, null_oid());
cp.dir = update_data->sm_path;
cp.git_cmd = 1;
prepare_submodule_repo_env(&cp.env);
update_data_to_args(&next, &cp.args);
/* die() if child process die()'d */
res = run_command(&cp);
if (!res)
return 0;
die_message(_("Failed to recurse into submodule path '%s'"),
update_data->displaypath);
if (res == 128)
exit(res);
else if (res)
return 1;
}
return 0;
}
static int update_submodules(struct update_data *update_data)
{
int i, res = 0;
struct submodule_update_clone suc = SUBMODULE_UPDATE_CLONE_INIT;
suc.update_data = update_data;
run_processes_parallel_tr2(suc.update_data->max_jobs, update_clone_get_next_task,
update_clone_start_failure,
update_clone_task_finished, &suc, "submodule",
"parallel/update");
/*
* We saved the output and put it out all at once now.
* That means:
* - the listener does not have to interleave their (checkout)
* work with our fetching. The writes involved in a
* checkout involve more straightforward sequential I/O.
* - the listener can avoid doing any work if fetching failed.
*/
if (suc.quickstop) {
res = 1;
goto cleanup;
}
for (i = 0; i < suc.update_clone_nr; i++) {
struct update_clone_data ucd = suc.update_clone[i];
oidcpy(&update_data->oid, &ucd.oid);
update_data->just_cloned = ucd.just_cloned;
update_data->sm_path = ucd.sub->path;
if (update_submodule(update_data))
res = 1;
}
cleanup:
string_list_clear(&update_data->references, 0);
return res;
}
static int module_update(int argc, const char **argv, const char *prefix)
{
struct pathspec pathspec;
struct update_data opt = UPDATE_DATA_INIT;
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
struct list_objects_filter_options filter_options;
int ret;
enum submodule_update_type update_type = SM_UPDATE_UNSPECIFIED;
struct option module_update_options[] = {
OPT__FORCE(&opt.force, N_("force checkout updates"), 0),
OPT_BOOL(0, "init", &opt.init,
N_("initialize uninitialized submodules before update")),
OPT_BOOL(0, "remote", &opt.remote,
N_("use SHA-1 of submodule's remote tracking branch")),
OPT_BOOL(0, "recursive", &opt.recursive,
N_("traverse submodules recursively")),
OPT_BOOL('N', "no-fetch", &opt.nofetch,
N_("don't fetch new objects from the remote site")),
OPT_STRING(0, "prefix", &opt.prefix,
N_("path"),
N_("path into the working tree")),
OPT_STRING(0, "recursive-prefix", &opt.recursive_prefix,
N_("path"),
N_("path into the working tree, across nested "
"submodule boundaries")),
OPT_STRING(0, "update", &opt.update_default,
N_("string"),
N_("rebase, merge, checkout or none")),
OPT_SET_INT(0, "checkout", &update_type,
N_("use the 'checkout' update strategy (default)"),
SM_UPDATE_CHECKOUT),
OPT_SET_INT('m', "merge", &update_type,
N_("use the 'merge' update strategy"),
SM_UPDATE_MERGE),
OPT_SET_INT('r', "rebase", &update_type,
N_("use the 'rebase' update strategy"),
SM_UPDATE_REBASE),
OPT_STRING_LIST(0, "reference", &opt.references, N_("repo"),
N_("reference repository")),
OPT_BOOL(0, "dissociate", &opt.dissociate,
N_("use --reference only while cloning")),
OPT_INTEGER(0, "depth", &opt.depth,
N_("create a shallow clone truncated to the "
"specified number of revisions")),
OPT_INTEGER('j', "jobs", &opt.max_jobs,
N_("parallel jobs")),
OPT_BOOL(0, "recommend-shallow", &opt.recommend_shallow,
N_("whether the initial clone should follow the shallow recommendation")),
OPT__QUIET(&opt.quiet, N_("don't print cloning progress")),
OPT_BOOL(0, "progress", &opt.progress,
clone: pass --progress decision to recursive submodules When cloning with "--recursive", we'd generally expect submodules to show progress reports if the main clone did, too. In older versions of git, this mostly worked out of the box. Since we show progress by default when stderr is a tty, and since the child clones inherit the parent stderr, then both processes would come to the same decision by default. If the parent clone was asked for "--quiet", we passed down "--quiet" to the child. However, if stderr was not a tty and the user specified "--progress", we did not propagate this to the child. That's a minor bug, but things got much worse when we switched recently to submodule--helper's update_clone command. With that change, the stderr of the child clones are always connected to a pipe, and we never output progress at all. This patch teaches git-submodule and git-submodule--helper how to pass down an explicit "--progress" flag when cloning. The clone command then decides to propagate that flag based on the cloning decision made earlier (which takes into account isatty(2) of the parent process, existing --progress or --quiet flags, etc). Since the child processes always run without a tty on stderr, we don't have to worry about passing an explicit "--no-progress"; it's the default for them. This fixes the recent loss of progress during recursive clones. And as a bonus, it makes: git clone --recursive --progress ... 2>&1 | cat work by triggering progress explicitly in the children. Signed-off-by: Jeff King <peff@peff.net> Acked-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-09-22 08:24:46 +03:00
N_("force cloning progress")),
OPT_BOOL(0, "require-init", &opt.require_init,
N_("disallow cloning into non-empty directory, implies --init")),
OPT_BOOL(0, "single-branch", &opt.single_branch,
N_("clone only one branch, HEAD or --branch")),
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
OPT_PARSE_LIST_OBJECTS_FILTER(&filter_options),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule [--quiet] update"
" [--init [--filter=<filter-spec>]] [--remote]"
" [-N|--no-fetch] [-f|--force]"
" [--checkout|--merge|--rebase]"
" [--[no-]recommend-shallow] [--reference <repository>]"
" [--recursive] [--[no-]single-branch] [--] [<path>...]"),
NULL
};
update_clone_config_from_gitmodules(&opt.max_jobs);
git_config(git_update_clone_config, &opt.max_jobs);
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
memset(&filter_options, 0, sizeof(filter_options));
argc = parse_options(argc, argv, prefix, module_update_options,
git_submodule_helper_usage, 0);
if (opt.require_init)
opt.init = 1;
if (filter_options.choice && !opt.init) {
usage_with_options(git_submodule_helper_usage,
module_update_options);
}
opt.filter_options = &filter_options;
if (update_type != SM_UPDATE_UNSPECIFIED)
opt.update_default = submodule_update_type_to_label(update_type);
if (opt.update_default)
if (parse_submodule_update_strategy(opt.update_default,
&opt.update_strategy) < 0)
die(_("bad value for update parameter"));
if (module_list_compute(argc, argv, prefix, &pathspec, &opt.list) < 0) {
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
list_objects_filter_release(&filter_options);
return 1;
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
}
if (pathspec.nr)
opt.warn_if_uninitialized = 1;
if (opt.init) {
struct module_list list = MODULE_LIST_INIT;
struct init_cb info = INIT_CB_INIT;
if (module_list_compute(argc, argv, opt.prefix,
&pathspec, &list) < 0)
return 1;
/*
* If there are no path args and submodule.active is set then,
* by default, only initialize 'active' modules.
*/
if (!argc && git_config_get_value_multi("submodule.active"))
module_list_active(&list);
info.prefix = opt.prefix;
info.superprefix = opt.recursive_prefix;
if (opt.quiet)
info.flags |= OPT_QUIET;
for_each_listed_submodule(&list, init_submodule_cb, &info);
}
ret = update_submodules(&opt);
clone, submodule: pass partial clone filters to submodules When cloning a repo with a --filter and with --recurse-submodules enabled, the partial clone filter only applies to the top-level repo. This can lead to unexpected bandwidth and disk usage for projects which include large submodules. For example, a user might wish to make a partial clone of Gerrit and would run: `git clone --recurse-submodules --filter=blob:5k https://gerrit.googlesource.com/gerrit`. However, only the superproject would be a partial clone; all the submodules would have all blobs downloaded regardless of their size. With this change, the same filter can also be applied to submodules, meaning the expected bandwidth and disk savings apply consistently. To avoid changing default behavior, add a new clone flag, `--also-filter-submodules`. When this is set along with `--filter` and `--recurse-submodules`, the filter spec is passed along to git-submodule and git-submodule--helper, such that submodule clones also have the filter applied. This applies the same filter to the superproject and all submodules. Users who need to customize the filter per-submodule would need to clone with `--no-recurse-submodules` and then manually initialize each submodule with the proper filter. Applying filters to submodules should be safe thanks to Jonathan Tan's recent work [1, 2, 3] eliminating the use of alternates as a method of accessing submodule objects, so any submodule object access now triggers a lazy fetch from the submodule's promisor remote if the accessed object is missing. This patch is a reworked version of [4], which was created prior to Jonathan Tan's work. [1]: 8721e2e (Merge branch 'jt/partial-clone-submodule-1', 2021-07-16) [2]: 11e5d0a (Merge branch 'jt/grep-wo-submodule-odb-as-alternate', 2021-09-20) [3]: 162a13b (Merge branch 'jt/no-abuse-alternate-odb-for-submodules', 2021-10-25) [4]: https://lore.kernel.org/git/52bf9d45b8e2b72ff32aa773f2415bf7b2b86da2.1563322192.git.steadmon@google.com/ Signed-off-by: Josh Steadmon <steadmon@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-02-05 08:00:49 +03:00
list_objects_filter_release(&filter_options);
return ret;
}
static int push_check(int argc, const char **argv, const char *prefix)
{
struct remote *remote;
const char *superproject_head;
char *head;
int detached_head = 0;
struct object_id head_oid;
if (argc < 3)
die("submodule--helper push-check requires at least 2 arguments");
/*
* superproject's resolved head ref.
* if HEAD then the superproject is in a detached head state, otherwise
* it will be the resolved head ref.
*/
superproject_head = argv[1];
argv++;
argc--;
/* Get the submodule's head ref and determine if it is detached */
head = resolve_refdup("HEAD", 0, &head_oid, NULL);
if (!head)
die(_("Failed to resolve HEAD as a valid ref."));
if (!strcmp(head, "HEAD"))
detached_head = 1;
/*
* The remote must be configured.
* This is to avoid pushing to the exact same URL as the parent.
*/
remote = pushremote_get(argv[1]);
if (!remote || remote->origin == REMOTE_UNCONFIGURED)
die("remote '%s' not configured", argv[1]);
/* Check the refspec */
if (argc > 2) {
int i;
struct ref *local_refs = get_local_heads();
struct refspec refspec = REFSPEC_INIT_PUSH;
refspec_appendn(&refspec, argv + 2, argc - 2);
for (i = 0; i < refspec.nr; i++) {
const struct refspec_item *rs = &refspec.items[i];
if (rs->pattern || rs->matching)
continue;
/* LHS must match a single ref */
switch (count_refspec_match(rs->src, local_refs, NULL)) {
case 1:
break;
case 0:
/*
* If LHS matches 'HEAD' then we need to ensure
* that it matches the same named branch
* checked out in the superproject.
*/
if (!strcmp(rs->src, "HEAD")) {
if (!detached_head &&
!strcmp(head, superproject_head))
break;
die("HEAD does not match the named branch in the superproject");
}
consistently use "fallthrough" comments in switches Gcc 7 adds -Wimplicit-fallthrough, which can warn when a switch case falls through to the next case. The general idea is that the compiler can't tell if this was intentional or not, so you should annotate any intentional fall-throughs as such, leaving it to complain about any unannotated ones. There's a GNU __attribute__ which can be used for annotation, but of course we'd have to #ifdef it away on non-gcc compilers. Gcc will also recognize specially-formatted comments, which matches our current practice. Let's extend that practice to all of the unannotated sites (which I did look over and verify that they were behaving as intended). Ideally in each case we'd actually give some reasons in the comment about why we're falling through, or what we're falling through to. And gcc does support that with -Wimplicit-fallthrough=2, which relaxes the comment pattern matching to anything that contains "fallthrough" (or a variety of spelling variants). However, this isn't the default for -Wimplicit-fallthrough, nor for -Wextra. In the name of simplicity, it's probably better for us to support the default level, which requires "fallthrough" to be the only thing in the comment (modulo some window dressing like "else" and some punctuation; see the gcc manual for the complete set of patterns). This patch suppresses all warnings due to -Wimplicit-fallthrough. We might eventually want to add that to the DEVELOPER Makefile knob, but we should probably wait until gcc 7 is more widely adopted (since earlier versions will complain about the unknown warning type). Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-09-21 09:25:41 +03:00
/* fallthrough */
default:
die("src refspec '%s' must name a ref",
rs->src);
}
}
refspec_clear(&refspec);
}
free(head);
return 0;
}
static int absorb_git_dirs(int argc, const char **argv, const char *prefix)
{
int i;
struct pathspec pathspec;
struct module_list list = MODULE_LIST_INIT;
unsigned flags = ABSORB_GITDIR_RECURSE_SUBMODULES;
struct option embed_gitdir_options[] = {
OPT_STRING(0, "prefix", &prefix,
N_("path"),
N_("path into the working tree")),
OPT_BIT(0, "--recursive", &flags, N_("recurse into submodules"),
ABSORB_GITDIR_RECURSE_SUBMODULES),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule absorbgitdirs [<options>] [<path>...]"),
NULL
};
argc = parse_options(argc, argv, prefix, embed_gitdir_options,
git_submodule_helper_usage, 0);
if (module_list_compute(argc, argv, prefix, &pathspec, &list) < 0)
return 1;
for (i = 0; i < list.nr; i++)
absorb_git_dir_into_superproject(list.entries[i]->name, flags);
return 0;
}
static int is_active(int argc, const char **argv, const char *prefix)
{
if (argc != 2)
die("submodule--helper is-active takes exactly 1 argument");
return !is_submodule_active(the_repository, argv[1]);
}
submodule-config: verify submodule names as paths Submodule "names" come from the untrusted .gitmodules file, but we blindly append them to $GIT_DIR/modules to create our on-disk repo paths. This means you can do bad things by putting "../" into the name (among other things). Let's sanity-check these names to avoid building a path that can be exploited. There are two main decisions: 1. What should the allowed syntax be? It's tempting to reuse verify_path(), since submodule names typically come from in-repo paths. But there are two reasons not to: a. It's technically more strict than what we need, as we really care only about breaking out of the $GIT_DIR/modules/ hierarchy. E.g., having a submodule named "foo/.git" isn't actually dangerous, and it's possible that somebody has manually given such a funny name. b. Since we'll eventually use this checking logic in fsck to prevent downstream repositories, it should be consistent across platforms. Because verify_path() relies on is_dir_sep(), it wouldn't block "foo\..\bar" on a non-Windows machine. 2. Where should we enforce it? These days most of the .gitmodules reads go through submodule-config.c, so I've put it there in the reading step. That should cover all of the C code. We also construct the name for "git submodule add" inside the git-submodule.sh script. This is probably not a big deal for security since the name is coming from the user anyway, but it would be polite to remind them if the name they pick is invalid (and we need to expose the name-checker to the shell anyway for our test scripts). This patch issues a warning when reading .gitmodules and just ignores the related config entry completely. This will generally end up producing a sensible error, as it works the same as a .gitmodules file which is missing a submodule entry (so "submodule update" will barf, but "git clone --recurse-submodules" will print an error but not abort the clone. There is one minor oddity, which is that we print the warning once per malformed config key (since that's how the config subsystem gives us the entries). So in the new test, for example, the user would see three warnings. That's OK, since the intent is that this case should never come up outside of malicious repositories (and then it might even benefit the user to see the message multiple times). Credit for finding this vulnerability and the proof of concept from which the test script was adapted goes to Etienne Stalmans. Signed-off-by: Jeff King <peff@peff.net>
2018-04-30 10:25:25 +03:00
/*
* Exit non-zero if any of the submodule names given on the command line is
* invalid. If no names are given, filter stdin to print only valid names
* (which is primarily intended for testing).
*/
static int check_name(int argc, const char **argv, const char *prefix)
{
if (argc > 1) {
while (*++argv) {
if (check_submodule_name(*argv) < 0)
return 1;
}
} else {
struct strbuf buf = STRBUF_INIT;
while (strbuf_getline(&buf, stdin) != EOF) {
if (!check_submodule_name(buf.buf))
printf("%s\n", buf.buf);
}
strbuf_release(&buf);
}
return 0;
}
static int module_config(int argc, const char **argv, const char *prefix)
{
enum {
CHECK_WRITEABLE = 1,
DO_UNSET = 2
} command = 0;
struct option module_config_options[] = {
OPT_CMDMODE(0, "check-writeable", &command,
N_("check if it is safe to write to the .gitmodules file"),
CHECK_WRITEABLE),
OPT_CMDMODE(0, "unset", &command,
N_("unset the config in the .gitmodules file"),
DO_UNSET),
OPT_END()
};
const char *const git_submodule_helper_usage[] = {
N_("git submodule--helper config <name> [<value>]"),
N_("git submodule--helper config --unset <name>"),
"git submodule--helper config --check-writeable",
NULL
};
argc = parse_options(argc, argv, prefix, module_config_options,
git_submodule_helper_usage, PARSE_OPT_KEEP_ARGV0);
if (argc == 1 && command == CHECK_WRITEABLE)
return is_writing_gitmodules_ok() ? 0 : -1;
/* Equivalent to ACTION_GET in builtin/config.c */
if (argc == 2 && command != DO_UNSET)
return print_config_from_gitmodules(the_repository, argv[1]);
/* Equivalent to ACTION_SET in builtin/config.c */
if (argc == 3 || (argc == 2 && command == DO_UNSET)) {
const char *value = (argc == 3) ? argv[2] : NULL;
submodule: support reading .gitmodules when it's not in the working tree When the .gitmodules file is not available in the working tree, try using the content from the index and from the current branch. This covers the case when the file is part of the repository but for some reason it is not checked out, for example because of a sparse checkout. This makes it possible to use at least the 'git submodule' commands which *read* the gitmodules configuration file without fully populating the working tree. Writing to .gitmodules will still require that the file is checked out, so check for that before calling config_set_in_gitmodules_file_gently. Add a similar check also in git-submodule.sh::cmd_add() to anticipate the eventual failure of the "git submodule add" command when .gitmodules is not safely writeable; this prevents the command from leaving the repository in a spurious state (e.g. the submodule repository was cloned but .gitmodules was not updated because config_set_in_gitmodules_file_gently failed). Moreover, since config_from_gitmodules() now accesses the global object store, it is necessary to protect all code paths which call the function against concurrent access to the global object store. Currently this only happens in builtin/grep.c::grep_submodules(), so call grep_read_lock() before invoking code involving config_from_gitmodules(). Finally, add t7418-submodule-sparse-gitmodules.sh to verify that reading from .gitmodules succeeds and that writing to it fails when the file is not checked out. NOTE: there is one rare case where this new feature does not work properly yet: nested submodules without .gitmodules in their working tree. This has been documented with a warning and a test_expect_failure item in t7814, and in this case the current behavior is not altered: no config is read. Signed-off-by: Antonio Ospite <ao2@ao2.it> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-25 19:18:12 +03:00
if (!is_writing_gitmodules_ok())
die(_("please make sure that the .gitmodules file is in the working tree"));
return config_set_in_gitmodules_file_gently(argv[1], value);
submodule: support reading .gitmodules when it's not in the working tree When the .gitmodules file is not available in the working tree, try using the content from the index and from the current branch. This covers the case when the file is part of the repository but for some reason it is not checked out, for example because of a sparse checkout. This makes it possible to use at least the 'git submodule' commands which *read* the gitmodules configuration file without fully populating the working tree. Writing to .gitmodules will still require that the file is checked out, so check for that before calling config_set_in_gitmodules_file_gently. Add a similar check also in git-submodule.sh::cmd_add() to anticipate the eventual failure of the "git submodule add" command when .gitmodules is not safely writeable; this prevents the command from leaving the repository in a spurious state (e.g. the submodule repository was cloned but .gitmodules was not updated because config_set_in_gitmodules_file_gently failed). Moreover, since config_from_gitmodules() now accesses the global object store, it is necessary to protect all code paths which call the function against concurrent access to the global object store. Currently this only happens in builtin/grep.c::grep_submodules(), so call grep_read_lock() before invoking code involving config_from_gitmodules(). Finally, add t7418-submodule-sparse-gitmodules.sh to verify that reading from .gitmodules succeeds and that writing to it fails when the file is not checked out. NOTE: there is one rare case where this new feature does not work properly yet: nested submodules without .gitmodules in their working tree. This has been documented with a warning and a test_expect_failure item in t7814, and in this case the current behavior is not altered: no config is read. Signed-off-by: Antonio Ospite <ao2@ao2.it> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-25 19:18:12 +03:00
}
usage_with_options(git_submodule_helper_usage, module_config_options);
}
static int module_set_url(int argc, const char **argv, const char *prefix)
{
int quiet = 0;
const char *newurl;
const char *path;
char *config_name;
struct option options[] = {
OPT__QUIET(&quiet, N_("suppress output for setting url of a submodule")),
OPT_END()
};
const char *const usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule set-url [--quiet] <path> <newurl>"),
NULL
};
argc = parse_options(argc, argv, prefix, options, usage, 0);
if (argc != 2 || !(path = argv[0]) || !(newurl = argv[1]))
usage_with_options(usage, options);
config_name = xstrfmt("submodule.%s.url", path);
config_set_in_gitmodules_file_gently(config_name, newurl);
sync_submodule(path, prefix, quiet ? OPT_QUIET : 0);
free(config_name);
return 0;
}
static int module_set_branch(int argc, const char **argv, const char *prefix)
{
int opt_default = 0, ret;
const char *opt_branch = NULL;
const char *path;
char *config_name;
/*
* We accept the `quiet` option for uniformity across subcommands,
* though there is nothing to make less verbose in this subcommand.
*/
struct option options[] = {
OPT_NOOP_NOARG('q', "quiet"),
OPT_BOOL('d', "default", &opt_default,
N_("set the default tracking branch to master")),
OPT_STRING('b', "branch", &opt_branch, N_("branch"),
N_("set the default tracking branch")),
OPT_END()
};
const char *const usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule set-branch [-q|--quiet] (-d|--default) <path>"),
N_("git submodule set-branch [-q|--quiet] (-b|--branch) <branch> <path>"),
NULL
};
argc = parse_options(argc, argv, prefix, options, usage, 0);
if (!opt_branch && !opt_default)
die(_("--branch or --default required"));
if (opt_branch && opt_default)
die(_("options '%s' and '%s' cannot be used together"), "--branch", "--default");
if (argc != 1 || !(path = argv[0]))
usage_with_options(usage, options);
config_name = xstrfmt("submodule.%s.branch", path);
ret = config_set_in_gitmodules_file_gently(config_name, opt_branch);
free(config_name);
return !!ret;
}
branch: add --recurse-submodules option for branch creation To improve the submodules UX, we would like to teach Git to handle branches in submodules. Start this process by teaching "git branch" the --recurse-submodules option so that "git branch --recurse-submodules topic" will create the `topic` branch in the superproject and its submodules. Although this commit does not introduce breaking changes, it does not work well with existing --recurse-submodules commands because "git branch --recurse-submodules" writes to the submodule ref store, but most commands only consider the superproject gitlink and ignore the submodule ref store. For example, "git checkout --recurse-submodules" will check out the commits in the superproject gitlinks (and put the submodules in detached HEAD) instead of checking out the submodule branches. Because of this, this commit introduces a new configuration value, `submodule.propagateBranches`. The plan is for Git commands to prioritize submodule ref store information over superproject gitlinks if this value is true. Because "git branch --recurse-submodules" writes to submodule ref stores, for the sake of clarity, it will not function unless this configuration value is set. This commit also includes changes that support working with submodules from a superproject commit because "branch --recurse-submodules" (and future commands) need to read .gitmodules and gitlinks from the superproject commit, but submodules are typically read from the filesystem's .gitmodules and the index's gitlinks. These changes are: * add a submodules_of_tree() helper that gives the relevant information of an in-tree submodule (e.g. path and oid) and initializes the repository * add is_tree_submodule_active() by adding a treeish_name parameter to is_submodule_active() * add the "submoduleNotUpdated" advice to advise users to update the submodules in their trees Incidentally, fix an incorrect usage string that combined the 'list' usage of git branch (-l) with the 'create' usage; this string has been incorrect since its inception, a8dfd5eac4 (Make builtin-branch.c use parse_options., 2007-10-07). Helped-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Glen Choo <chooglen@google.com> Reviewed-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-01-29 03:04:45 +03:00
static int module_create_branch(int argc, const char **argv, const char *prefix)
{
enum branch_track track;
int quiet = 0, force = 0, reflog = 0, dry_run = 0;
struct option options[] = {
OPT__QUIET(&quiet, N_("print only error messages")),
OPT__FORCE(&force, N_("force creation"), 0),
OPT_BOOL(0, "create-reflog", &reflog,
N_("create the branch's reflog")),
OPT_CALLBACK_F('t', "track", &track, "(direct|inherit)",
N_("set branch tracking configuration"),
PARSE_OPT_OPTARG,
parse_opt_tracking_mode),
branch: add --recurse-submodules option for branch creation To improve the submodules UX, we would like to teach Git to handle branches in submodules. Start this process by teaching "git branch" the --recurse-submodules option so that "git branch --recurse-submodules topic" will create the `topic` branch in the superproject and its submodules. Although this commit does not introduce breaking changes, it does not work well with existing --recurse-submodules commands because "git branch --recurse-submodules" writes to the submodule ref store, but most commands only consider the superproject gitlink and ignore the submodule ref store. For example, "git checkout --recurse-submodules" will check out the commits in the superproject gitlinks (and put the submodules in detached HEAD) instead of checking out the submodule branches. Because of this, this commit introduces a new configuration value, `submodule.propagateBranches`. The plan is for Git commands to prioritize submodule ref store information over superproject gitlinks if this value is true. Because "git branch --recurse-submodules" writes to submodule ref stores, for the sake of clarity, it will not function unless this configuration value is set. This commit also includes changes that support working with submodules from a superproject commit because "branch --recurse-submodules" (and future commands) need to read .gitmodules and gitlinks from the superproject commit, but submodules are typically read from the filesystem's .gitmodules and the index's gitlinks. These changes are: * add a submodules_of_tree() helper that gives the relevant information of an in-tree submodule (e.g. path and oid) and initializes the repository * add is_tree_submodule_active() by adding a treeish_name parameter to is_submodule_active() * add the "submoduleNotUpdated" advice to advise users to update the submodules in their trees Incidentally, fix an incorrect usage string that combined the 'list' usage of git branch (-l) with the 'create' usage; this string has been incorrect since its inception, a8dfd5eac4 (Make builtin-branch.c use parse_options., 2007-10-07). Helped-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Glen Choo <chooglen@google.com> Reviewed-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-01-29 03:04:45 +03:00
OPT__DRY_RUN(&dry_run,
N_("show whether the branch would be created")),
OPT_END()
};
const char *const usage[] = {
N_("git submodule--helper create-branch [-f|--force] [--create-reflog] [-q|--quiet] [-t|--track] [-n|--dry-run] <name> <start-oid> <start-name>"),
branch: add --recurse-submodules option for branch creation To improve the submodules UX, we would like to teach Git to handle branches in submodules. Start this process by teaching "git branch" the --recurse-submodules option so that "git branch --recurse-submodules topic" will create the `topic` branch in the superproject and its submodules. Although this commit does not introduce breaking changes, it does not work well with existing --recurse-submodules commands because "git branch --recurse-submodules" writes to the submodule ref store, but most commands only consider the superproject gitlink and ignore the submodule ref store. For example, "git checkout --recurse-submodules" will check out the commits in the superproject gitlinks (and put the submodules in detached HEAD) instead of checking out the submodule branches. Because of this, this commit introduces a new configuration value, `submodule.propagateBranches`. The plan is for Git commands to prioritize submodule ref store information over superproject gitlinks if this value is true. Because "git branch --recurse-submodules" writes to submodule ref stores, for the sake of clarity, it will not function unless this configuration value is set. This commit also includes changes that support working with submodules from a superproject commit because "branch --recurse-submodules" (and future commands) need to read .gitmodules and gitlinks from the superproject commit, but submodules are typically read from the filesystem's .gitmodules and the index's gitlinks. These changes are: * add a submodules_of_tree() helper that gives the relevant information of an in-tree submodule (e.g. path and oid) and initializes the repository * add is_tree_submodule_active() by adding a treeish_name parameter to is_submodule_active() * add the "submoduleNotUpdated" advice to advise users to update the submodules in their trees Incidentally, fix an incorrect usage string that combined the 'list' usage of git branch (-l) with the 'create' usage; this string has been incorrect since its inception, a8dfd5eac4 (Make builtin-branch.c use parse_options., 2007-10-07). Helped-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Glen Choo <chooglen@google.com> Reviewed-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-01-29 03:04:45 +03:00
NULL
};
git_config(git_default_config, NULL);
track = git_branch_track;
argc = parse_options(argc, argv, prefix, options, usage, 0);
if (argc != 3)
usage_with_options(usage, options);
if (!quiet && !dry_run)
printf_ln(_("creating branch '%s'"), argv[0]);
create_branches_recursively(the_repository, argv[0], argv[1], argv[2],
force, reflog, quiet, track, dry_run);
return 0;
}
struct add_data {
const char *prefix;
const char *branch;
const char *reference_path;
char *sm_path;
const char *sm_name;
const char *repo;
const char *realrepo;
int depth;
unsigned int force: 1;
unsigned int quiet: 1;
unsigned int progress: 1;
unsigned int dissociate: 1;
};
#define ADD_DATA_INIT { .depth = -1 }
static void append_fetch_remotes(struct strbuf *msg, const char *git_dir_path)
{
struct child_process cp_remote = CHILD_PROCESS_INIT;
struct strbuf sb_remote_out = STRBUF_INIT;
cp_remote.git_cmd = 1;
strvec_pushf(&cp_remote.env,
"GIT_DIR=%s", git_dir_path);
strvec_push(&cp_remote.env, "GIT_WORK_TREE=.");
strvec_pushl(&cp_remote.args, "remote", "-v", NULL);
if (!capture_command(&cp_remote, &sb_remote_out, 0)) {
char *next_line;
char *line = sb_remote_out.buf;
while ((next_line = strchr(line, '\n')) != NULL) {
size_t len = next_line - line;
if (strip_suffix_mem(line, &len, " (fetch)"))
submodule--helper: fix incorrect newlines in an error message A refactoring[1] done as part of the recent conversion of 'git submodule add' to builtin, changed the error message shown when a Git directory already exists locally for a submodule name. Before the refactoring, the error used to appear like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- After the refactoring the error started appearing like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub fatal: If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or if you are unsure what this means, choose another name with the '--name' option. --- END OF OUTPUT --- As one could observe the remote information is printed along with the first line rather than on its own line. Also, there's an additional newline following output. Make the error message consistent with the error message that used to be printed before the refactoring. This also moves the 'fatal:' prefix that appears in the middle of the error message to the first line as it would more appropriate to have it in the first line. The output after the change would look like: --- START OF OUTPUT --- $ git submodule add ../sub/ subm fatal: A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- [1]: https://lore.kernel.org/git/20210710074801.19917-5-raykar.ath@gmail.com/#t Signed-off-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2021-10-23 15:57:22 +03:00
strbuf_addf(msg, " %.*s\n", (int)len, line);
line = next_line + 1;
}
}
strbuf_release(&sb_remote_out);
}
static int add_submodule(const struct add_data *add_data)
{
char *submod_gitdir_path;
struct module_clone_data clone_data = MODULE_CLONE_DATA_INIT;
/* perhaps the path already exists and is already a git repo, else clone it */
if (is_directory(add_data->sm_path)) {
struct strbuf sm_path = STRBUF_INIT;
strbuf_addstr(&sm_path, add_data->sm_path);
submod_gitdir_path = xstrfmt("%s/.git", add_data->sm_path);
if (is_nonbare_repository_dir(&sm_path))
printf(_("Adding existing repo at '%s' to the index\n"),
add_data->sm_path);
else
die(_("'%s' already exists and is not a valid git repo"),
add_data->sm_path);
strbuf_release(&sm_path);
free(submod_gitdir_path);
} else {
struct child_process cp = CHILD_PROCESS_INIT;
submod_gitdir_path = xstrfmt(".git/modules/%s", add_data->sm_name);
if (is_directory(submod_gitdir_path)) {
if (!add_data->force) {
submodule--helper: fix incorrect newlines in an error message A refactoring[1] done as part of the recent conversion of 'git submodule add' to builtin, changed the error message shown when a Git directory already exists locally for a submodule name. Before the refactoring, the error used to appear like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- After the refactoring the error started appearing like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub fatal: If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or if you are unsure what this means, choose another name with the '--name' option. --- END OF OUTPUT --- As one could observe the remote information is printed along with the first line rather than on its own line. Also, there's an additional newline following output. Make the error message consistent with the error message that used to be printed before the refactoring. This also moves the 'fatal:' prefix that appears in the middle of the error message to the first line as it would more appropriate to have it in the first line. The output after the change would look like: --- START OF OUTPUT --- $ git submodule add ../sub/ subm fatal: A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- [1]: https://lore.kernel.org/git/20210710074801.19917-5-raykar.ath@gmail.com/#t Signed-off-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2021-10-23 15:57:22 +03:00
struct strbuf msg = STRBUF_INIT;
char *die_msg;
strbuf_addf(&msg, _("A git directory for '%s' is found "
"locally with remote(s):\n"),
add_data->sm_name);
append_fetch_remotes(&msg, submod_gitdir_path);
free(submod_gitdir_path);
submodule--helper: fix incorrect newlines in an error message A refactoring[1] done as part of the recent conversion of 'git submodule add' to builtin, changed the error message shown when a Git directory already exists locally for a submodule name. Before the refactoring, the error used to appear like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- After the refactoring the error started appearing like so: --- START OF OUTPUT --- $ git submodule add ../sub/ subm A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub fatal: If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or if you are unsure what this means, choose another name with the '--name' option. --- END OF OUTPUT --- As one could observe the remote information is printed along with the first line rather than on its own line. Also, there's an additional newline following output. Make the error message consistent with the error message that used to be printed before the refactoring. This also moves the 'fatal:' prefix that appears in the middle of the error message to the first line as it would more appropriate to have it in the first line. The output after the change would look like: --- START OF OUTPUT --- $ git submodule add ../sub/ subm fatal: A git directory for 'subm' is found locally with remote(s): origin /me/git-repos-for-test/sub If you want to reuse this local git directory instead of cloning again from /me/git-repos-for-test/sub use the '--force' option. If the local git directory is not the correct repo or you are unsure what this means choose another name with the '--name' option. --- END OF OUTPUT --- [1]: https://lore.kernel.org/git/20210710074801.19917-5-raykar.ath@gmail.com/#t Signed-off-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2021-10-23 15:57:22 +03:00
strbuf_addf(&msg, _("If you want to reuse this local git "
"directory instead of cloning again from\n"
" %s\n"
"use the '--force' option. If the local git "
"directory is not the correct repo\n"
"or you are unsure what this means choose "
"another name with the '--name' option."),
add_data->realrepo);
die_msg = strbuf_detach(&msg, NULL);
die("%s", die_msg);
} else {
printf(_("Reactivating local git directory for "
"submodule '%s'\n"), add_data->sm_name);
}
}
free(submod_gitdir_path);
clone_data.prefix = add_data->prefix;
clone_data.path = add_data->sm_path;
clone_data.name = add_data->sm_name;
clone_data.url = add_data->realrepo;
clone_data.quiet = add_data->quiet;
clone_data.progress = add_data->progress;
if (add_data->reference_path)
string_list_append(&clone_data.reference,
xstrdup(add_data->reference_path));
clone_data.dissociate = add_data->dissociate;
if (add_data->depth >= 0)
clone_data.depth = xstrfmt("%d", add_data->depth);
if (clone_submodule(&clone_data))
return -1;
prepare_submodule_repo_env(&cp.env);
cp.git_cmd = 1;
cp.dir = add_data->sm_path;
Comment important codepaths regarding nuking untracked files/dirs In the last few commits we focused on code in unpack-trees.c that mistakenly removed untracked files or directories. There may be more of those, but in this commit we change our focus: callers of toplevel commands that are expected to remove untracked files or directories. As noted previously, we have toplevel commands that are expected to delete untracked files such as 'read-tree --reset', 'reset --hard', and 'checkout --force'. However, that does not mean that other highlevel commands that happen to call these other commands thought about or conveyed to users the possibility that untracked files could be removed. Audit the code for such callsites, and add comments near existing callsites to mention whether these are safe or not. My auditing is somewhat incomplete, though; it skipped several cases: * git-rebase--preserve-merges.sh: is in the process of being deprecated/removed, so I won't leave a note that there are likely more bugs in that script. * contrib/git-new-workdir: why is the -f flag being used in a new empty directory?? It shouldn't hurt, but it seems useless. * git-p4.py: Don't see why -f is needed for a new dir (maybe it's not and is just superfluous), but I'm not at all familiar with the p4 stuff * git-archimport.perl: Don't care; arch is long since dead * git-cvs*.perl: Don't care; cvs is long since dead Also, the reset --hard in builtin/worktree.c looks safe, due to only running in an empty directory. Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2021-09-27 19:33:47 +03:00
/*
* NOTE: we only get here if add_data->force is true, so
* passing --force to checkout is reasonable.
*/
strvec_pushl(&cp.args, "checkout", "-f", "-q", NULL);
if (add_data->branch) {
strvec_pushl(&cp.args, "-B", add_data->branch, NULL);
strvec_pushf(&cp.args, "origin/%s", add_data->branch);
}
if (run_command(&cp))
die(_("unable to checkout submodule '%s'"), add_data->sm_path);
}
return 0;
}
static int config_submodule_in_gitmodules(const char *name, const char *var, const char *value)
{
char *key;
int ret;
if (!is_writing_gitmodules_ok())
die(_("please make sure that the .gitmodules file is in the working tree"));
key = xstrfmt("submodule.%s.%s", name, var);
ret = config_set_in_gitmodules_file_gently(key, value);
free(key);
return ret;
}
static void configure_added_submodule(struct add_data *add_data)
{
char *key;
char *val = NULL;
struct child_process add_submod = CHILD_PROCESS_INIT;
struct child_process add_gitmodules = CHILD_PROCESS_INIT;
key = xstrfmt("submodule.%s.url", add_data->sm_name);
git_config_set_gently(key, add_data->realrepo);
free(key);
add_submod.git_cmd = 1;
strvec_pushl(&add_submod.args, "add",
"--no-warn-embedded-repo", NULL);
if (add_data->force)
strvec_push(&add_submod.args, "--force");
strvec_pushl(&add_submod.args, "--", add_data->sm_path, NULL);
if (run_command(&add_submod))
die(_("Failed to add submodule '%s'"), add_data->sm_path);
if (config_submodule_in_gitmodules(add_data->sm_name, "path", add_data->sm_path) ||
config_submodule_in_gitmodules(add_data->sm_name, "url", add_data->repo))
die(_("Failed to register submodule '%s'"), add_data->sm_path);
if (add_data->branch) {
if (config_submodule_in_gitmodules(add_data->sm_name,
"branch", add_data->branch))
die(_("Failed to register submodule '%s'"), add_data->sm_path);
}
add_gitmodules.git_cmd = 1;
strvec_pushl(&add_gitmodules.args,
"add", "--force", "--", ".gitmodules", NULL);
if (run_command(&add_gitmodules))
die(_("Failed to register submodule '%s'"), add_data->sm_path);
/*
* NEEDSWORK: In a multi-working-tree world this needs to be
* set in the per-worktree config.
*/
/*
* NEEDSWORK: In the longer run, we need to get rid of this
* pattern of querying "submodule.active" before calling
* is_submodule_active(), since that function needs to find
* out the value of "submodule.active" again anyway.
*/
if (!git_config_get_string("submodule.active", &val) && val) {
/*
* If the submodule being added isn't already covered by the
* current configured pathspec, set the submodule's active flag
*/
if (!is_submodule_active(the_repository, add_data->sm_path)) {
key = xstrfmt("submodule.%s.active", add_data->sm_name);
git_config_set_gently(key, "true");
free(key);
}
} else {
key = xstrfmt("submodule.%s.active", add_data->sm_name);
git_config_set_gently(key, "true");
free(key);
}
}
static void die_on_index_match(const char *path, int force)
{
struct pathspec ps;
const char *args[] = { path, NULL };
parse_pathspec(&ps, 0, PATHSPEC_PREFER_CWD, NULL, args);
if (read_cache_preload(NULL) < 0)
die(_("index file corrupt"));
if (ps.nr) {
int i;
char *ps_matched = xcalloc(ps.nr, 1);
/* TODO: audit for interaction with sparse-index. */
ensure_full_index(&the_index);
/*
* Since there is only one pathspec, we just need
* need to check ps_matched[0] to know if a cache
* entry matched.
*/
for (i = 0; i < active_nr; i++) {
ce_path_match(&the_index, active_cache[i], &ps,
ps_matched);
if (ps_matched[0]) {
if (!force)
die(_("'%s' already exists in the index"),
path);
if (!S_ISGITLINK(active_cache[i]->ce_mode))
die(_("'%s' already exists in the index "
"and is not a submodule"), path);
break;
}
}
free(ps_matched);
}
clear_pathspec(&ps);
}
static void die_on_repo_without_commits(const char *path)
{
struct strbuf sb = STRBUF_INIT;
strbuf_addstr(&sb, path);
if (is_nonbare_repository_dir(&sb)) {
struct object_id oid;
if (resolve_gitlink_ref(path, "HEAD", &oid) < 0)
die(_("'%s' does not have a commit checked out"), path);
}
strbuf_release(&sb);
}
static int module_add(int argc, const char **argv, const char *prefix)
{
int force = 0, quiet = 0, progress = 0, dissociate = 0;
struct add_data add_data = ADD_DATA_INIT;
char *to_free = NULL;
struct option options[] = {
OPT_STRING('b', "branch", &add_data.branch, N_("branch"),
N_("branch of repository to add as submodule")),
OPT__FORCE(&force, N_("allow adding an otherwise ignored submodule path"),
PARSE_OPT_NOCOMPLETE),
OPT__QUIET(&quiet, N_("print only error messages")),
OPT_BOOL(0, "progress", &progress, N_("force cloning progress")),
OPT_STRING(0, "reference", &add_data.reference_path, N_("repository"),
N_("reference repository")),
OPT_BOOL(0, "dissociate", &dissociate, N_("borrow the objects from reference repositories")),
OPT_STRING(0, "name", &add_data.sm_name, N_("name"),
i18n CI: stop allowing non-ASCII source messages in po/git.pot In the preceding commit we moved away from using xgettext(1) to both generate the po/git.pot, and to merge the incrementally generated po/git.pot+ file as we sourced translations from C, shell and Perl. Doing it this way, which dates back to my initial implementation[1][2][3] was conflating two things: With xgettext(1) the --from-code both controls what encoding is specified in the po/git.pot's header, and what encoding we allow in source messages. We don't ever want to allow non-ASCII in *source messages*, and doing so has hid e.g. a buggy message introduced in a6226fd772b (submodule--helper: convert the bulk of cmd_add() to C, 2021-08-10) from us, we'd warn about it before, but only when running "make pot", but the operation would still succeed. Now we'll error out on it when running "make pot". Since the preceding Makefile changes made this easy: let's add a "make check-pot" target with the same prerequisites as the "po/git.pot" target, but without changing the file "po/git.pot". Running it as part of the "static-analysis" CI target will ensure that we catch any such issues in the future. E.g.: $ make check-pot XGETTEXT .build/pot/po/builtin/submodule--helper.c.po xgettext: Non-ASCII string at builtin/submodule--helper.c:3381. Please specify the source encoding through --from-code. make: *** [.build/pot/po/builtin/submodule--helper.c.po] Error 1 1. cd5513a7168 (i18n: Makefile: "pot" target to extract messages marked for translation, 2011-02-22) 2. adc3b2b2767 (Makefile: add xgettext target for *.sh files, 2011-05-14) 3. 5e9637c6297 (i18n: add infrastructure for translating Git with gettext, 2011-11-18) Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Jiang Xin <zhiyou.jx@alibaba-inc.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-05-26 17:50:29 +03:00
N_("sets the submodule's name to the given string "
"instead of defaulting to its path")),
OPT_INTEGER(0, "depth", &add_data.depth, N_("depth for shallow clones")),
OPT_END()
};
const char *const usage[] = {
submodule--helper: report "submodule" as our name in some "-h" output Change the user-facing "git submodule--helper" commands so that they'll report their name as being "git submodule". To a user these commands are internal implementation details, and it doesn't make sense to emit usage about an internal helper when "git submodule" is invoked with invalid options. Before this we'd emit e.g.: $ git submodule absorbgitdirs --blah error: unknown option `blah' usage: git submodule--helper absorbgitdirs [<options>] [<path>...] [...] And: $ git submodule set-url -- -- usage: git submodule--helper set-url [--quiet] <path> <newurl> [...] Now we'll start with "usage: git submodule [...]" in both of those cases. This change does not alter the "list", "name", "clone", "config" and "create-branch" commands, those are internal-only (as an aside; their usage info should probably invoke BUG(...)). This only changes the user-facing commands. The "status", "deinit" and "update" commands are not included in this change, because their usage information already used "submodule" rather than "submodule--helper". I don't think it's currently possible to emit some of this usage information in practice, as git-submodule.sh will catch unknown options, and e.g. it doesn't seem to be possible to get "add" to emit its usage information from "submodule--helper". Though that change may be superfluous now, it's also harmless, and will allow us to eventually dispatch further into "git submodule--helper" from git-submodule.sh, while emitting the correct usage output. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-06-28 13:05:30 +03:00
N_("git submodule add [<options>] [--] <repository> [<path>]"),
NULL
};
argc = parse_options(argc, argv, prefix, options, usage, 0);
if (!is_writing_gitmodules_ok())
die(_("please make sure that the .gitmodules file is in the working tree"));
if (prefix && *prefix &&
add_data.reference_path && !is_absolute_path(add_data.reference_path))
add_data.reference_path = xstrfmt("%s%s", prefix, add_data.reference_path);
if (argc == 0 || argc > 2)
usage_with_options(usage, options);
add_data.repo = argv[0];
if (argc == 1)
add_data.sm_path = git_url_basename(add_data.repo, 0, 0);
else
add_data.sm_path = xstrdup(argv[1]);
if (prefix && *prefix && !is_absolute_path(add_data.sm_path))
add_data.sm_path = xstrfmt("%s%s", prefix, add_data.sm_path);
if (starts_with_dot_dot_slash(add_data.repo) ||
starts_with_dot_slash(add_data.repo)) {
if (prefix)
die(_("Relative path can only be used from the toplevel "
"of the working tree"));
/* dereference source url relative to parent's url */
to_free = resolve_relative_url(add_data.repo, NULL, 1);
add_data.realrepo = to_free;
} else if (is_dir_sep(add_data.repo[0]) || strchr(add_data.repo, ':')) {
add_data.realrepo = add_data.repo;
} else {
die(_("repo URL: '%s' must be absolute or begin with ./|../"),
add_data.repo);
}
/*
* normalize path:
* multiple //; leading ./; /./; /../;
*/
normalize_path_copy(add_data.sm_path, add_data.sm_path);
strip_dir_trailing_slashes(add_data.sm_path);
die_on_index_match(add_data.sm_path, force);
die_on_repo_without_commits(add_data.sm_path);
if (!force) {
int exit_code = -1;
struct strbuf sb = STRBUF_INIT;
struct child_process cp = CHILD_PROCESS_INIT;
cp.git_cmd = 1;
cp.no_stdout = 1;
strvec_pushl(&cp.args, "add", "--dry-run", "--ignore-missing",
"--no-warn-embedded-repo", add_data.sm_path, NULL);
if ((exit_code = pipe_command(&cp, NULL, 0, NULL, 0, &sb, 0))) {
strbuf_complete_line(&sb);
fputs(sb.buf, stderr);
free(add_data.sm_path);
return exit_code;
}
strbuf_release(&sb);
}
if(!add_data.sm_name)
add_data.sm_name = add_data.sm_path;
if (check_submodule_name(add_data.sm_name))
die(_("'%s' is not a valid submodule name"), add_data.sm_name);
add_data.prefix = prefix;
add_data.force = !!force;
add_data.quiet = !!quiet;
add_data.progress = !!progress;
add_data.dissociate = !!dissociate;
if (add_submodule(&add_data)) {
free(add_data.sm_path);
return 1;
}
configure_added_submodule(&add_data);
free(add_data.sm_path);
free(to_free);
return 0;
}
#define SUPPORT_SUPER_PREFIX (1<<0)
struct cmd_struct {
const char *cmd;
int (*fn)(int, const char **, const char *);
unsigned option;
};
static struct cmd_struct commands[] = {
{"list", module_list, 0},
{"name", module_name, 0},
{"clone", module_clone, 0},
{"add", module_add, SUPPORT_SUPER_PREFIX},
{"update", module_update, 0},
{"resolve-relative-url-test", resolve_relative_url_test, 0},
{"foreach", module_foreach, SUPPORT_SUPER_PREFIX},
{"init", module_init, SUPPORT_SUPER_PREFIX},
{"status", module_status, SUPPORT_SUPER_PREFIX},
{"sync", module_sync, SUPPORT_SUPER_PREFIX},
{"deinit", module_deinit, 0},
submodule: port submodule subcommand 'summary' from shell to C Convert submodule subcommand 'summary' to a builtin and call it via 'git-submodule.sh'. The shell version had to call $diff_cmd twice, once to find the modified modules cared by the user and then again, with that list of modules to do various operations for computing the summary of those modules. On the other hand, the C version does not need a second call to $diff_cmd since it reuses the module list from the first call to do the aforementioned tasks. In the C version, we use the combination of setting a child process' working directory to the submodule path and then calling 'prepare_submodule_repo_env()' which also sets the 'GIT_DIR' to '.git', so that we can be certain that those spawned processes will not access the superproject's ODB by mistake. A behavioural difference between the C and the shell version is that the shell version outputs two line feeds after the 'git log' output when run outside of the tests while the C version outputs one line feed in any case. The reason for this is that the shell version calls log with '--pretty=format:<fmt>' whose output is followed by two echo calls; 'format' does not have "terminator" semantics like its 'tformat' counterpart. So, the log output is terminated by a newline only when invoked by the user and not when invoked from the scripts. This results in the one & two line feed differences in the shell version. On the other hand, the C version calls log with '--pretty=<fmt>' which is equivalent to '--pretty:tformat:<fmt>' which is then followed by a 'printf("\n")'. Due to its "terminator" semantics the log output is always terminated by newline and hence one line feed in any case. Also, when we try to pass an option-like argument after a non-option argument, for instance: git submodule summary HEAD --foo-bar (or) git submodule summary HEAD --cached That argument would be treated like a path to the submodule for which the user is requesting a summary. So, the option ends up having no effect. Though, passing '--quiet' is an exception to this: git submodule summary HEAD --quiet While 'summary' doesn't support '--quiet', we don't get an output for the above command as '--quiet' is treated as a path which means we get an output only if a submodule whose path is '--quiet' exists. The error message in case of computing a summary for non-existent submodules in the C version is different from that of the shell version. Since the new error message is not marked for translation, change the 'test_i18ngrep' in t7421.4 to 'grep'. Mentored-by: Christian Couder <chriscool@tuxfamily.org> Mentored-by: Stefan Beller <stefanbeller@gmail.com> Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com> Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Prathamesh Chavan <pc44800@gmail.com> Signed-off-by: Shourya Shukla <shouryashukla.oo@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-08-12 22:44:04 +03:00
{"summary", module_summary, SUPPORT_SUPER_PREFIX},
{"push-check", push_check, 0},
{"absorbgitdirs", absorb_git_dirs, SUPPORT_SUPER_PREFIX},
{"is-active", is_active, 0},
submodule-config: verify submodule names as paths Submodule "names" come from the untrusted .gitmodules file, but we blindly append them to $GIT_DIR/modules to create our on-disk repo paths. This means you can do bad things by putting "../" into the name (among other things). Let's sanity-check these names to avoid building a path that can be exploited. There are two main decisions: 1. What should the allowed syntax be? It's tempting to reuse verify_path(), since submodule names typically come from in-repo paths. But there are two reasons not to: a. It's technically more strict than what we need, as we really care only about breaking out of the $GIT_DIR/modules/ hierarchy. E.g., having a submodule named "foo/.git" isn't actually dangerous, and it's possible that somebody has manually given such a funny name. b. Since we'll eventually use this checking logic in fsck to prevent downstream repositories, it should be consistent across platforms. Because verify_path() relies on is_dir_sep(), it wouldn't block "foo\..\bar" on a non-Windows machine. 2. Where should we enforce it? These days most of the .gitmodules reads go through submodule-config.c, so I've put it there in the reading step. That should cover all of the C code. We also construct the name for "git submodule add" inside the git-submodule.sh script. This is probably not a big deal for security since the name is coming from the user anyway, but it would be polite to remind them if the name they pick is invalid (and we need to expose the name-checker to the shell anyway for our test scripts). This patch issues a warning when reading .gitmodules and just ignores the related config entry completely. This will generally end up producing a sensible error, as it works the same as a .gitmodules file which is missing a submodule entry (so "submodule update" will barf, but "git clone --recurse-submodules" will print an error but not abort the clone. There is one minor oddity, which is that we print the warning once per malformed config key (since that's how the config subsystem gives us the entries). So in the new test, for example, the user would see three warnings. That's OK, since the intent is that this case should never come up outside of malicious repositories (and then it might even benefit the user to see the message multiple times). Credit for finding this vulnerability and the proof of concept from which the test script was adapted goes to Etienne Stalmans. Signed-off-by: Jeff King <peff@peff.net>
2018-04-30 10:25:25 +03:00
{"check-name", check_name, 0},
{"config", module_config, 0},
{"set-url", module_set_url, 0},
{"set-branch", module_set_branch, 0},
branch: add --recurse-submodules option for branch creation To improve the submodules UX, we would like to teach Git to handle branches in submodules. Start this process by teaching "git branch" the --recurse-submodules option so that "git branch --recurse-submodules topic" will create the `topic` branch in the superproject and its submodules. Although this commit does not introduce breaking changes, it does not work well with existing --recurse-submodules commands because "git branch --recurse-submodules" writes to the submodule ref store, but most commands only consider the superproject gitlink and ignore the submodule ref store. For example, "git checkout --recurse-submodules" will check out the commits in the superproject gitlinks (and put the submodules in detached HEAD) instead of checking out the submodule branches. Because of this, this commit introduces a new configuration value, `submodule.propagateBranches`. The plan is for Git commands to prioritize submodule ref store information over superproject gitlinks if this value is true. Because "git branch --recurse-submodules" writes to submodule ref stores, for the sake of clarity, it will not function unless this configuration value is set. This commit also includes changes that support working with submodules from a superproject commit because "branch --recurse-submodules" (and future commands) need to read .gitmodules and gitlinks from the superproject commit, but submodules are typically read from the filesystem's .gitmodules and the index's gitlinks. These changes are: * add a submodules_of_tree() helper that gives the relevant information of an in-tree submodule (e.g. path and oid) and initializes the repository * add is_tree_submodule_active() by adding a treeish_name parameter to is_submodule_active() * add the "submoduleNotUpdated" advice to advise users to update the submodules in their trees Incidentally, fix an incorrect usage string that combined the 'list' usage of git branch (-l) with the 'create' usage; this string has been incorrect since its inception, a8dfd5eac4 (Make builtin-branch.c use parse_options., 2007-10-07). Helped-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Glen Choo <chooglen@google.com> Reviewed-by: Jonathan Tan <jonathantanmy@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-01-29 03:04:45 +03:00
{"create-branch", module_create_branch, 0},
};
int cmd_submodule__helper(int argc, const char **argv, const char *prefix)
{
int i;
if (argc < 2 || !strcmp(argv[1], "-h"))
usage("git submodule--helper <command>");
for (i = 0; i < ARRAY_SIZE(commands); i++) {
if (!strcmp(argv[1], commands[i].cmd)) {
if (get_super_prefix() &&
!(commands[i].option & SUPPORT_SUPER_PREFIX))
die(_("%s doesn't support --super-prefix"),
commands[i].cmd);
return commands[i].fn(argc - 1, argv + 1, prefix);
}
}
die(_("'%s' is not a valid submodule--helper "
"subcommand"), argv[1]);
}