2015-06-13 22:37:27 +03:00
|
|
|
#include "builtin.h"
|
|
|
|
#include "cache.h"
|
|
|
|
#include "parse-options.h"
|
|
|
|
#include "refs.h"
|
|
|
|
#include "wildmatch.h"
|
2018-05-16 02:42:15 +03:00
|
|
|
#include "object-store.h"
|
2018-06-29 04:21:51 +03:00
|
|
|
#include "repository.h"
|
2015-06-13 22:37:27 +03:00
|
|
|
#include "commit.h"
|
|
|
|
#include "remote.h"
|
|
|
|
#include "color.h"
|
|
|
|
#include "tag.h"
|
|
|
|
#include "quote.h"
|
|
|
|
#include "ref-filter.h"
|
2015-07-07 19:06:12 +03:00
|
|
|
#include "revision.h"
|
2015-09-11 18:03:07 +03:00
|
|
|
#include "utf8.h"
|
2015-09-10 18:48:25 +03:00
|
|
|
#include "git-compat-util.h"
|
|
|
|
#include "version.h"
|
2016-11-19 03:58:15 +03:00
|
|
|
#include "trailer.h"
|
2017-01-10 11:49:38 +03:00
|
|
|
#include "wt-status.h"
|
2017-03-09 16:29:49 +03:00
|
|
|
#include "commit-slab.h"
|
2018-05-01 15:47:15 +03:00
|
|
|
#include "commit-graph.h"
|
2018-07-20 19:33:04 +03:00
|
|
|
#include "commit-reach.h"
|
2019-04-29 08:19:42 +03:00
|
|
|
#include "worktree.h"
|
|
|
|
#include "hashmap.h"
|
2020-07-28 23:23:39 +03:00
|
|
|
#include "strvec.h"
|
2015-06-13 22:37:27 +03:00
|
|
|
|
2017-01-10 11:49:50 +03:00
|
|
|
static struct ref_msg {
|
|
|
|
const char *gone;
|
|
|
|
const char *ahead;
|
|
|
|
const char *behind;
|
|
|
|
const char *ahead_behind;
|
|
|
|
} msgs = {
|
|
|
|
/* Untranslated plumbing messages: */
|
|
|
|
"gone",
|
|
|
|
"ahead %d",
|
|
|
|
"behind %d",
|
|
|
|
"ahead %d, behind %d"
|
|
|
|
};
|
|
|
|
|
|
|
|
void setup_ref_filter_porcelain_msg(void)
|
|
|
|
{
|
|
|
|
msgs.gone = _("gone");
|
|
|
|
msgs.ahead = _("ahead %d");
|
|
|
|
msgs.behind = _("behind %d");
|
|
|
|
msgs.ahead_behind = _("ahead %d, behind %d");
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
typedef enum { FIELD_STR, FIELD_ULONG, FIELD_TIME } cmp_type;
|
2017-01-10 11:49:36 +03:00
|
|
|
typedef enum { COMPARE_EQUAL, COMPARE_UNEQUAL, COMPARE_NONE } cmp_status;
|
2018-07-17 11:22:57 +03:00
|
|
|
typedef enum { SOURCE_NONE = 0, SOURCE_OBJ, SOURCE_OTHER } info_source;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
2016-02-17 21:06:15 +03:00
|
|
|
struct align {
|
|
|
|
align_type position;
|
|
|
|
unsigned int width;
|
|
|
|
};
|
|
|
|
|
2017-01-10 11:49:34 +03:00
|
|
|
struct if_then_else {
|
2017-01-10 11:49:36 +03:00
|
|
|
cmp_status cmp_status;
|
|
|
|
const char *str;
|
2017-01-10 11:49:34 +03:00
|
|
|
unsigned int then_atom_seen : 1,
|
|
|
|
else_atom_seen : 1,
|
|
|
|
condition_satisfied : 1;
|
|
|
|
};
|
|
|
|
|
2017-01-10 11:49:43 +03:00
|
|
|
struct refname_atom {
|
2017-01-10 11:49:49 +03:00
|
|
|
enum { R_NORMAL, R_SHORT, R_LSTRIP, R_RSTRIP } option;
|
|
|
|
int lstrip, rstrip;
|
2017-01-10 11:49:43 +03:00
|
|
|
};
|
|
|
|
|
2018-07-17 11:22:57 +03:00
|
|
|
static struct expand_data {
|
|
|
|
struct object_id oid;
|
|
|
|
enum object_type type;
|
|
|
|
unsigned long size;
|
|
|
|
off_t disk_size;
|
|
|
|
struct object_id delta_base_oid;
|
|
|
|
void *content;
|
|
|
|
|
|
|
|
struct object_info info;
|
|
|
|
} oi, oi_deref;
|
|
|
|
|
2019-04-29 08:19:42 +03:00
|
|
|
struct ref_to_worktree_entry {
|
2019-10-07 02:30:43 +03:00
|
|
|
struct hashmap_entry ent;
|
2019-04-29 08:19:42 +03:00
|
|
|
struct worktree *wt; /* key is wt->head_ref */
|
|
|
|
};
|
|
|
|
|
|
|
|
static int ref_to_worktree_map_cmpfnc(const void *unused_lookupdata,
|
2019-10-07 02:30:37 +03:00
|
|
|
const struct hashmap_entry *eptr,
|
|
|
|
const struct hashmap_entry *kptr,
|
2019-04-29 08:19:42 +03:00
|
|
|
const void *keydata_aka_refname)
|
|
|
|
{
|
2019-10-07 02:30:37 +03:00
|
|
|
const struct ref_to_worktree_entry *e, *k;
|
|
|
|
|
|
|
|
e = container_of(eptr, const struct ref_to_worktree_entry, ent);
|
|
|
|
k = container_of(kptr, const struct ref_to_worktree_entry, ent);
|
|
|
|
|
2019-04-29 08:19:42 +03:00
|
|
|
return strcmp(e->wt->head_ref,
|
|
|
|
keydata_aka_refname ? keydata_aka_refname : k->wt->head_ref);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct ref_to_worktree_map {
|
|
|
|
struct hashmap map;
|
|
|
|
struct worktree **worktrees;
|
|
|
|
} ref_to_worktree_map;
|
|
|
|
|
2016-02-17 21:06:10 +03:00
|
|
|
/*
|
|
|
|
* An atom is a valid field atom listed below, possibly prefixed with
|
|
|
|
* a "*" to denote deref_tag().
|
|
|
|
*
|
|
|
|
* We parse given format string and sort specifiers, and make a list
|
|
|
|
* of properties that we need to extract out of objects. ref_array_item
|
|
|
|
* structure will hold an array of values extracted that can be
|
|
|
|
* indexed with the "atom number", which is an index into this
|
|
|
|
* array.
|
|
|
|
*/
|
2016-02-17 21:06:11 +03:00
|
|
|
static struct used_atom {
|
|
|
|
const char *name;
|
|
|
|
cmp_type type;
|
2018-07-17 11:22:57 +03:00
|
|
|
info_source source;
|
2016-02-17 21:06:13 +03:00
|
|
|
union {
|
|
|
|
char color[COLOR_MAXLEN];
|
2016-02-17 21:06:15 +03:00
|
|
|
struct align align;
|
2017-01-10 11:49:41 +03:00
|
|
|
struct {
|
2017-10-05 15:19:09 +03:00
|
|
|
enum {
|
2017-11-07 19:31:08 +03:00
|
|
|
RR_REF, RR_TRACK, RR_TRACKSHORT, RR_REMOTE_NAME, RR_REMOTE_REF
|
2017-10-05 15:19:09 +03:00
|
|
|
} option;
|
2017-01-10 11:49:45 +03:00
|
|
|
struct refname_atom refname;
|
2017-10-05 15:19:09 +03:00
|
|
|
unsigned int nobracket : 1, push : 1, push_remote : 1;
|
2017-01-10 11:49:41 +03:00
|
|
|
} remote_ref;
|
2016-02-17 21:06:18 +03:00
|
|
|
struct {
|
2020-08-22 00:41:50 +03:00
|
|
|
enum { C_BARE, C_BODY, C_BODY_DEP, C_LENGTH, C_LINES,
|
|
|
|
C_SIG, C_SUB, C_SUB_SANITIZE, C_TRAILERS } option;
|
2017-10-02 08:25:23 +03:00
|
|
|
struct process_trailer_options trailer_opts;
|
2016-02-17 21:06:18 +03:00
|
|
|
unsigned int nlines;
|
|
|
|
} contents;
|
2017-01-10 11:49:36 +03:00
|
|
|
struct {
|
|
|
|
cmp_status cmp_status;
|
|
|
|
const char *str;
|
|
|
|
} if_then_else;
|
2017-01-10 11:49:37 +03:00
|
|
|
struct {
|
|
|
|
enum { O_FULL, O_LENGTH, O_SHORT } option;
|
|
|
|
unsigned int length;
|
2020-08-22 00:41:46 +03:00
|
|
|
} oid;
|
2020-08-22 00:41:43 +03:00
|
|
|
struct email_option {
|
|
|
|
enum { EO_RAW, EO_TRIM, EO_LOCALPART } option;
|
|
|
|
} email_option;
|
2017-01-10 11:49:43 +03:00
|
|
|
struct refname_atom refname;
|
2017-05-19 09:12:12 +03:00
|
|
|
char *head;
|
2016-02-17 21:06:13 +03:00
|
|
|
} u;
|
2016-02-17 21:06:11 +03:00
|
|
|
} *used_atom;
|
2016-02-17 21:06:10 +03:00
|
|
|
static int used_atom_cnt, need_tagged, need_symref;
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
/*
|
|
|
|
* Expand string, append it to strbuf *sb, then return error code ret.
|
|
|
|
* Allow to save few lines of code.
|
|
|
|
*/
|
|
|
|
static int strbuf_addf_ret(struct strbuf *sb, int ret, const char *fmt, ...)
|
|
|
|
{
|
|
|
|
va_list ap;
|
|
|
|
va_start(ap, fmt);
|
|
|
|
strbuf_vaddf(sb, fmt, ap);
|
|
|
|
va_end(ap);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int color_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *color_value, struct strbuf *err)
|
2016-02-17 21:06:13 +03:00
|
|
|
{
|
|
|
|
if (!color_value)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("expected format: %%(color:<color>)"));
|
2016-02-17 21:06:13 +03:00
|
|
|
if (color_parse(color_value, atom->u.color) < 0)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized color: %%(color:%s)"),
|
|
|
|
color_value);
|
ref-filter: consult want_color() before emitting colors
When color placeholders like %(color:red) are used in a
ref-filter format, we unconditionally output the colors,
even if the user has asked us for no colors. This usually
isn't a problem when the user is constructing a --format on
the command line, but it means we may do the wrong thing
when the format is fed from a script or alias. For example:
$ git config alias.b 'branch --format=%(color:green)%(refname)'
$ git b --no-color
should probably omit the green color. Likewise, running:
$ git b >branches
should probably also omit the color, just as we would for
all baked-in coloring (and as we recently started to do for
user-specified colors in --pretty formats).
This commit makes both of those cases work by teaching
the ref-filter code to consult want_color() before
outputting any color. The color flag in ref_format defaults
to "-1", which means we'll consult color.ui, which in turn
defaults to the usual isatty() check on stdout. However,
callers like git-branch which support their own color config
(and command-line options) can override that.
The new tests independently cover all three of the callers
of ref-filter (for-each-ref, tag, and branch). Even though
these seem redundant, it confirms that we've correctly
plumbed through all of the necessary config to make colors
work by default.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-07-13 18:09:32 +03:00
|
|
|
/*
|
|
|
|
* We check this after we've parsed the color, which lets us complain
|
|
|
|
* about syntactically bogus color names even if they won't be used.
|
|
|
|
*/
|
|
|
|
if (!want_color(format->use_color))
|
|
|
|
color_parse("", atom->u.color);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:13 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int refname_atom_parser_internal(struct refname_atom *atom, const char *arg,
|
|
|
|
const char *name, struct strbuf *err)
|
2016-02-17 21:06:17 +03:00
|
|
|
{
|
|
|
|
if (!arg)
|
2017-01-10 11:49:43 +03:00
|
|
|
atom->option = R_NORMAL;
|
2016-02-17 21:06:17 +03:00
|
|
|
else if (!strcmp(arg, "short"))
|
2017-01-10 11:49:43 +03:00
|
|
|
atom->option = R_SHORT;
|
2017-02-07 22:50:34 +03:00
|
|
|
else if (skip_prefix(arg, "lstrip=", &arg) ||
|
|
|
|
skip_prefix(arg, "strip=", &arg)) {
|
2017-01-10 11:49:46 +03:00
|
|
|
atom->option = R_LSTRIP;
|
2017-01-10 11:49:48 +03:00
|
|
|
if (strtol_i(arg, 10, &atom->lstrip))
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("Integer value expected refname:lstrip=%s"), arg);
|
2017-01-10 11:49:49 +03:00
|
|
|
} else if (skip_prefix(arg, "rstrip=", &arg)) {
|
|
|
|
atom->option = R_RSTRIP;
|
|
|
|
if (strtol_i(arg, 10, &atom->rstrip))
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("Integer value expected refname:rstrip=%s"), arg);
|
2017-01-10 11:49:43 +03:00
|
|
|
} else
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized %%(%s) argument: %s"), name, arg);
|
|
|
|
return 0;
|
2017-01-10 11:49:43 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int remote_ref_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:17 +03:00
|
|
|
{
|
2017-01-10 11:49:41 +03:00
|
|
|
struct string_list params = STRING_LIST_INIT_DUP;
|
|
|
|
int i;
|
|
|
|
|
2017-10-05 15:19:09 +03:00
|
|
|
if (!strcmp(atom->name, "push") || starts_with(atom->name, "push:"))
|
|
|
|
atom->u.remote_ref.push = 1;
|
|
|
|
|
2017-01-10 11:49:41 +03:00
|
|
|
if (!arg) {
|
2017-01-10 11:49:45 +03:00
|
|
|
atom->u.remote_ref.option = RR_REF;
|
2018-03-29 15:49:45 +03:00
|
|
|
return refname_atom_parser_internal(&atom->u.remote_ref.refname,
|
|
|
|
arg, atom->name, err);
|
2017-01-10 11:49:41 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
atom->u.remote_ref.nobracket = 0;
|
|
|
|
string_list_split(¶ms, arg, ',', -1);
|
|
|
|
|
|
|
|
for (i = 0; i < params.nr; i++) {
|
|
|
|
const char *s = params.items[i].string;
|
|
|
|
|
2017-01-10 11:49:45 +03:00
|
|
|
if (!strcmp(s, "track"))
|
2017-01-10 11:49:41 +03:00
|
|
|
atom->u.remote_ref.option = RR_TRACK;
|
|
|
|
else if (!strcmp(s, "trackshort"))
|
|
|
|
atom->u.remote_ref.option = RR_TRACKSHORT;
|
|
|
|
else if (!strcmp(s, "nobracket"))
|
|
|
|
atom->u.remote_ref.nobracket = 1;
|
2017-10-05 15:19:09 +03:00
|
|
|
else if (!strcmp(s, "remotename")) {
|
|
|
|
atom->u.remote_ref.option = RR_REMOTE_NAME;
|
|
|
|
atom->u.remote_ref.push_remote = 1;
|
2017-11-07 19:31:08 +03:00
|
|
|
} else if (!strcmp(s, "remoteref")) {
|
|
|
|
atom->u.remote_ref.option = RR_REMOTE_REF;
|
|
|
|
atom->u.remote_ref.push_remote = 1;
|
2017-10-05 15:19:09 +03:00
|
|
|
} else {
|
2017-01-10 11:49:45 +03:00
|
|
|
atom->u.remote_ref.option = RR_REF;
|
2018-03-29 15:49:45 +03:00
|
|
|
if (refname_atom_parser_internal(&atom->u.remote_ref.refname,
|
|
|
|
arg, atom->name, err)) {
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return -1;
|
|
|
|
}
|
2017-01-10 11:49:45 +03:00
|
|
|
}
|
2017-01-10 11:49:41 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
string_list_clear(¶ms, 0);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:17 +03:00
|
|
|
}
|
|
|
|
|
2018-07-17 11:22:57 +03:00
|
|
|
static int objecttype_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
|
|
|
{
|
|
|
|
if (arg)
|
|
|
|
return strbuf_addf_ret(err, -1, _("%%(objecttype) does not take arguments"));
|
|
|
|
if (*atom->name == '*')
|
|
|
|
oi_deref.info.typep = &oi_deref.type;
|
|
|
|
else
|
|
|
|
oi.info.typep = &oi.type;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int objectsize_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
|
|
|
{
|
2018-12-24 16:24:30 +03:00
|
|
|
if (!arg) {
|
|
|
|
if (*atom->name == '*')
|
|
|
|
oi_deref.info.sizep = &oi_deref.size;
|
|
|
|
else
|
|
|
|
oi.info.sizep = &oi.size;
|
|
|
|
} else if (!strcmp(arg, "disk")) {
|
|
|
|
if (*atom->name == '*')
|
|
|
|
oi_deref.info.disk_sizep = &oi_deref.disk_size;
|
|
|
|
else
|
|
|
|
oi.info.disk_sizep = &oi.disk_size;
|
|
|
|
} else
|
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized %%(objectsize) argument: %s"), arg);
|
2018-07-17 11:22:57 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-12-24 16:24:30 +03:00
|
|
|
static int deltabase_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2018-07-17 11:22:57 +03:00
|
|
|
{
|
|
|
|
if (arg)
|
2018-12-24 16:24:30 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("%%(deltabase) does not take arguments"));
|
2018-07-17 11:22:57 +03:00
|
|
|
if (*atom->name == '*')
|
2020-02-24 07:36:56 +03:00
|
|
|
oi_deref.info.delta_base_oid = &oi_deref.delta_base_oid;
|
2018-07-17 11:22:57 +03:00
|
|
|
else
|
2020-02-24 07:36:56 +03:00
|
|
|
oi.info.delta_base_oid = &oi.delta_base_oid;
|
2018-07-17 11:22:57 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int body_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:18 +03:00
|
|
|
{
|
|
|
|
if (arg)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("%%(body) does not take arguments"));
|
2016-02-17 21:06:18 +03:00
|
|
|
atom->u.contents.option = C_BODY_DEP;
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:18 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int subject_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:18 +03:00
|
|
|
{
|
2020-08-22 00:41:50 +03:00
|
|
|
if (!arg)
|
|
|
|
atom->u.contents.option = C_SUB;
|
|
|
|
else if (!strcmp(arg, "sanitize"))
|
|
|
|
atom->u.contents.option = C_SUB_SANITIZE;
|
|
|
|
else
|
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized %%(subject) argument: %s"), arg);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:18 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int trailers_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-11-19 03:58:15 +03:00
|
|
|
{
|
2017-10-02 08:25:23 +03:00
|
|
|
struct string_list params = STRING_LIST_INIT_DUP;
|
|
|
|
int i;
|
|
|
|
|
2018-08-23 03:50:17 +03:00
|
|
|
atom->u.contents.trailer_opts.no_divider = 1;
|
|
|
|
|
2017-10-02 08:25:23 +03:00
|
|
|
if (arg) {
|
|
|
|
string_list_split(¶ms, arg, ',', -1);
|
|
|
|
for (i = 0; i < params.nr; i++) {
|
|
|
|
const char *s = params.items[i].string;
|
|
|
|
if (!strcmp(s, "unfold"))
|
|
|
|
atom->u.contents.trailer_opts.unfold = 1;
|
|
|
|
else if (!strcmp(s, "only"))
|
|
|
|
atom->u.contents.trailer_opts.only_trailers = 1;
|
2018-03-29 15:49:45 +03:00
|
|
|
else {
|
|
|
|
strbuf_addf(err, _("unknown %%(trailers) argument: %s"), s);
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return -1;
|
|
|
|
}
|
2017-10-02 08:25:23 +03:00
|
|
|
}
|
|
|
|
}
|
2016-11-19 03:58:15 +03:00
|
|
|
atom->u.contents.option = C_TRAILERS;
|
2017-10-02 08:25:23 +03:00
|
|
|
string_list_clear(¶ms, 0);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-11-19 03:58:15 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int contents_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:18 +03:00
|
|
|
{
|
|
|
|
if (!arg)
|
|
|
|
atom->u.contents.option = C_BARE;
|
|
|
|
else if (!strcmp(arg, "body"))
|
|
|
|
atom->u.contents.option = C_BODY;
|
2020-07-16 15:19:40 +03:00
|
|
|
else if (!strcmp(arg, "size"))
|
|
|
|
atom->u.contents.option = C_LENGTH;
|
2016-02-17 21:06:18 +03:00
|
|
|
else if (!strcmp(arg, "signature"))
|
|
|
|
atom->u.contents.option = C_SIG;
|
|
|
|
else if (!strcmp(arg, "subject"))
|
|
|
|
atom->u.contents.option = C_SUB;
|
2020-08-22 00:06:14 +03:00
|
|
|
else if (!strcmp(arg, "trailers")) {
|
|
|
|
if (trailers_atom_parser(format, atom, NULL, err))
|
|
|
|
return -1;
|
|
|
|
} else if (skip_prefix(arg, "trailers:", &arg)) {
|
|
|
|
if (trailers_atom_parser(format, atom, arg, err))
|
2018-03-29 15:49:45 +03:00
|
|
|
return -1;
|
2017-10-02 08:25:24 +03:00
|
|
|
} else if (skip_prefix(arg, "lines=", &arg)) {
|
2016-02-17 21:06:18 +03:00
|
|
|
atom->u.contents.option = C_LINES;
|
|
|
|
if (strtoul_ui(arg, 10, &atom->u.contents.nlines))
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("positive value expected contents:lines=%s"), arg);
|
2016-02-17 21:06:18 +03:00
|
|
|
} else
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized %%(contents) argument: %s"), arg);
|
|
|
|
return 0;
|
2016-02-17 21:06:18 +03:00
|
|
|
}
|
|
|
|
|
2020-08-22 00:41:46 +03:00
|
|
|
static int oid_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:19 +03:00
|
|
|
{
|
|
|
|
if (!arg)
|
2020-08-22 00:41:46 +03:00
|
|
|
atom->u.oid.option = O_FULL;
|
2016-02-17 21:06:19 +03:00
|
|
|
else if (!strcmp(arg, "short"))
|
2020-08-22 00:41:46 +03:00
|
|
|
atom->u.oid.option = O_SHORT;
|
2017-01-10 11:49:37 +03:00
|
|
|
else if (skip_prefix(arg, "short=", &arg)) {
|
2020-08-22 00:41:46 +03:00
|
|
|
atom->u.oid.option = O_LENGTH;
|
|
|
|
if (strtoul_ui(arg, 10, &atom->u.oid.length) ||
|
|
|
|
atom->u.oid.length == 0)
|
2020-08-22 00:41:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("positive value expected '%s' in %%(%s)"), arg, atom->name);
|
2020-08-22 00:41:46 +03:00
|
|
|
if (atom->u.oid.length < MINIMUM_ABBREV)
|
|
|
|
atom->u.oid.length = MINIMUM_ABBREV;
|
2017-01-10 11:49:37 +03:00
|
|
|
} else
|
2020-08-22 00:41:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized argument '%s' in %%(%s)"), arg, atom->name);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:19 +03:00
|
|
|
}
|
|
|
|
|
2020-08-22 00:41:43 +03:00
|
|
|
static int person_email_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
|
|
|
{
|
|
|
|
if (!arg)
|
|
|
|
atom->u.email_option.option = EO_RAW;
|
|
|
|
else if (!strcmp(arg, "trim"))
|
|
|
|
atom->u.email_option.option = EO_TRIM;
|
|
|
|
else if (!strcmp(arg, "localpart"))
|
|
|
|
atom->u.email_option.option = EO_LOCALPART;
|
|
|
|
else
|
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized email option: %s"), arg);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:19 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int refname_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2017-01-10 11:49:44 +03:00
|
|
|
{
|
2018-03-29 15:49:45 +03:00
|
|
|
return refname_atom_parser_internal(&atom->u.refname, arg, atom->name, err);
|
2017-01-10 11:49:44 +03:00
|
|
|
}
|
|
|
|
|
2016-02-17 21:06:14 +03:00
|
|
|
static align_type parse_align_position(const char *s)
|
|
|
|
{
|
|
|
|
if (!strcmp(s, "right"))
|
|
|
|
return ALIGN_RIGHT;
|
|
|
|
else if (!strcmp(s, "middle"))
|
|
|
|
return ALIGN_MIDDLE;
|
|
|
|
else if (!strcmp(s, "left"))
|
|
|
|
return ALIGN_LEFT;
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int align_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2016-02-17 21:06:15 +03:00
|
|
|
{
|
|
|
|
struct align *align = &atom->u.align;
|
|
|
|
struct string_list params = STRING_LIST_INIT_DUP;
|
|
|
|
int i;
|
|
|
|
unsigned int width = ~0U;
|
|
|
|
|
|
|
|
if (!arg)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("expected format: %%(align:<width>,<position>)"));
|
2016-02-17 21:06:15 +03:00
|
|
|
|
|
|
|
align->position = ALIGN_LEFT;
|
|
|
|
|
|
|
|
string_list_split(¶ms, arg, ',', -1);
|
|
|
|
for (i = 0; i < params.nr; i++) {
|
|
|
|
const char *s = params.items[i].string;
|
|
|
|
int position;
|
|
|
|
|
2016-02-17 21:06:16 +03:00
|
|
|
if (skip_prefix(s, "position=", &s)) {
|
|
|
|
position = parse_align_position(s);
|
2018-03-29 15:49:45 +03:00
|
|
|
if (position < 0) {
|
|
|
|
strbuf_addf(err, _("unrecognized position:%s"), s);
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return -1;
|
|
|
|
}
|
2016-02-17 21:06:16 +03:00
|
|
|
align->position = position;
|
|
|
|
} else if (skip_prefix(s, "width=", &s)) {
|
2018-03-29 15:49:45 +03:00
|
|
|
if (strtoul_ui(s, 10, &width)) {
|
|
|
|
strbuf_addf(err, _("unrecognized width:%s"), s);
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return -1;
|
|
|
|
}
|
2016-02-17 21:06:16 +03:00
|
|
|
} else if (!strtoul_ui(s, 10, &width))
|
2016-02-17 21:06:15 +03:00
|
|
|
;
|
|
|
|
else if ((position = parse_align_position(s)) >= 0)
|
|
|
|
align->position = position;
|
2018-03-29 15:49:45 +03:00
|
|
|
else {
|
|
|
|
strbuf_addf(err, _("unrecognized %%(align) argument: %s"), s);
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return -1;
|
|
|
|
}
|
2016-02-17 21:06:15 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
if (width == ~0U) {
|
|
|
|
string_list_clear(¶ms, 0);
|
|
|
|
return strbuf_addf_ret(err, -1, _("positive width expected with the %%(align) atom"));
|
|
|
|
}
|
2016-02-17 21:06:15 +03:00
|
|
|
align->width = width;
|
|
|
|
string_list_clear(¶ms, 0);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2016-02-17 21:06:15 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int if_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err)
|
2017-01-10 11:49:36 +03:00
|
|
|
{
|
|
|
|
if (!arg) {
|
|
|
|
atom->u.if_then_else.cmp_status = COMPARE_NONE;
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2017-01-10 11:49:36 +03:00
|
|
|
} else if (skip_prefix(arg, "equals=", &atom->u.if_then_else.str)) {
|
|
|
|
atom->u.if_then_else.cmp_status = COMPARE_EQUAL;
|
|
|
|
} else if (skip_prefix(arg, "notequals=", &atom->u.if_then_else.str)) {
|
|
|
|
atom->u.if_then_else.cmp_status = COMPARE_UNEQUAL;
|
2018-03-29 15:49:45 +03:00
|
|
|
} else
|
|
|
|
return strbuf_addf_ret(err, -1, _("unrecognized %%(if) argument: %s"), arg);
|
|
|
|
return 0;
|
2017-01-10 11:49:36 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int head_atom_parser(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *unused_err)
|
2017-05-19 09:12:12 +03:00
|
|
|
{
|
2017-10-01 10:29:03 +03:00
|
|
|
atom->u.head = resolve_refdup("HEAD", RESOLVE_REF_READING, NULL, NULL);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2017-05-19 09:12:12 +03:00
|
|
|
}
|
2017-01-10 11:49:36 +03:00
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
static struct {
|
|
|
|
const char *name;
|
2018-07-17 11:22:57 +03:00
|
|
|
info_source source;
|
2015-06-13 22:37:27 +03:00
|
|
|
cmp_type cmp_type;
|
2018-03-29 15:49:45 +03:00
|
|
|
int (*parser)(const struct ref_format *format, struct used_atom *atom,
|
|
|
|
const char *arg, struct strbuf *err);
|
2015-06-13 22:37:27 +03:00
|
|
|
} valid_atom[] = {
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "refname", SOURCE_NONE, FIELD_STR, refname_atom_parser },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "objecttype", SOURCE_OTHER, FIELD_STR, objecttype_atom_parser },
|
|
|
|
{ "objectsize", SOURCE_OTHER, FIELD_ULONG, objectsize_atom_parser },
|
2020-08-22 00:41:46 +03:00
|
|
|
{ "objectname", SOURCE_OTHER, FIELD_STR, oid_atom_parser },
|
2018-12-24 16:24:30 +03:00
|
|
|
{ "deltabase", SOURCE_OTHER, FIELD_STR, deltabase_atom_parser },
|
2020-08-22 00:41:47 +03:00
|
|
|
{ "tree", SOURCE_OBJ, FIELD_STR, oid_atom_parser },
|
2020-08-22 00:41:48 +03:00
|
|
|
{ "parent", SOURCE_OBJ, FIELD_STR, oid_atom_parser },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "numparent", SOURCE_OBJ, FIELD_ULONG },
|
|
|
|
{ "object", SOURCE_OBJ },
|
|
|
|
{ "type", SOURCE_OBJ },
|
|
|
|
{ "tag", SOURCE_OBJ },
|
|
|
|
{ "author", SOURCE_OBJ },
|
|
|
|
{ "authorname", SOURCE_OBJ },
|
2020-08-22 00:41:43 +03:00
|
|
|
{ "authoremail", SOURCE_OBJ, FIELD_STR, person_email_atom_parser },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "authordate", SOURCE_OBJ, FIELD_TIME },
|
|
|
|
{ "committer", SOURCE_OBJ },
|
|
|
|
{ "committername", SOURCE_OBJ },
|
2020-08-22 00:41:43 +03:00
|
|
|
{ "committeremail", SOURCE_OBJ, FIELD_STR, person_email_atom_parser },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "committerdate", SOURCE_OBJ, FIELD_TIME },
|
|
|
|
{ "tagger", SOURCE_OBJ },
|
|
|
|
{ "taggername", SOURCE_OBJ },
|
2020-08-22 00:41:43 +03:00
|
|
|
{ "taggeremail", SOURCE_OBJ, FIELD_STR, person_email_atom_parser },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "taggerdate", SOURCE_OBJ, FIELD_TIME },
|
|
|
|
{ "creator", SOURCE_OBJ },
|
|
|
|
{ "creatordate", SOURCE_OBJ, FIELD_TIME },
|
|
|
|
{ "subject", SOURCE_OBJ, FIELD_STR, subject_atom_parser },
|
|
|
|
{ "body", SOURCE_OBJ, FIELD_STR, body_atom_parser },
|
|
|
|
{ "trailers", SOURCE_OBJ, FIELD_STR, trailers_atom_parser },
|
|
|
|
{ "contents", SOURCE_OBJ, FIELD_STR, contents_atom_parser },
|
|
|
|
{ "upstream", SOURCE_NONE, FIELD_STR, remote_ref_atom_parser },
|
|
|
|
{ "push", SOURCE_NONE, FIELD_STR, remote_ref_atom_parser },
|
|
|
|
{ "symref", SOURCE_NONE, FIELD_STR, refname_atom_parser },
|
|
|
|
{ "flag", SOURCE_NONE },
|
|
|
|
{ "HEAD", SOURCE_NONE, FIELD_STR, head_atom_parser },
|
|
|
|
{ "color", SOURCE_NONE, FIELD_STR, color_atom_parser },
|
2019-04-29 08:19:42 +03:00
|
|
|
{ "worktreepath", SOURCE_NONE },
|
2018-07-17 11:22:57 +03:00
|
|
|
{ "align", SOURCE_NONE, FIELD_STR, align_atom_parser },
|
|
|
|
{ "end", SOURCE_NONE },
|
|
|
|
{ "if", SOURCE_NONE, FIELD_STR, if_atom_parser },
|
|
|
|
{ "then", SOURCE_NONE },
|
|
|
|
{ "else", SOURCE_NONE },
|
2019-02-16 14:24:41 +03:00
|
|
|
/*
|
|
|
|
* Please update $__git_ref_fieldlist in git-completion.bash
|
|
|
|
* when you add new atoms
|
|
|
|
*/
|
2015-06-13 22:37:27 +03:00
|
|
|
};
|
|
|
|
|
2015-09-10 18:48:18 +03:00
|
|
|
#define REF_FORMATTING_STATE_INIT { 0, NULL }
|
|
|
|
|
|
|
|
struct ref_formatting_stack {
|
|
|
|
struct ref_formatting_stack *prev;
|
|
|
|
struct strbuf output;
|
2017-01-10 11:49:34 +03:00
|
|
|
void (*at_end)(struct ref_formatting_stack **stack);
|
2015-09-11 18:03:07 +03:00
|
|
|
void *at_end_data;
|
2015-09-10 18:48:18 +03:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ref_formatting_state {
|
|
|
|
int quote_style;
|
|
|
|
struct ref_formatting_stack *stack;
|
|
|
|
};
|
|
|
|
|
2015-08-22 06:39:37 +03:00
|
|
|
struct atom_value {
|
|
|
|
const char *s;
|
2018-03-29 15:49:45 +03:00
|
|
|
int (*handler)(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *err);
|
2017-04-20 23:52:09 +03:00
|
|
|
uintmax_t value; /* used for sorting when not FIELD_STR */
|
2017-01-10 11:49:35 +03:00
|
|
|
struct used_atom *atom;
|
2015-08-22 06:39:37 +03:00
|
|
|
};
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* Used to parse format string and sort specifiers
|
|
|
|
*/
|
2017-07-13 18:06:40 +03:00
|
|
|
static int parse_ref_filter_atom(const struct ref_format *format,
|
2018-03-29 15:49:45 +03:00
|
|
|
const char *atom, const char *ep,
|
|
|
|
struct strbuf *err)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
const char *sp;
|
2016-02-17 21:06:12 +03:00
|
|
|
const char *arg;
|
2016-10-02 19:35:11 +03:00
|
|
|
int i, at, atom_len;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
sp = atom;
|
|
|
|
if (*sp == '*' && sp < ep)
|
|
|
|
sp++; /* deref */
|
|
|
|
if (ep <= sp)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("malformed field name: %.*s"),
|
|
|
|
(int)(ep-atom), atom);
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
/* Do we have the atom already used elsewhere? */
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
int len = strlen(used_atom[i].name);
|
|
|
|
if (len == ep - atom && !memcmp(used_atom[i].name, atom, len))
|
2015-06-13 22:37:27 +03:00
|
|
|
return i;
|
|
|
|
}
|
|
|
|
|
2016-10-02 19:35:11 +03:00
|
|
|
/*
|
|
|
|
* If the atom name has a colon, strip it and everything after
|
|
|
|
* it off - it specifies the format for this entry, and
|
|
|
|
* shouldn't be used for checking against the valid_atom
|
|
|
|
* table.
|
|
|
|
*/
|
|
|
|
arg = memchr(sp, ':', ep - sp);
|
|
|
|
atom_len = (arg ? arg : ep) - sp;
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/* Is the atom a valid one? */
|
|
|
|
for (i = 0; i < ARRAY_SIZE(valid_atom); i++) {
|
|
|
|
int len = strlen(valid_atom[i].name);
|
2016-10-02 19:35:11 +03:00
|
|
|
if (len == atom_len && !memcmp(valid_atom[i].name, sp, len))
|
2015-06-13 22:37:27 +03:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ARRAY_SIZE(valid_atom) <= i)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("unknown field name: %.*s"),
|
|
|
|
(int)(ep-atom), atom);
|
2018-11-14 15:27:25 +03:00
|
|
|
if (valid_atom[i].source != SOURCE_NONE && !have_git_dir())
|
|
|
|
return strbuf_addf_ret(err, -1,
|
|
|
|
_("not a git repository, but the field '%.*s' requires access to object data"),
|
|
|
|
(int)(ep-atom), atom);
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
/* Add it in, including the deref prefix */
|
|
|
|
at = used_atom_cnt;
|
|
|
|
used_atom_cnt++;
|
|
|
|
REALLOC_ARRAY(used_atom, used_atom_cnt);
|
2016-02-17 21:06:11 +03:00
|
|
|
used_atom[at].name = xmemdupz(atom, ep - atom);
|
|
|
|
used_atom[at].type = valid_atom[i].cmp_type;
|
2018-07-17 11:22:57 +03:00
|
|
|
used_atom[at].source = valid_atom[i].source;
|
2018-07-17 11:22:57 +03:00
|
|
|
if (used_atom[at].source == SOURCE_OBJ) {
|
|
|
|
if (*atom == '*')
|
|
|
|
oi_deref.info.contentp = &oi_deref.content;
|
|
|
|
else
|
|
|
|
oi.info.contentp = &oi.content;
|
|
|
|
}
|
2017-10-02 19:10:34 +03:00
|
|
|
if (arg) {
|
2016-02-17 21:06:12 +03:00
|
|
|
arg = used_atom[at].name + (arg - atom) + 1;
|
2017-10-02 19:10:34 +03:00
|
|
|
if (!*arg) {
|
|
|
|
/*
|
|
|
|
* Treat empty sub-arguments list as NULL (i.e.,
|
|
|
|
* "%(atom:)" is equivalent to "%(atom)").
|
|
|
|
*/
|
|
|
|
arg = NULL;
|
|
|
|
}
|
|
|
|
}
|
2016-02-17 21:06:13 +03:00
|
|
|
memset(&used_atom[at].u, 0, sizeof(used_atom[at].u));
|
2018-03-29 15:49:45 +03:00
|
|
|
if (valid_atom[i].parser && valid_atom[i].parser(format, &used_atom[at], arg, err))
|
|
|
|
return -1;
|
2015-06-13 22:37:27 +03:00
|
|
|
if (*atom == '*')
|
|
|
|
need_tagged = 1;
|
2017-01-10 11:49:42 +03:00
|
|
|
if (!strcmp(valid_atom[i].name, "symref"))
|
2015-06-13 22:37:27 +03:00
|
|
|
need_symref = 1;
|
|
|
|
return at;
|
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:20 +03:00
|
|
|
static void quote_formatting(struct strbuf *s, const char *str, int quote_style)
|
|
|
|
{
|
|
|
|
switch (quote_style) {
|
|
|
|
case QUOTE_NONE:
|
|
|
|
strbuf_addstr(s, str);
|
|
|
|
break;
|
|
|
|
case QUOTE_SHELL:
|
|
|
|
sq_quote_buf(s, str);
|
|
|
|
break;
|
|
|
|
case QUOTE_PERL:
|
|
|
|
perl_quote_buf(s, str);
|
|
|
|
break;
|
|
|
|
case QUOTE_PYTHON:
|
|
|
|
python_quote_buf(s, str);
|
|
|
|
break;
|
|
|
|
case QUOTE_TCL:
|
|
|
|
tcl_quote_buf(s, str);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int append_atom(struct atom_value *v, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *unused_err)
|
2015-09-10 18:48:20 +03:00
|
|
|
{
|
2015-09-11 18:03:07 +03:00
|
|
|
/*
|
|
|
|
* Quote formatting is only done when the stack has a single
|
|
|
|
* element. Otherwise quote formatting is done on the
|
|
|
|
* element's entire output strbuf when the %(end) atom is
|
|
|
|
* encountered.
|
|
|
|
*/
|
|
|
|
if (!state->stack->prev)
|
|
|
|
quote_formatting(&state->stack->output, v->s, state->quote_style);
|
|
|
|
else
|
|
|
|
strbuf_addstr(&state->stack->output, v->s);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-09-10 18:48:20 +03:00
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:18 +03:00
|
|
|
static void push_stack_element(struct ref_formatting_stack **stack)
|
|
|
|
{
|
|
|
|
struct ref_formatting_stack *s = xcalloc(1, sizeof(struct ref_formatting_stack));
|
|
|
|
|
|
|
|
strbuf_init(&s->output, 0);
|
|
|
|
s->prev = *stack;
|
|
|
|
*stack = s;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pop_stack_element(struct ref_formatting_stack **stack)
|
|
|
|
{
|
|
|
|
struct ref_formatting_stack *current = *stack;
|
|
|
|
struct ref_formatting_stack *prev = current->prev;
|
|
|
|
|
|
|
|
if (prev)
|
|
|
|
strbuf_addbuf(&prev->output, ¤t->output);
|
|
|
|
strbuf_release(¤t->output);
|
|
|
|
free(current);
|
|
|
|
*stack = prev;
|
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:34 +03:00
|
|
|
static void end_align_handler(struct ref_formatting_stack **stack)
|
2015-09-11 18:03:07 +03:00
|
|
|
{
|
2017-01-10 11:49:34 +03:00
|
|
|
struct ref_formatting_stack *cur = *stack;
|
|
|
|
struct align *align = (struct align *)cur->at_end_data;
|
2015-09-11 18:03:07 +03:00
|
|
|
struct strbuf s = STRBUF_INIT;
|
|
|
|
|
2017-01-10 11:49:34 +03:00
|
|
|
strbuf_utf8_align(&s, align->position, align->width, cur->output.buf);
|
|
|
|
strbuf_swap(&cur->output, &s);
|
2015-09-11 18:03:07 +03:00
|
|
|
strbuf_release(&s);
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int align_atom_handler(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *unused_err)
|
2015-09-11 18:03:07 +03:00
|
|
|
{
|
2018-02-14 21:59:46 +03:00
|
|
|
struct ref_formatting_stack *new_stack;
|
2015-09-11 18:03:07 +03:00
|
|
|
|
|
|
|
push_stack_element(&state->stack);
|
2018-02-14 21:59:46 +03:00
|
|
|
new_stack = state->stack;
|
|
|
|
new_stack->at_end = end_align_handler;
|
|
|
|
new_stack->at_end_data = &atomv->atom->u.align;
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-09-11 18:03:07 +03:00
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:34 +03:00
|
|
|
static void if_then_else_handler(struct ref_formatting_stack **stack)
|
|
|
|
{
|
|
|
|
struct ref_formatting_stack *cur = *stack;
|
|
|
|
struct ref_formatting_stack *prev = cur->prev;
|
|
|
|
struct if_then_else *if_then_else = (struct if_then_else *)cur->at_end_data;
|
|
|
|
|
|
|
|
if (!if_then_else->then_atom_seen)
|
|
|
|
die(_("format: %%(if) atom used without a %%(then) atom"));
|
|
|
|
|
|
|
|
if (if_then_else->else_atom_seen) {
|
|
|
|
/*
|
|
|
|
* There is an %(else) atom: we need to drop one state from the
|
|
|
|
* stack, either the %(else) branch if the condition is satisfied, or
|
|
|
|
* the %(then) branch if it isn't.
|
|
|
|
*/
|
|
|
|
if (if_then_else->condition_satisfied) {
|
|
|
|
strbuf_reset(&cur->output);
|
|
|
|
pop_stack_element(&cur);
|
|
|
|
} else {
|
|
|
|
strbuf_swap(&cur->output, &prev->output);
|
|
|
|
strbuf_reset(&cur->output);
|
|
|
|
pop_stack_element(&cur);
|
|
|
|
}
|
|
|
|
} else if (!if_then_else->condition_satisfied) {
|
|
|
|
/*
|
|
|
|
* No %(else) atom: just drop the %(then) branch if the
|
|
|
|
* condition is not satisfied.
|
|
|
|
*/
|
|
|
|
strbuf_reset(&cur->output);
|
|
|
|
}
|
|
|
|
|
|
|
|
*stack = cur;
|
|
|
|
free(if_then_else);
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int if_atom_handler(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *unused_err)
|
2017-01-10 11:49:34 +03:00
|
|
|
{
|
2018-02-14 21:59:46 +03:00
|
|
|
struct ref_formatting_stack *new_stack;
|
2017-01-10 11:49:34 +03:00
|
|
|
struct if_then_else *if_then_else = xcalloc(sizeof(struct if_then_else), 1);
|
|
|
|
|
2017-01-10 11:49:36 +03:00
|
|
|
if_then_else->str = atomv->atom->u.if_then_else.str;
|
|
|
|
if_then_else->cmp_status = atomv->atom->u.if_then_else.cmp_status;
|
|
|
|
|
2017-01-10 11:49:34 +03:00
|
|
|
push_stack_element(&state->stack);
|
2018-02-14 21:59:46 +03:00
|
|
|
new_stack = state->stack;
|
|
|
|
new_stack->at_end = if_then_else_handler;
|
|
|
|
new_stack->at_end_data = if_then_else;
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2017-01-10 11:49:34 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
static int is_empty(const char *s)
|
|
|
|
{
|
|
|
|
while (*s != '\0') {
|
|
|
|
if (!isspace(*s))
|
|
|
|
return 0;
|
|
|
|
s++;
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int then_atom_handler(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *err)
|
2017-01-10 11:49:34 +03:00
|
|
|
{
|
|
|
|
struct ref_formatting_stack *cur = state->stack;
|
|
|
|
struct if_then_else *if_then_else = NULL;
|
|
|
|
|
|
|
|
if (cur->at_end == if_then_else_handler)
|
|
|
|
if_then_else = (struct if_then_else *)cur->at_end_data;
|
|
|
|
if (!if_then_else)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(then) atom used without an %%(if) atom"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if (if_then_else->then_atom_seen)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(then) atom used more than once"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if (if_then_else->else_atom_seen)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(then) atom used after %%(else)"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if_then_else->then_atom_seen = 1;
|
|
|
|
/*
|
2017-01-10 11:49:36 +03:00
|
|
|
* If the 'equals' or 'notequals' attribute is used then
|
|
|
|
* perform the required comparison. If not, only non-empty
|
|
|
|
* strings satisfy the 'if' condition.
|
2017-01-10 11:49:34 +03:00
|
|
|
*/
|
2017-01-10 11:49:36 +03:00
|
|
|
if (if_then_else->cmp_status == COMPARE_EQUAL) {
|
|
|
|
if (!strcmp(if_then_else->str, cur->output.buf))
|
|
|
|
if_then_else->condition_satisfied = 1;
|
|
|
|
} else if (if_then_else->cmp_status == COMPARE_UNEQUAL) {
|
|
|
|
if (strcmp(if_then_else->str, cur->output.buf))
|
|
|
|
if_then_else->condition_satisfied = 1;
|
|
|
|
} else if (cur->output.len && !is_empty(cur->output.buf))
|
2017-01-10 11:49:34 +03:00
|
|
|
if_then_else->condition_satisfied = 1;
|
|
|
|
strbuf_reset(&cur->output);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2017-01-10 11:49:34 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int else_atom_handler(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *err)
|
2017-01-10 11:49:34 +03:00
|
|
|
{
|
|
|
|
struct ref_formatting_stack *prev = state->stack;
|
|
|
|
struct if_then_else *if_then_else = NULL;
|
|
|
|
|
|
|
|
if (prev->at_end == if_then_else_handler)
|
|
|
|
if_then_else = (struct if_then_else *)prev->at_end_data;
|
|
|
|
if (!if_then_else)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(else) atom used without an %%(if) atom"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if (!if_then_else->then_atom_seen)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(else) atom used without a %%(then) atom"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if (if_then_else->else_atom_seen)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(else) atom used more than once"));
|
2017-01-10 11:49:34 +03:00
|
|
|
if_then_else->else_atom_seen = 1;
|
|
|
|
push_stack_element(&state->stack);
|
|
|
|
state->stack->at_end_data = prev->at_end_data;
|
|
|
|
state->stack->at_end = prev->at_end;
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-09-11 18:03:07 +03:00
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
static int end_atom_handler(struct atom_value *atomv, struct ref_formatting_state *state,
|
|
|
|
struct strbuf *err)
|
2015-09-11 18:03:07 +03:00
|
|
|
{
|
|
|
|
struct ref_formatting_stack *current = state->stack;
|
|
|
|
struct strbuf s = STRBUF_INIT;
|
|
|
|
|
|
|
|
if (!current->at_end)
|
2018-03-29 15:49:45 +03:00
|
|
|
return strbuf_addf_ret(err, -1, _("format: %%(end) atom used without corresponding atom"));
|
2017-01-10 11:49:34 +03:00
|
|
|
current->at_end(&state->stack);
|
|
|
|
|
|
|
|
/* Stack may have been popped within at_end(), hence reset the current pointer */
|
|
|
|
current = state->stack;
|
2015-09-11 18:03:07 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Perform quote formatting when the stack element is that of
|
|
|
|
* a supporting atom. If nested then perform quote formatting
|
|
|
|
* only on the topmost supporting atom.
|
|
|
|
*/
|
2017-01-10 11:49:34 +03:00
|
|
|
if (!current->prev->prev) {
|
2015-09-11 18:03:07 +03:00
|
|
|
quote_formatting(&s, current->output.buf, state->quote_style);
|
|
|
|
strbuf_swap(¤t->output, &s);
|
|
|
|
}
|
|
|
|
strbuf_release(&s);
|
|
|
|
pop_stack_element(&state->stack);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-09-11 18:03:07 +03:00
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* In a format string, find the next occurrence of %(atom).
|
|
|
|
*/
|
|
|
|
static const char *find_next(const char *cp)
|
|
|
|
{
|
|
|
|
while (*cp) {
|
|
|
|
if (*cp == '%') {
|
|
|
|
/*
|
|
|
|
* %( is the start of an atom;
|
|
|
|
* %% is a quoted per-cent.
|
|
|
|
*/
|
|
|
|
if (cp[1] == '(')
|
|
|
|
return cp;
|
|
|
|
else if (cp[1] == '%')
|
|
|
|
cp++; /* skip over two % */
|
|
|
|
/* otherwise this is a singleton, literal % */
|
|
|
|
}
|
|
|
|
cp++;
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Make sure the format string is well formed, and parse out
|
|
|
|
* the used atoms.
|
|
|
|
*/
|
2017-07-13 18:01:18 +03:00
|
|
|
int verify_ref_format(struct ref_format *format)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
const char *cp, *sp;
|
|
|
|
|
2017-07-13 18:02:30 +03:00
|
|
|
format->need_color_reset_at_eol = 0;
|
2017-07-13 18:01:18 +03:00
|
|
|
for (cp = format->format; *cp && (sp = find_next(cp)); ) {
|
2018-03-29 15:49:45 +03:00
|
|
|
struct strbuf err = STRBUF_INIT;
|
2015-06-13 22:37:27 +03:00
|
|
|
const char *color, *ep = strchr(sp, ')');
|
|
|
|
int at;
|
|
|
|
|
|
|
|
if (!ep)
|
2016-02-27 09:42:04 +03:00
|
|
|
return error(_("malformed format string %s"), sp);
|
2015-06-13 22:37:27 +03:00
|
|
|
/* sp points at "%(" and ep points at the closing ")" */
|
2018-03-29 15:49:45 +03:00
|
|
|
at = parse_ref_filter_atom(format, sp + 2, ep, &err);
|
|
|
|
if (at < 0)
|
|
|
|
die("%s", err.buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
cp = ep + 1;
|
|
|
|
|
2016-02-17 21:06:11 +03:00
|
|
|
if (skip_prefix(used_atom[at].name, "color:", &color))
|
2017-07-13 18:02:30 +03:00
|
|
|
format->need_color_reset_at_eol = !!strcmp(color, "reset");
|
2018-03-29 15:49:45 +03:00
|
|
|
strbuf_release(&err);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
ref-filter: consult want_color() before emitting colors
When color placeholders like %(color:red) are used in a
ref-filter format, we unconditionally output the colors,
even if the user has asked us for no colors. This usually
isn't a problem when the user is constructing a --format on
the command line, but it means we may do the wrong thing
when the format is fed from a script or alias. For example:
$ git config alias.b 'branch --format=%(color:green)%(refname)'
$ git b --no-color
should probably omit the green color. Likewise, running:
$ git b >branches
should probably also omit the color, just as we would for
all baked-in coloring (and as we recently started to do for
user-specified colors in --pretty formats).
This commit makes both of those cases work by teaching
the ref-filter code to consult want_color() before
outputting any color. The color flag in ref_format defaults
to "-1", which means we'll consult color.ui, which in turn
defaults to the usual isatty() check on stdout. However,
callers like git-branch which support their own color config
(and command-line options) can override that.
The new tests independently cover all three of the callers
of ref-filter (for-each-ref, tag, and branch). Even though
these seem redundant, it confirms that we've correctly
plumbed through all of the necessary config to make colors
work by default.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-07-13 18:09:32 +03:00
|
|
|
if (format->need_color_reset_at_eol && !want_color(format->use_color))
|
|
|
|
format->need_color_reset_at_eol = 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2020-08-22 00:41:46 +03:00
|
|
|
static const char *do_grab_oid(const char *field, const struct object_id *oid,
|
|
|
|
struct used_atom *atom)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
2020-08-22 00:41:46 +03:00
|
|
|
switch (atom->u.oid.option) {
|
2020-08-22 00:41:44 +03:00
|
|
|
case O_FULL:
|
|
|
|
return oid_to_hex(oid);
|
|
|
|
case O_LENGTH:
|
2020-08-22 00:41:46 +03:00
|
|
|
return find_unique_abbrev(oid, atom->u.oid.length);
|
2020-08-22 00:41:44 +03:00
|
|
|
case O_SHORT:
|
|
|
|
return find_unique_abbrev(oid, DEFAULT_ABBREV);
|
|
|
|
default:
|
|
|
|
BUG("unknown %%(%s) option", field);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-08-22 00:41:46 +03:00
|
|
|
static int grab_oid(const char *name, const char *field, const struct object_id *oid,
|
|
|
|
struct atom_value *v, struct used_atom *atom)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
2020-08-22 00:41:44 +03:00
|
|
|
if (starts_with(name, field)) {
|
2020-08-22 00:41:46 +03:00
|
|
|
v->s = xstrdup(do_grab_oid(field, oid, atom));
|
2020-08-22 00:41:44 +03:00
|
|
|
return 1;
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* See grab_values */
|
2018-07-17 11:22:57 +03:00
|
|
|
static void grab_common_values(struct atom_value *val, int deref, struct expand_data *oi)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
|
|
|
if (!strcmp(name, "objecttype"))
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(type_name(oi->type));
|
2018-12-24 16:24:30 +03:00
|
|
|
else if (!strcmp(name, "objectsize:disk")) {
|
|
|
|
v->value = oi->disk_size;
|
2019-01-10 21:15:49 +03:00
|
|
|
v->s = xstrfmt("%"PRIuMAX, (uintmax_t)oi->disk_size);
|
2018-12-24 16:24:30 +03:00
|
|
|
} else if (!strcmp(name, "objectsize")) {
|
2018-07-17 11:22:57 +03:00
|
|
|
v->value = oi->size;
|
2018-11-11 10:05:04 +03:00
|
|
|
v->s = xstrfmt("%"PRIuMAX , (uintmax_t)oi->size);
|
2018-12-24 16:24:30 +03:00
|
|
|
} else if (!strcmp(name, "deltabase"))
|
|
|
|
v->s = xstrdup(oid_to_hex(&oi->delta_base_oid));
|
2015-06-13 22:37:27 +03:00
|
|
|
else if (deref)
|
2020-08-22 00:41:46 +03:00
|
|
|
grab_oid(name, "objectname", &oi->oid, v, &used_atom[i]);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* See grab_values */
|
2019-02-14 08:50:54 +03:00
|
|
|
static void grab_tag_values(struct atom_value *val, int deref, struct object *obj)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
struct tag *tag = (struct tag *) obj;
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
|
|
|
if (!strcmp(name, "tag"))
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(tag->tag);
|
2015-06-13 22:37:27 +03:00
|
|
|
else if (!strcmp(name, "type") && tag->tagged)
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(type_name(tag->tagged->type));
|
2015-09-25 00:07:12 +03:00
|
|
|
else if (!strcmp(name, "object") && tag->tagged)
|
2015-11-10 05:22:28 +03:00
|
|
|
v->s = xstrdup(oid_to_hex(&tag->tagged->oid));
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* See grab_values */
|
2019-02-14 08:50:54 +03:00
|
|
|
static void grab_commit_values(struct atom_value *val, int deref, struct object *obj)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
struct commit *commit = (struct commit *) obj;
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
2020-08-22 00:41:47 +03:00
|
|
|
if (grab_oid(name, "tree", get_commit_tree_oid(commit), v, &used_atom[i]))
|
|
|
|
continue;
|
|
|
|
if (!strcmp(name, "numparent")) {
|
2017-04-20 23:52:09 +03:00
|
|
|
v->value = commit_list_count(commit->parents);
|
|
|
|
v->s = xstrfmt("%lu", (unsigned long)v->value);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2020-08-22 00:41:48 +03:00
|
|
|
else if (starts_with(name, "parent")) {
|
2015-06-13 22:37:27 +03:00
|
|
|
struct commit_list *parents;
|
2015-09-25 00:07:12 +03:00
|
|
|
struct strbuf s = STRBUF_INIT;
|
|
|
|
for (parents = commit->parents; parents; parents = parents->next) {
|
2020-08-22 00:41:48 +03:00
|
|
|
struct object_id *oid = &parents->item->object.oid;
|
2015-09-25 00:07:12 +03:00
|
|
|
if (parents != commit->parents)
|
|
|
|
strbuf_addch(&s, ' ');
|
2020-08-22 00:41:48 +03:00
|
|
|
strbuf_addstr(&s, do_grab_oid("parent", oid, &used_atom[i]));
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2015-09-25 00:07:12 +03:00
|
|
|
v->s = strbuf_detach(&s, NULL);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-02-14 08:51:03 +03:00
|
|
|
static const char *find_wholine(const char *who, int wholen, const char *buf)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
const char *eol;
|
|
|
|
while (*buf) {
|
|
|
|
if (!strncmp(buf, who, wholen) &&
|
|
|
|
buf[wholen] == ' ')
|
|
|
|
return buf + wholen + 1;
|
|
|
|
eol = strchr(buf, '\n');
|
|
|
|
if (!eol)
|
|
|
|
return "";
|
|
|
|
eol++;
|
|
|
|
if (*eol == '\n')
|
|
|
|
return ""; /* end of header */
|
|
|
|
buf = eol;
|
|
|
|
}
|
|
|
|
return "";
|
|
|
|
}
|
|
|
|
|
|
|
|
static const char *copy_line(const char *buf)
|
|
|
|
{
|
|
|
|
const char *eol = strchrnul(buf, '\n');
|
|
|
|
return xmemdupz(buf, eol - buf);
|
|
|
|
}
|
|
|
|
|
|
|
|
static const char *copy_name(const char *buf)
|
|
|
|
{
|
|
|
|
const char *cp;
|
|
|
|
for (cp = buf; *cp && *cp != '\n'; cp++) {
|
|
|
|
if (!strncmp(cp, " <", 2))
|
|
|
|
return xmemdupz(buf, cp - buf);
|
|
|
|
}
|
2019-08-18 00:51:07 +03:00
|
|
|
return xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
2020-08-22 00:41:43 +03:00
|
|
|
static const char *copy_email(const char *buf, struct used_atom *atom)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
const char *email = strchr(buf, '<');
|
|
|
|
const char *eoemail;
|
|
|
|
if (!email)
|
2019-08-18 00:51:07 +03:00
|
|
|
return xstrdup("");
|
2020-08-22 00:41:43 +03:00
|
|
|
switch (atom->u.email_option.option) {
|
|
|
|
case EO_RAW:
|
|
|
|
eoemail = strchr(email, '>');
|
|
|
|
if (eoemail)
|
|
|
|
eoemail++;
|
|
|
|
break;
|
|
|
|
case EO_TRIM:
|
|
|
|
email++;
|
|
|
|
eoemail = strchr(email, '>');
|
|
|
|
break;
|
|
|
|
case EO_LOCALPART:
|
|
|
|
email++;
|
|
|
|
eoemail = strchr(email, '@');
|
|
|
|
if (!eoemail)
|
|
|
|
eoemail = strchr(email, '>');
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
BUG("unknown email option");
|
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
if (!eoemail)
|
2019-08-18 00:51:07 +03:00
|
|
|
return xstrdup("");
|
2020-08-22 00:41:43 +03:00
|
|
|
return xmemdupz(email, eoemail - email);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
static char *copy_subject(const char *buf, unsigned long len)
|
|
|
|
{
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
struct strbuf sb = STRBUF_INIT;
|
2015-06-13 22:37:27 +03:00
|
|
|
int i;
|
|
|
|
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
for (i = 0; i < len; i++) {
|
|
|
|
if (buf[i] == '\r' && i + 1 < len && buf[i + 1] == '\n')
|
|
|
|
continue; /* ignore CR in CRLF */
|
2015-06-13 22:37:27 +03:00
|
|
|
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
if (buf[i] == '\n')
|
|
|
|
strbuf_addch(&sb, ' ');
|
|
|
|
else
|
|
|
|
strbuf_addch(&sb, buf[i]);
|
|
|
|
}
|
|
|
|
return strbuf_detach(&sb, NULL);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
static void grab_date(const char *buf, struct atom_value *v, const char *atomname)
|
|
|
|
{
|
|
|
|
const char *eoemail = strstr(buf, "> ");
|
|
|
|
char *zone;
|
2017-04-26 22:29:31 +03:00
|
|
|
timestamp_t timestamp;
|
2015-06-13 22:37:27 +03:00
|
|
|
long tz;
|
2015-08-03 21:01:27 +03:00
|
|
|
struct date_mode date_mode = { DATE_NORMAL };
|
2015-06-13 22:37:27 +03:00
|
|
|
const char *formatp;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We got here because atomname ends in "date" or "date<something>";
|
|
|
|
* it's not possible that <something> is not ":<format>" because
|
|
|
|
* parse_ref_filter_atom() wouldn't have allowed it, so we can assume that no
|
|
|
|
* ":" means no format is specified, and use the default.
|
|
|
|
*/
|
|
|
|
formatp = strchr(atomname, ':');
|
|
|
|
if (formatp != NULL) {
|
|
|
|
formatp++;
|
2015-08-03 21:01:27 +03:00
|
|
|
parse_date_format(formatp, &date_mode);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
if (!eoemail)
|
|
|
|
goto bad;
|
2017-04-21 13:45:44 +03:00
|
|
|
timestamp = parse_timestamp(eoemail + 2, &zone, 10);
|
2017-04-26 22:29:31 +03:00
|
|
|
if (timestamp == TIME_MAX)
|
2015-06-13 22:37:27 +03:00
|
|
|
goto bad;
|
|
|
|
tz = strtol(zone, NULL, 10);
|
|
|
|
if ((tz == LONG_MIN || tz == LONG_MAX) && errno == ERANGE)
|
|
|
|
goto bad;
|
2015-08-03 21:01:27 +03:00
|
|
|
v->s = xstrdup(show_date(timestamp, tz, &date_mode));
|
2017-04-20 23:52:09 +03:00
|
|
|
v->value = timestamp;
|
2015-06-13 22:37:27 +03:00
|
|
|
return;
|
|
|
|
bad:
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2017-04-20 23:52:09 +03:00
|
|
|
v->value = 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/* See grab_values */
|
2019-02-14 08:51:03 +03:00
|
|
|
static void grab_person(const char *who, struct atom_value *val, int deref, void *buf)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
int wholen = strlen(who);
|
|
|
|
const char *wholine = NULL;
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
|
|
|
if (strncmp(who, name, wholen))
|
|
|
|
continue;
|
|
|
|
if (name[wholen] != 0 &&
|
|
|
|
strcmp(name + wholen, "name") &&
|
2020-08-22 00:41:43 +03:00
|
|
|
!starts_with(name + wholen, "email") &&
|
2015-06-13 22:37:27 +03:00
|
|
|
!starts_with(name + wholen, "date"))
|
|
|
|
continue;
|
|
|
|
if (!wholine)
|
2019-02-14 08:51:03 +03:00
|
|
|
wholine = find_wholine(who, wholen, buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
if (!wholine)
|
|
|
|
return; /* no point looking for it */
|
|
|
|
if (name[wholen] == 0)
|
|
|
|
v->s = copy_line(wholine);
|
|
|
|
else if (!strcmp(name + wholen, "name"))
|
|
|
|
v->s = copy_name(wholine);
|
2020-08-22 00:41:43 +03:00
|
|
|
else if (starts_with(name + wholen, "email"))
|
|
|
|
v->s = copy_email(wholine, &used_atom[i]);
|
2015-06-13 22:37:27 +03:00
|
|
|
else if (starts_with(name + wholen, "date"))
|
|
|
|
grab_date(wholine, v, name);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* For a tag or a commit object, if "creator" or "creatordate" is
|
|
|
|
* requested, do something special.
|
|
|
|
*/
|
|
|
|
if (strcmp(who, "tagger") && strcmp(who, "committer"))
|
|
|
|
return; /* "author" for commit object is not wanted */
|
|
|
|
if (!wholine)
|
2019-02-14 08:51:03 +03:00
|
|
|
wholine = find_wholine(who, wholen, buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
if (!wholine)
|
|
|
|
return;
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
|
|
|
|
|
|
|
if (starts_with(name, "creatordate"))
|
|
|
|
grab_date(wholine, v, name);
|
|
|
|
else if (!strcmp(name, "creator"))
|
|
|
|
v->s = copy_line(wholine);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-02-14 08:51:03 +03:00
|
|
|
static void find_subpos(const char *buf,
|
2015-06-13 22:37:27 +03:00
|
|
|
const char **sub, unsigned long *sublen,
|
|
|
|
const char **body, unsigned long *bodylen,
|
|
|
|
unsigned long *nonsiglen,
|
|
|
|
const char **sig, unsigned long *siglen)
|
|
|
|
{
|
|
|
|
const char *eol;
|
|
|
|
/* skip past header until we hit empty line */
|
|
|
|
while (*buf && *buf != '\n') {
|
|
|
|
eol = strchrnul(buf, '\n');
|
|
|
|
if (*eol)
|
|
|
|
eol++;
|
|
|
|
buf = eol;
|
|
|
|
}
|
|
|
|
/* skip any empty lines */
|
|
|
|
while (*buf == '\n')
|
|
|
|
buf++;
|
|
|
|
|
|
|
|
/* parse signature first; we might not even have a subject line */
|
|
|
|
*sig = buf + parse_signature(buf, strlen(buf));
|
|
|
|
*siglen = strlen(*sig);
|
|
|
|
|
|
|
|
/* subject is first non-empty line */
|
|
|
|
*sub = buf;
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
/* subject goes to first empty line before signature begins */
|
|
|
|
if ((eol = strstr(*sub, "\n\n"))) {
|
|
|
|
eol = eol < *sig ? eol : *sig;
|
|
|
|
/* check if message uses CRLF */
|
|
|
|
} else if (! (eol = strstr(*sub, "\r\n\r\n"))) {
|
|
|
|
/* treat whole message as subject */
|
|
|
|
eol = strrchr(*sub, '\0');
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
buf = eol;
|
2015-06-13 22:37:27 +03:00
|
|
|
*sublen = buf - *sub;
|
|
|
|
/* drop trailing newline, if present */
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
while (*sublen && ((*sub)[*sublen - 1] == '\n' ||
|
|
|
|
(*sub)[*sublen - 1] == '\r'))
|
2015-06-13 22:37:27 +03:00
|
|
|
*sublen -= 1;
|
|
|
|
|
|
|
|
/* skip any empty lines */
|
ref-filter: handle CRLF at end-of-line more gracefully
The ref-filter code does not correctly handle commit or tag messages
that use CRLF as the line terminator. Such messages can be created with
the `--cleanup=verbatim` option of `git commit` and `git tag`, or by
using `git commit-tree` directly.
The function `find_subpos` in ref-filter.c looks for two consecutive
LFs to find the end of the subject line, a sequence which is absent in
messages using CRLF. This results in the whole message being parsed as
the subject line (`%(contents:subject)`), and the body of the message
(`%(contents:body)`) being empty.
Moreover, in `copy_subject`, which wants to return the subject as a
single line, '\n' is replaced by space, but '\r' is
untouched.
This impacts the output of `git branch`, `git tag` and `git
for-each-ref`.
This behaviour is a regression for `git branch --verbose`, which
bisects down to 949af0684c (branch: use ref-filter printing APIs,
2017-01-10).
Adjust the ref-filter code to be more lenient by hardening the logic in
`copy_subject` and `find_subpos` to correctly parse messages containing
CRLF.
Add a new test script, 't3920-crlf-messages.sh', to test the behaviour
of commands using either the ref-filter or the pretty APIs with messages
using CRLF line endings. The function `test_crlf_subject_body_and_contents`
can be used to test that the `--format` option of `branch`, `tag`,
`for-each-ref`, `log` and `show` correctly displays the subject, body
and raw content of commit and tag messages using CRLF. Test the
output of `branch`, `tag` and `for-each-ref` with such commits.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-29 15:48:28 +03:00
|
|
|
while (*buf == '\n' || *buf == '\r')
|
2015-06-13 22:37:27 +03:00
|
|
|
buf++;
|
|
|
|
*body = buf;
|
|
|
|
*bodylen = strlen(buf);
|
|
|
|
*nonsiglen = *sig - buf;
|
|
|
|
}
|
|
|
|
|
2015-09-11 18:04:16 +03:00
|
|
|
/*
|
|
|
|
* If 'lines' is greater than 0, append that many lines from the given
|
|
|
|
* 'buf' of length 'size' to the given strbuf.
|
|
|
|
*/
|
|
|
|
static void append_lines(struct strbuf *out, const char *buf, unsigned long size, int lines)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
const char *sp, *eol;
|
|
|
|
size_t len;
|
|
|
|
|
|
|
|
sp = buf;
|
|
|
|
|
|
|
|
for (i = 0; i < lines && sp < buf + size; i++) {
|
|
|
|
if (i)
|
|
|
|
strbuf_addstr(out, "\n ");
|
|
|
|
eol = memchr(sp, '\n', size - (sp - buf));
|
|
|
|
len = eol ? eol - sp : size - (sp - buf);
|
|
|
|
strbuf_add(out, sp, len);
|
|
|
|
if (!eol)
|
|
|
|
break;
|
|
|
|
sp = eol + 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/* See grab_values */
|
2019-02-14 08:51:03 +03:00
|
|
|
static void grab_sub_body_contents(struct atom_value *val, int deref, void *buf)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
const char *subpos = NULL, *bodypos = NULL, *sigpos = NULL;
|
|
|
|
unsigned long sublen = 0, bodylen = 0, nonsiglen = 0, siglen = 0;
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:18 +03:00
|
|
|
struct used_atom *atom = &used_atom[i];
|
|
|
|
const char *name = atom->name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (!!deref != (*name == '*'))
|
|
|
|
continue;
|
|
|
|
if (deref)
|
|
|
|
name++;
|
2020-08-22 00:41:50 +03:00
|
|
|
if (strcmp(name, "body") &&
|
|
|
|
!starts_with(name, "subject") &&
|
2017-10-02 08:25:23 +03:00
|
|
|
!starts_with(name, "trailers") &&
|
2016-02-17 21:06:18 +03:00
|
|
|
!starts_with(name, "contents"))
|
2015-06-13 22:37:27 +03:00
|
|
|
continue;
|
|
|
|
if (!subpos)
|
2019-02-14 08:51:03 +03:00
|
|
|
find_subpos(buf,
|
2015-06-13 22:37:27 +03:00
|
|
|
&subpos, &sublen,
|
|
|
|
&bodypos, &bodylen, &nonsiglen,
|
|
|
|
&sigpos, &siglen);
|
|
|
|
|
2016-02-17 21:06:18 +03:00
|
|
|
if (atom->u.contents.option == C_SUB)
|
2015-06-13 22:37:27 +03:00
|
|
|
v->s = copy_subject(subpos, sublen);
|
2020-08-22 00:41:50 +03:00
|
|
|
else if (atom->u.contents.option == C_SUB_SANITIZE) {
|
|
|
|
struct strbuf sb = STRBUF_INIT;
|
|
|
|
format_sanitized_subject(&sb, subpos, sublen);
|
|
|
|
v->s = strbuf_detach(&sb, NULL);
|
|
|
|
} else if (atom->u.contents.option == C_BODY_DEP)
|
2015-06-13 22:37:27 +03:00
|
|
|
v->s = xmemdupz(bodypos, bodylen);
|
2020-07-16 15:19:40 +03:00
|
|
|
else if (atom->u.contents.option == C_LENGTH)
|
|
|
|
v->s = xstrfmt("%"PRIuMAX, (uintmax_t)strlen(subpos));
|
2016-02-17 21:06:18 +03:00
|
|
|
else if (atom->u.contents.option == C_BODY)
|
2015-06-13 22:37:27 +03:00
|
|
|
v->s = xmemdupz(bodypos, nonsiglen);
|
2016-02-17 21:06:18 +03:00
|
|
|
else if (atom->u.contents.option == C_SIG)
|
2015-06-13 22:37:27 +03:00
|
|
|
v->s = xmemdupz(sigpos, siglen);
|
2016-02-17 21:06:18 +03:00
|
|
|
else if (atom->u.contents.option == C_LINES) {
|
2015-09-11 18:04:16 +03:00
|
|
|
struct strbuf s = STRBUF_INIT;
|
|
|
|
const char *contents_end = bodylen + bodypos - siglen;
|
|
|
|
|
|
|
|
/* Size is the length of the message after removing the signature */
|
2016-02-17 21:06:18 +03:00
|
|
|
append_lines(&s, subpos, contents_end - subpos, atom->u.contents.nlines);
|
2015-09-11 18:04:16 +03:00
|
|
|
v->s = strbuf_detach(&s, NULL);
|
2016-11-19 03:58:15 +03:00
|
|
|
} else if (atom->u.contents.option == C_TRAILERS) {
|
2017-10-02 08:25:23 +03:00
|
|
|
struct strbuf s = STRBUF_INIT;
|
2016-11-19 03:58:15 +03:00
|
|
|
|
2017-10-02 08:25:23 +03:00
|
|
|
/* Format the trailer info according to the trailer_opts given */
|
|
|
|
format_trailers_from_commit(&s, subpos, &atom->u.contents.trailer_opts);
|
|
|
|
|
|
|
|
v->s = strbuf_detach(&s, NULL);
|
2016-02-17 21:06:18 +03:00
|
|
|
} else if (atom->u.contents.option == C_BARE)
|
|
|
|
v->s = xstrdup(subpos);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We want to have empty print-string for field requests
|
|
|
|
* that do not apply (e.g. "authordate" for a tag object)
|
|
|
|
*/
|
|
|
|
static void fill_missing_values(struct atom_value *val)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
|
|
|
struct atom_value *v = &val[i];
|
|
|
|
if (v->s == NULL)
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* val is a list of atom_value to hold returned values. Extract
|
|
|
|
* the values for atoms in used_atom array out of (obj, buf, sz).
|
|
|
|
* when deref is false, (obj, buf, sz) is the object that is
|
|
|
|
* pointed at by the ref itself; otherwise it is the object the
|
|
|
|
* ref (which is a tag) refers to.
|
|
|
|
*/
|
2019-02-14 08:51:03 +03:00
|
|
|
static void grab_values(struct atom_value *val, int deref, struct object *obj, void *buf)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
switch (obj->type) {
|
|
|
|
case OBJ_TAG:
|
2019-02-14 08:50:54 +03:00
|
|
|
grab_tag_values(val, deref, obj);
|
2019-02-14 08:51:03 +03:00
|
|
|
grab_sub_body_contents(val, deref, buf);
|
|
|
|
grab_person("tagger", val, deref, buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
break;
|
|
|
|
case OBJ_COMMIT:
|
2019-02-14 08:50:54 +03:00
|
|
|
grab_commit_values(val, deref, obj);
|
2019-02-14 08:51:03 +03:00
|
|
|
grab_sub_body_contents(val, deref, buf);
|
|
|
|
grab_person("author", val, deref, buf);
|
|
|
|
grab_person("committer", val, deref, buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
break;
|
|
|
|
case OBJ_TREE:
|
|
|
|
/* grab_tree_values(val, deref, obj, buf, sz); */
|
|
|
|
break;
|
|
|
|
case OBJ_BLOB:
|
|
|
|
/* grab_blob_values(val, deref, obj, buf, sz); */
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
die("Eh? Object of type %d?", obj->type);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline char *copy_advance(char *dst, const char *src)
|
|
|
|
{
|
|
|
|
while (*src)
|
|
|
|
*dst++ = *src++;
|
|
|
|
return dst;
|
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:48 +03:00
|
|
|
static const char *lstrip_ref_components(const char *refname, int len)
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
{
|
2017-01-10 11:49:44 +03:00
|
|
|
long remaining = len;
|
2018-10-18 10:28:54 +03:00
|
|
|
const char *start = xstrdup(refname);
|
|
|
|
const char *to_free = start;
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
|
2017-01-10 11:49:48 +03:00
|
|
|
if (len < 0) {
|
|
|
|
int i;
|
|
|
|
const char *p = refname;
|
|
|
|
|
|
|
|
/* Find total no of '/' separated path-components */
|
|
|
|
for (i = 0; p[i]; p[i] == '/' ? i++ : *p++)
|
|
|
|
;
|
|
|
|
/*
|
|
|
|
* The number of components we need to strip is now
|
|
|
|
* the total minus the components to be left (Plus one
|
|
|
|
* because we count the number of '/', but the number
|
|
|
|
* of components is one more than the no of '/').
|
|
|
|
*/
|
|
|
|
remaining = i + len + 1;
|
|
|
|
}
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
|
2017-01-10 11:49:48 +03:00
|
|
|
while (remaining > 0) {
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
switch (*start++) {
|
|
|
|
case '\0':
|
2018-10-18 10:28:54 +03:00
|
|
|
free((char *)to_free);
|
|
|
|
return xstrdup("");
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
case '/':
|
|
|
|
remaining--;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2017-01-10 11:49:48 +03:00
|
|
|
|
2018-10-18 10:28:54 +03:00
|
|
|
start = xstrdup(start);
|
|
|
|
free((char *)to_free);
|
tag: do not show ambiguous tag names as "tags/foo"
Since b7cc53e9 (tag.c: use 'ref-filter' APIs, 2015-07-11),
git-tag has started showing tags with ambiguous names (i.e.,
when both "heads/foo" and "tags/foo" exists) as "tags/foo"
instead of just "foo". This is both:
- pointless; the output of "git tag" includes only
refs/tags, so we know that "foo" means the one in
"refs/tags".
and
- ambiguous; in the original output, we know that the line
"foo" means that "refs/tags/foo" exists. In the new
output, it is unclear whether we mean "refs/tags/foo" or
"refs/tags/tags/foo".
The reason this happens is that commit b7cc53e9 switched
git-tag to use ref-filter's "%(refname:short)" output
formatting, which was adapted from for-each-ref. This more
general code does not know that we care only about tags, and
uses shorten_unambiguous_ref to get the short-name. We need
to tell it that we care only about "refs/tags/", and it
should shorten with respect to that value.
In theory, the ref-filter code could figure this out by us
passing FILTER_REFS_TAGS. But there are two complications
there:
1. The handling of refname:short is deep in formatting
code that does not even have our ref_filter struct, let
alone the arguments to the filter_ref struct.
2. In git v2.7.0, we expose the formatting language to the
user. If we follow this path, it will mean that
"%(refname:short)" behaves differently for "tag" versus
"for-each-ref" (including "for-each-ref refs/tags/"),
which can lead to confusion.
Instead, let's add a new modifier to the formatting
language, "strip", to remove a specific set of prefix
components. This fixes "git tag", and lets users invoke the
same behavior from their own custom formats (for "tag" or
"for-each-ref") while leaving ":short" with its same
consistent meaning in all places.
We introduce a test in t7004 for "git tag", which fails
without this patch. We also add a similar test in t3203 for
"git branch", which does not actually fail. But since it is
likely that "branch" will eventually use the same formatting
code, the test helps defend against future regressions.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-01-26 06:00:05 +03:00
|
|
|
return start;
|
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:49 +03:00
|
|
|
static const char *rstrip_ref_components(const char *refname, int len)
|
|
|
|
{
|
|
|
|
long remaining = len;
|
2018-10-18 10:28:54 +03:00
|
|
|
const char *start = xstrdup(refname);
|
|
|
|
const char *to_free = start;
|
2017-01-10 11:49:49 +03:00
|
|
|
|
|
|
|
if (len < 0) {
|
|
|
|
int i;
|
|
|
|
const char *p = refname;
|
|
|
|
|
|
|
|
/* Find total no of '/' separated path-components */
|
|
|
|
for (i = 0; p[i]; p[i] == '/' ? i++ : *p++)
|
|
|
|
;
|
|
|
|
/*
|
|
|
|
* The number of components we need to strip is now
|
|
|
|
* the total minus the components to be left (Plus one
|
|
|
|
* because we count the number of '/', but the number
|
|
|
|
* of components is one more than the no of '/').
|
|
|
|
*/
|
|
|
|
remaining = i + len + 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (remaining-- > 0) {
|
|
|
|
char *p = strrchr(start, '/');
|
2018-10-18 10:28:54 +03:00
|
|
|
if (p == NULL) {
|
|
|
|
free((char *)to_free);
|
|
|
|
return xstrdup("");
|
|
|
|
} else
|
2017-01-10 11:49:49 +03:00
|
|
|
p[0] = '\0';
|
|
|
|
}
|
|
|
|
return start;
|
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:44 +03:00
|
|
|
static const char *show_ref(struct refname_atom *atom, const char *refname)
|
|
|
|
{
|
|
|
|
if (atom->option == R_SHORT)
|
|
|
|
return shorten_unambiguous_ref(refname, warn_ambiguous_refs);
|
2017-01-10 11:49:46 +03:00
|
|
|
else if (atom->option == R_LSTRIP)
|
|
|
|
return lstrip_ref_components(refname, atom->lstrip);
|
2017-01-10 11:49:49 +03:00
|
|
|
else if (atom->option == R_RSTRIP)
|
|
|
|
return rstrip_ref_components(refname, atom->rstrip);
|
2017-01-10 11:49:44 +03:00
|
|
|
else
|
2018-10-18 10:28:54 +03:00
|
|
|
return xstrdup(refname);
|
2017-01-10 11:49:44 +03:00
|
|
|
}
|
|
|
|
|
2016-02-17 21:06:17 +03:00
|
|
|
static void fill_remote_ref_details(struct used_atom *atom, const char *refname,
|
|
|
|
struct branch *branch, const char **s)
|
|
|
|
{
|
|
|
|
int num_ours, num_theirs;
|
2017-01-10 11:49:45 +03:00
|
|
|
if (atom->u.remote_ref.option == RR_REF)
|
|
|
|
*s = show_ref(&atom->u.remote_ref.refname, refname);
|
2017-01-10 11:49:41 +03:00
|
|
|
else if (atom->u.remote_ref.option == RR_TRACK) {
|
2018-01-09 21:50:15 +03:00
|
|
|
if (stat_tracking_info(branch, &num_ours, &num_theirs,
|
2019-04-16 15:16:46 +03:00
|
|
|
NULL, atom->u.remote_ref.push,
|
|
|
|
AHEAD_BEHIND_FULL) < 0) {
|
2017-01-10 11:49:50 +03:00
|
|
|
*s = xstrdup(msgs.gone);
|
2017-01-10 11:49:41 +03:00
|
|
|
} else if (!num_ours && !num_theirs)
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup("");
|
2016-02-17 21:06:17 +03:00
|
|
|
else if (!num_ours)
|
2017-01-10 11:49:50 +03:00
|
|
|
*s = xstrfmt(msgs.behind, num_theirs);
|
2016-02-17 21:06:17 +03:00
|
|
|
else if (!num_theirs)
|
2017-01-10 11:49:50 +03:00
|
|
|
*s = xstrfmt(msgs.ahead, num_ours);
|
2016-02-17 21:06:17 +03:00
|
|
|
else
|
2017-01-10 11:49:50 +03:00
|
|
|
*s = xstrfmt(msgs.ahead_behind,
|
2016-02-17 21:06:17 +03:00
|
|
|
num_ours, num_theirs);
|
2017-01-10 11:49:41 +03:00
|
|
|
if (!atom->u.remote_ref.nobracket && *s[0]) {
|
|
|
|
const char *to_free = *s;
|
|
|
|
*s = xstrfmt("[%s]", *s);
|
|
|
|
free((void *)to_free);
|
|
|
|
}
|
|
|
|
} else if (atom->u.remote_ref.option == RR_TRACKSHORT) {
|
2018-01-09 21:50:15 +03:00
|
|
|
if (stat_tracking_info(branch, &num_ours, &num_theirs,
|
2019-04-16 15:16:46 +03:00
|
|
|
NULL, atom->u.remote_ref.push,
|
|
|
|
AHEAD_BEHIND_FULL) < 0) {
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup("");
|
2016-02-17 21:06:17 +03:00
|
|
|
return;
|
2018-10-18 10:28:54 +03:00
|
|
|
}
|
2016-02-17 21:06:17 +03:00
|
|
|
if (!num_ours && !num_theirs)
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup("=");
|
2016-02-17 21:06:17 +03:00
|
|
|
else if (!num_ours)
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup("<");
|
2016-02-17 21:06:17 +03:00
|
|
|
else if (!num_theirs)
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup(">");
|
2016-02-17 21:06:17 +03:00
|
|
|
else
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup("<>");
|
2017-10-05 15:19:09 +03:00
|
|
|
} else if (atom->u.remote_ref.option == RR_REMOTE_NAME) {
|
|
|
|
int explicit;
|
|
|
|
const char *remote = atom->u.remote_ref.push ?
|
|
|
|
pushremote_for_branch(branch, &explicit) :
|
|
|
|
remote_for_branch(branch, &explicit);
|
2018-10-18 10:28:54 +03:00
|
|
|
*s = xstrdup(explicit ? remote : "");
|
2017-11-07 19:31:08 +03:00
|
|
|
} else if (atom->u.remote_ref.option == RR_REMOTE_REF) {
|
|
|
|
const char *merge;
|
|
|
|
|
2020-03-03 19:12:22 +03:00
|
|
|
merge = remote_ref_for_branch(branch, atom->u.remote_ref.push);
|
|
|
|
*s = xstrdup(merge ? merge : "");
|
2017-01-10 11:49:45 +03:00
|
|
|
} else
|
2018-05-02 12:38:39 +03:00
|
|
|
BUG("unhandled RR_* enum");
|
2016-02-17 21:06:17 +03:00
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:38 +03:00
|
|
|
char *get_head_description(void)
|
|
|
|
{
|
|
|
|
struct strbuf desc = STRBUF_INIT;
|
|
|
|
struct wt_status_state state;
|
|
|
|
memset(&state, 0, sizeof(state));
|
2018-11-10 08:48:50 +03:00
|
|
|
wt_status_get_state(the_repository, &state, 1);
|
2019-06-19 01:29:15 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The ( character must be hard-coded and not part of a localizable
|
|
|
|
* string, since the description is used as a sort key and compared
|
|
|
|
* with ref names.
|
|
|
|
*/
|
|
|
|
strbuf_addch(&desc, '(');
|
2017-01-10 11:49:38 +03:00
|
|
|
if (state.rebase_in_progress ||
|
2018-04-03 07:31:00 +03:00
|
|
|
state.rebase_interactive_in_progress) {
|
|
|
|
if (state.branch)
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addf(&desc, _("no branch, rebasing %s"),
|
2018-04-03 07:31:00 +03:00
|
|
|
state.branch);
|
|
|
|
else
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addf(&desc, _("no branch, rebasing detached HEAD %s"),
|
2018-04-03 07:31:00 +03:00
|
|
|
state.detached_from);
|
|
|
|
} else if (state.bisect_in_progress)
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addf(&desc, _("no branch, bisect started on %s"),
|
2017-01-10 11:49:38 +03:00
|
|
|
state.branch);
|
|
|
|
else if (state.detached_from) {
|
|
|
|
if (state.detached_at)
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addstr(&desc, HEAD_DETACHED_AT);
|
2017-01-10 11:49:38 +03:00
|
|
|
else
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addstr(&desc, HEAD_DETACHED_FROM);
|
|
|
|
strbuf_addstr(&desc, state.detached_from);
|
2017-01-10 11:49:38 +03:00
|
|
|
}
|
|
|
|
else
|
2019-06-19 01:29:15 +03:00
|
|
|
strbuf_addstr(&desc, _("no branch"));
|
|
|
|
strbuf_addch(&desc, ')');
|
|
|
|
|
wt-status: introduce wt_status_state_free_buffers()
When we have a `struct wt_status_state`, we manually free its `branch`,
`onto` and `detached_from`, or sometimes just one or two of them.
Provide a function `wt_status_state_free_buffers()` which does the
freeing.
The callers are still aware of these fields, e.g., they check whether
`branch` was populated or not. But this way, they don't need to know
about *all* of them, and if `struct wt_status_state` gets more fields,
they will not need to learn to free them.
Users of `struct wt_status` (which contains a `wt_status_state`) already
have `wt_status_collect_free_buffers()` (corresponding to
`wt_status_collect()`) which we can also teach to use this new helper.
Finally, note that we're currently leaving dangling pointers behind.
Some callers work on a stack-allocated struct, where this is obviously
ok. But for the users of `run_status()` in builtin/commit.c, there are
ample opportunities for someone to mistakenly use those dangling
pointers. We seem to be ok for now, but it's a use-after-free waiting to
happen. Let's leave NULL-pointers behind instead.
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-27 16:15:43 +03:00
|
|
|
wt_status_state_free_buffers(&state);
|
2017-01-10 11:49:38 +03:00
|
|
|
return strbuf_detach(&desc, NULL);
|
|
|
|
}
|
|
|
|
|
2017-01-10 11:49:44 +03:00
|
|
|
static const char *get_symref(struct used_atom *atom, struct ref_array_item *ref)
|
|
|
|
{
|
|
|
|
if (!ref->symref)
|
2018-10-18 10:28:54 +03:00
|
|
|
return xstrdup("");
|
2017-01-10 11:49:44 +03:00
|
|
|
else
|
|
|
|
return show_ref(&atom->u.refname, ref->symref);
|
|
|
|
}
|
|
|
|
|
|
|
|
static const char *get_refname(struct used_atom *atom, struct ref_array_item *ref)
|
|
|
|
{
|
|
|
|
if (ref->kind & FILTER_REFS_DETACHED_HEAD)
|
|
|
|
return get_head_description();
|
|
|
|
return show_ref(&atom->u.refname, ref->refname);
|
2016-02-17 21:06:17 +03:00
|
|
|
}
|
|
|
|
|
2018-07-17 11:22:57 +03:00
|
|
|
static int get_object(struct ref_array_item *ref, int deref, struct object **obj,
|
|
|
|
struct expand_data *oi, struct strbuf *err)
|
2018-02-21 09:59:00 +03:00
|
|
|
{
|
2018-07-17 11:22:57 +03:00
|
|
|
/* parse_object_buffer() will set eaten to 0 if free() will be needed */
|
|
|
|
int eaten = 1;
|
2018-07-17 11:22:57 +03:00
|
|
|
if (oi->info.contentp) {
|
|
|
|
/* We need to know that to use parse_object_buffer properly */
|
|
|
|
oi->info.sizep = &oi->size;
|
|
|
|
oi->info.typep = &oi->type;
|
|
|
|
}
|
|
|
|
if (oid_object_info_extended(the_repository, &oi->oid, &oi->info,
|
|
|
|
OBJECT_INFO_LOOKUP_REPLACE))
|
|
|
|
return strbuf_addf_ret(err, -1, _("missing object %s for %s"),
|
|
|
|
oid_to_hex(&oi->oid), ref->refname);
|
2018-12-24 16:24:30 +03:00
|
|
|
if (oi->info.disk_sizep && oi->disk_size < 0)
|
|
|
|
BUG("Object size is less than zero.");
|
2018-07-17 11:22:57 +03:00
|
|
|
|
|
|
|
if (oi->info.contentp) {
|
2018-08-17 23:09:57 +03:00
|
|
|
*obj = parse_object_buffer(the_repository, &oi->oid, oi->type, oi->size, oi->content, &eaten);
|
2018-07-17 11:22:57 +03:00
|
|
|
if (!obj) {
|
|
|
|
if (!eaten)
|
|
|
|
free(oi->content);
|
|
|
|
return strbuf_addf_ret(err, -1, _("parse_object_buffer failed on %s for %s"),
|
|
|
|
oid_to_hex(&oi->oid), ref->refname);
|
|
|
|
}
|
2019-02-14 08:51:03 +03:00
|
|
|
grab_values(ref->value, deref, *obj, oi->content);
|
2018-07-17 11:22:57 +03:00
|
|
|
}
|
2018-07-17 11:22:57 +03:00
|
|
|
|
|
|
|
grab_common_values(ref->value, deref, oi);
|
2018-02-21 09:59:00 +03:00
|
|
|
if (!eaten)
|
2018-07-17 11:22:57 +03:00
|
|
|
free(oi->content);
|
|
|
|
return 0;
|
2018-02-21 09:59:00 +03:00
|
|
|
}
|
|
|
|
|
2019-04-29 08:19:42 +03:00
|
|
|
static void populate_worktree_map(struct hashmap *map, struct worktree **worktrees)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; worktrees[i]; i++) {
|
|
|
|
if (worktrees[i]->head_ref) {
|
|
|
|
struct ref_to_worktree_entry *entry;
|
|
|
|
entry = xmalloc(sizeof(*entry));
|
|
|
|
entry->wt = worktrees[i];
|
2019-10-07 02:30:27 +03:00
|
|
|
hashmap_entry_init(&entry->ent,
|
|
|
|
strhash(worktrees[i]->head_ref));
|
2019-04-29 08:19:42 +03:00
|
|
|
|
2019-10-07 02:30:29 +03:00
|
|
|
hashmap_add(map, &entry->ent);
|
2019-04-29 08:19:42 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void lazy_init_worktree_map(void)
|
|
|
|
{
|
|
|
|
if (ref_to_worktree_map.worktrees)
|
|
|
|
return;
|
|
|
|
|
2020-06-20 02:35:44 +03:00
|
|
|
ref_to_worktree_map.worktrees = get_worktrees();
|
2019-04-29 08:19:42 +03:00
|
|
|
hashmap_init(&(ref_to_worktree_map.map), ref_to_worktree_map_cmpfnc, NULL, 0);
|
|
|
|
populate_worktree_map(&(ref_to_worktree_map.map), ref_to_worktree_map.worktrees);
|
|
|
|
}
|
|
|
|
|
|
|
|
static char *get_worktree_path(const struct used_atom *atom, const struct ref_array_item *ref)
|
|
|
|
{
|
2019-10-07 02:30:36 +03:00
|
|
|
struct hashmap_entry entry, *e;
|
2019-04-29 08:19:42 +03:00
|
|
|
struct ref_to_worktree_entry *lookup_result;
|
|
|
|
|
|
|
|
lazy_init_worktree_map();
|
|
|
|
|
|
|
|
hashmap_entry_init(&entry, strhash(ref->refname));
|
2019-10-07 02:30:36 +03:00
|
|
|
e = hashmap_get(&(ref_to_worktree_map.map), &entry, ref->refname);
|
2019-04-29 08:19:42 +03:00
|
|
|
|
2019-10-07 02:30:36 +03:00
|
|
|
if (!e)
|
2019-04-29 08:19:42 +03:00
|
|
|
return xstrdup("");
|
2019-10-07 02:30:36 +03:00
|
|
|
|
|
|
|
lookup_result = container_of(e, struct ref_to_worktree_entry, ent);
|
|
|
|
|
|
|
|
return xstrdup(lookup_result->wt->path);
|
2019-04-29 08:19:42 +03:00
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* Parse the object referred by ref, and grab needed value.
|
|
|
|
*/
|
2018-03-29 15:49:45 +03:00
|
|
|
static int populate_value(struct ref_array_item *ref, struct strbuf *err)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
struct object *obj;
|
2018-02-21 09:59:00 +03:00
|
|
|
int i;
|
2018-07-17 11:22:57 +03:00
|
|
|
struct object_info empty = OBJECT_INFO_INIT;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
ref->value = xcalloc(used_atom_cnt, sizeof(struct atom_value));
|
|
|
|
|
|
|
|
if (need_symref && (ref->flag & REF_ISSYMREF) && !ref->symref) {
|
|
|
|
ref->symref = resolve_refdup(ref->refname, RESOLVE_REF_READING,
|
2017-10-01 10:29:03 +03:00
|
|
|
NULL, NULL);
|
2015-06-13 22:37:27 +03:00
|
|
|
if (!ref->symref)
|
2018-10-18 10:28:54 +03:00
|
|
|
ref->symref = xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill in specials first */
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
2016-02-17 21:06:13 +03:00
|
|
|
struct used_atom *atom = &used_atom[i];
|
2016-02-17 21:06:11 +03:00
|
|
|
const char *name = used_atom[i].name;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *v = &ref->value[i];
|
|
|
|
int deref = 0;
|
|
|
|
const char *refname;
|
|
|
|
struct branch *branch = NULL;
|
|
|
|
|
2015-09-10 18:48:20 +03:00
|
|
|
v->handler = append_atom;
|
2017-01-10 11:49:35 +03:00
|
|
|
v->atom = atom;
|
2015-09-10 18:48:20 +03:00
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
if (*name == '*') {
|
|
|
|
deref = 1;
|
|
|
|
name++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (starts_with(name, "refname"))
|
2017-01-10 11:49:44 +03:00
|
|
|
refname = get_refname(atom, ref);
|
2019-04-29 08:19:42 +03:00
|
|
|
else if (!strcmp(name, "worktreepath")) {
|
|
|
|
if (ref->kind == FILTER_REFS_BRANCHES)
|
|
|
|
v->s = get_worktree_path(atom, ref);
|
|
|
|
else
|
|
|
|
v->s = xstrdup("");
|
|
|
|
continue;
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
else if (starts_with(name, "symref"))
|
2017-01-10 11:49:44 +03:00
|
|
|
refname = get_symref(atom, ref);
|
2015-06-13 22:37:27 +03:00
|
|
|
else if (starts_with(name, "upstream")) {
|
|
|
|
const char *branch_name;
|
|
|
|
/* only local branches may have an upstream */
|
|
|
|
if (!skip_prefix(ref->refname, "refs/heads/",
|
2018-10-18 10:28:54 +03:00
|
|
|
&branch_name)) {
|
|
|
|
v->s = xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
continue;
|
2018-10-18 10:28:54 +03:00
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
branch = branch_get(branch_name);
|
|
|
|
|
|
|
|
refname = branch_get_upstream(branch, NULL);
|
2016-02-17 21:06:17 +03:00
|
|
|
if (refname)
|
|
|
|
fill_remote_ref_details(atom, refname, branch, &v->s);
|
2018-10-18 10:28:54 +03:00
|
|
|
else
|
|
|
|
v->s = xstrdup("");
|
2016-02-17 21:06:17 +03:00
|
|
|
continue;
|
2017-10-05 15:19:09 +03:00
|
|
|
} else if (atom->u.remote_ref.push) {
|
2015-06-13 22:37:27 +03:00
|
|
|
const char *branch_name;
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
if (!skip_prefix(ref->refname, "refs/heads/",
|
|
|
|
&branch_name))
|
|
|
|
continue;
|
|
|
|
branch = branch_get(branch_name);
|
|
|
|
|
2017-10-05 15:19:09 +03:00
|
|
|
if (atom->u.remote_ref.push_remote)
|
|
|
|
refname = NULL;
|
|
|
|
else {
|
|
|
|
refname = branch_get_push(branch, NULL);
|
|
|
|
if (!refname)
|
|
|
|
continue;
|
|
|
|
}
|
2018-10-18 10:28:54 +03:00
|
|
|
/* We will definitely re-init v->s on the next line. */
|
|
|
|
free((char *)v->s);
|
2016-02-17 21:06:17 +03:00
|
|
|
fill_remote_ref_details(atom, refname, branch, &v->s);
|
|
|
|
continue;
|
2016-02-17 21:06:13 +03:00
|
|
|
} else if (starts_with(name, "color:")) {
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(atom->u.color);
|
2015-06-13 22:37:27 +03:00
|
|
|
continue;
|
|
|
|
} else if (!strcmp(name, "flag")) {
|
|
|
|
char buf[256], *cp = buf;
|
|
|
|
if (ref->flag & REF_ISSYMREF)
|
|
|
|
cp = copy_advance(cp, ",symref");
|
|
|
|
if (ref->flag & REF_ISPACKED)
|
|
|
|
cp = copy_advance(cp, ",packed");
|
|
|
|
if (cp == buf)
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2015-06-13 22:37:27 +03:00
|
|
|
else {
|
|
|
|
*cp = '\0';
|
|
|
|
v->s = xstrdup(buf + 1);
|
|
|
|
}
|
|
|
|
continue;
|
2020-08-22 00:41:46 +03:00
|
|
|
} else if (!deref && grab_oid(name, "objectname", &ref->objectname, v, atom)) {
|
2015-06-13 22:37:27 +03:00
|
|
|
continue;
|
|
|
|
} else if (!strcmp(name, "HEAD")) {
|
2017-05-19 09:12:12 +03:00
|
|
|
if (atom->u.head && !strcmp(ref->refname, atom->u.head))
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("*");
|
2015-06-13 22:37:27 +03:00
|
|
|
else
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(" ");
|
2015-06-13 22:37:27 +03:00
|
|
|
continue;
|
2016-02-17 21:06:15 +03:00
|
|
|
} else if (starts_with(name, "align")) {
|
2015-09-11 18:03:07 +03:00
|
|
|
v->handler = align_atom_handler;
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2015-09-11 18:03:07 +03:00
|
|
|
continue;
|
|
|
|
} else if (!strcmp(name, "end")) {
|
|
|
|
v->handler = end_atom_handler;
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2015-09-11 18:03:07 +03:00
|
|
|
continue;
|
2017-01-10 11:49:36 +03:00
|
|
|
} else if (starts_with(name, "if")) {
|
|
|
|
const char *s;
|
|
|
|
if (skip_prefix(name, "if:", &s))
|
|
|
|
v->s = xstrdup(s);
|
2018-10-18 10:28:54 +03:00
|
|
|
else
|
|
|
|
v->s = xstrdup("");
|
2017-01-10 11:49:34 +03:00
|
|
|
v->handler = if_atom_handler;
|
|
|
|
continue;
|
|
|
|
} else if (!strcmp(name, "then")) {
|
|
|
|
v->handler = then_atom_handler;
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2017-01-10 11:49:34 +03:00
|
|
|
continue;
|
|
|
|
} else if (!strcmp(name, "else")) {
|
|
|
|
v->handler = else_atom_handler;
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup("");
|
2017-01-10 11:49:34 +03:00
|
|
|
continue;
|
2015-06-13 22:37:27 +03:00
|
|
|
} else
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (!deref)
|
2018-10-18 10:28:54 +03:00
|
|
|
v->s = xstrdup(refname);
|
2015-09-25 00:07:12 +03:00
|
|
|
else
|
|
|
|
v->s = xstrfmt("%s^{}", refname);
|
2018-10-18 10:28:54 +03:00
|
|
|
free((char *)refname);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++) {
|
|
|
|
struct atom_value *v = &ref->value[i];
|
2018-07-17 11:22:57 +03:00
|
|
|
if (v->s == NULL && used_atom[i].source == SOURCE_NONE)
|
|
|
|
return strbuf_addf_ret(err, -1, _("missing object %s for %s"),
|
|
|
|
oid_to_hex(&ref->objectname), ref->refname);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2018-07-17 11:22:57 +03:00
|
|
|
|
|
|
|
if (need_tagged)
|
|
|
|
oi.info.contentp = &oi.content;
|
|
|
|
if (!memcmp(&oi.info, &empty, sizeof(empty)) &&
|
|
|
|
!memcmp(&oi_deref.info, &empty, sizeof(empty)))
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
2018-07-17 11:22:57 +03:00
|
|
|
|
|
|
|
oi.oid = ref->objectname;
|
|
|
|
if (get_object(ref, 0, &obj, &oi, err))
|
2018-03-29 15:49:45 +03:00
|
|
|
return -1;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If there is no atom that wants to know about tagged
|
|
|
|
* object, we are done.
|
|
|
|
*/
|
|
|
|
if (!need_tagged || (obj->type != OBJ_TAG))
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If it is a tag object, see if we use a value that derefs
|
|
|
|
* the object, and if we do grab the object it refers to.
|
|
|
|
*/
|
2019-09-05 22:59:42 +03:00
|
|
|
oi_deref.oid = *get_tagged_oid((struct tag *)obj);
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* NEEDSWORK: This derefs tag only once, which
|
|
|
|
* is good to deal with chains of trust, but
|
|
|
|
* is not consistent with what deref_tag() does
|
|
|
|
* which peels the onion to the core.
|
|
|
|
*/
|
2018-07-17 11:22:57 +03:00
|
|
|
return get_object(ref, 1, &obj, &oi_deref, err);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Given a ref, return the value for the atom. This lazily gets value
|
|
|
|
* out of the object by calling populate value.
|
|
|
|
*/
|
2018-03-29 15:49:45 +03:00
|
|
|
static int get_ref_atom_value(struct ref_array_item *ref, int atom,
|
|
|
|
struct atom_value **v, struct strbuf *err)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
if (!ref->value) {
|
2018-03-29 15:49:45 +03:00
|
|
|
if (populate_value(ref, err))
|
|
|
|
return -1;
|
2015-06-13 22:37:27 +03:00
|
|
|
fill_missing_values(ref->value);
|
|
|
|
}
|
|
|
|
*v = &ref->value[atom];
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:26 +03:00
|
|
|
/*
|
|
|
|
* Return 1 if the refname matches one of the patterns, otherwise 0.
|
|
|
|
* A pattern can be a literal prefix (e.g. a refname "refs/heads/master"
|
|
|
|
* matches a pattern "refs/heads/mas") or a wildcard (e.g. the same ref
|
|
|
|
* matches "refs/heads/mas*", too).
|
|
|
|
*/
|
2016-12-04 05:52:25 +03:00
|
|
|
static int match_pattern(const struct ref_filter *filter, const char *refname)
|
2015-09-10 18:48:26 +03:00
|
|
|
{
|
2016-12-04 05:52:25 +03:00
|
|
|
const char **patterns = filter->name_patterns;
|
|
|
|
unsigned flags = 0;
|
|
|
|
|
|
|
|
if (filter->ignore_case)
|
|
|
|
flags |= WM_CASEFOLD;
|
|
|
|
|
2015-09-10 18:48:26 +03:00
|
|
|
/*
|
|
|
|
* When no '--format' option is given we need to skip the prefix
|
|
|
|
* for matching refs of tags and branches.
|
|
|
|
*/
|
|
|
|
(void)(skip_prefix(refname, "refs/tags/", &refname) ||
|
|
|
|
skip_prefix(refname, "refs/heads/", &refname) ||
|
|
|
|
skip_prefix(refname, "refs/remotes/", &refname) ||
|
|
|
|
skip_prefix(refname, "refs/", &refname));
|
|
|
|
|
|
|
|
for (; *patterns; patterns++) {
|
2017-06-23 00:38:08 +03:00
|
|
|
if (!wildmatch(*patterns, refname, flags))
|
2015-09-10 18:48:26 +03:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* Return 1 if the refname matches one of the patterns, otherwise 0.
|
|
|
|
* A pattern can be path prefix (e.g. a refname "refs/heads/master"
|
2015-09-10 18:48:26 +03:00
|
|
|
* matches a pattern "refs/heads/" but not "refs/heads/m") or a
|
|
|
|
* wildcard (e.g. the same ref matches "refs/heads/m*", too).
|
2015-06-13 22:37:27 +03:00
|
|
|
*/
|
2016-12-04 05:52:25 +03:00
|
|
|
static int match_name_as_path(const struct ref_filter *filter, const char *refname)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
2016-12-04 05:52:25 +03:00
|
|
|
const char **pattern = filter->name_patterns;
|
2015-06-13 22:37:27 +03:00
|
|
|
int namelen = strlen(refname);
|
2016-12-04 05:52:25 +03:00
|
|
|
unsigned flags = WM_PATHNAME;
|
|
|
|
|
|
|
|
if (filter->ignore_case)
|
|
|
|
flags |= WM_CASEFOLD;
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
for (; *pattern; pattern++) {
|
|
|
|
const char *p = *pattern;
|
|
|
|
int plen = strlen(p);
|
|
|
|
|
|
|
|
if ((plen <= namelen) &&
|
|
|
|
!strncmp(refname, p, plen) &&
|
|
|
|
(refname[plen] == '\0' ||
|
|
|
|
refname[plen] == '/' ||
|
|
|
|
p[plen-1] == '/'))
|
|
|
|
return 1;
|
2018-07-03 00:11:59 +03:00
|
|
|
if (!wildmatch(p, refname, flags))
|
2015-06-13 22:37:27 +03:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:26 +03:00
|
|
|
/* Return 1 if the refname matches one of the patterns, otherwise 0. */
|
|
|
|
static int filter_pattern_match(struct ref_filter *filter, const char *refname)
|
|
|
|
{
|
|
|
|
if (!*filter->name_patterns)
|
|
|
|
return 1; /* No pattern always matches */
|
|
|
|
if (filter->match_as_path)
|
2016-12-04 05:52:25 +03:00
|
|
|
return match_name_as_path(filter, refname);
|
|
|
|
return match_pattern(filter, refname);
|
2015-09-10 18:48:26 +03:00
|
|
|
}
|
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
static int qsort_strcmp(const void *va, const void *vb)
|
|
|
|
{
|
|
|
|
const char *a = *(const char **)va;
|
|
|
|
const char *b = *(const char **)vb;
|
|
|
|
|
|
|
|
return strcmp(a, b);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void find_longest_prefixes_1(struct string_list *out,
|
|
|
|
struct strbuf *prefix,
|
|
|
|
const char **patterns, size_t nr)
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
{
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
size_t i;
|
|
|
|
|
|
|
|
for (i = 0; i < nr; i++) {
|
|
|
|
char c = patterns[i][prefix->len];
|
|
|
|
if (!c || is_glob_special(c)) {
|
|
|
|
string_list_append(out, prefix->buf);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
i = 0;
|
|
|
|
while (i < nr) {
|
|
|
|
size_t end;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Set "end" to the index of the element _after_ the last one
|
|
|
|
* in our group.
|
|
|
|
*/
|
|
|
|
for (end = i + 1; end < nr; end++) {
|
|
|
|
if (patterns[i][prefix->len] != patterns[end][prefix->len])
|
|
|
|
break;
|
|
|
|
}
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
strbuf_addch(prefix, patterns[i][prefix->len]);
|
|
|
|
find_longest_prefixes_1(out, prefix, patterns + i, end - i);
|
|
|
|
strbuf_setlen(prefix, prefix->len - 1);
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
i = end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void find_longest_prefixes(struct string_list *out,
|
|
|
|
const char **patterns)
|
|
|
|
{
|
2020-07-28 23:25:12 +03:00
|
|
|
struct strvec sorted = STRVEC_INIT;
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
struct strbuf prefix = STRBUF_INIT;
|
|
|
|
|
2020-07-28 23:25:12 +03:00
|
|
|
strvec_pushv(&sorted, patterns);
|
2020-07-29 03:37:20 +03:00
|
|
|
QSORT(sorted.v, sorted.nr, qsort_strcmp);
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
|
2020-07-29 03:37:20 +03:00
|
|
|
find_longest_prefixes_1(out, &prefix, sorted.v, sorted.nr);
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
|
2020-07-28 23:25:12 +03:00
|
|
|
strvec_clear(&sorted);
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
strbuf_release(&prefix);
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This is the same as for_each_fullref_in(), but it tries to iterate
|
|
|
|
* only over the patterns we'll care about. Note that it _doesn't_ do a full
|
|
|
|
* pattern match, so the callback still has to match each ref individually.
|
|
|
|
*/
|
|
|
|
static int for_each_fullref_in_pattern(struct ref_filter *filter,
|
|
|
|
each_ref_fn cb,
|
|
|
|
void *cb_data,
|
|
|
|
int broken)
|
|
|
|
{
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
struct string_list prefixes = STRING_LIST_INIT_DUP;
|
|
|
|
struct string_list_item *prefix;
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (!filter->match_as_path) {
|
|
|
|
/*
|
|
|
|
* in this case, the patterns are applied after
|
|
|
|
* prefixes like "refs/heads/" etc. are stripped off,
|
|
|
|
* so we have to look at everything:
|
|
|
|
*/
|
|
|
|
return for_each_fullref_in("", cb, cb_data, broken);
|
|
|
|
}
|
|
|
|
|
2018-07-03 00:12:42 +03:00
|
|
|
if (filter->ignore_case) {
|
|
|
|
/*
|
|
|
|
* we can't handle case-insensitive comparisons,
|
|
|
|
* so just return everything and let the caller
|
|
|
|
* sort it out.
|
|
|
|
*/
|
|
|
|
return for_each_fullref_in("", cb, cb_data, broken);
|
|
|
|
}
|
|
|
|
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
if (!filter->name_patterns[0]) {
|
|
|
|
/* no patterns; we have to look at everything */
|
|
|
|
return for_each_fullref_in("", cb, cb_data, broken);
|
|
|
|
}
|
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
find_longest_prefixes(&prefixes, filter->name_patterns);
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
for_each_string_list_item(prefix, &prefixes) {
|
|
|
|
ret = for_each_fullref_in(prefix->string, cb, cb_data, broken);
|
|
|
|
if (ret)
|
|
|
|
break;
|
|
|
|
}
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
|
ref-filter.c: find disjoint pattern prefixes
Since cfe004a5a9 (ref-filter: limit traversal to prefix, 2017-05-22),
the ref-filter code has sought to limit the traversals to a prefix of
the given patterns.
That code stopped short of handling more than one pattern, because it
means invoking 'for_each_ref_in' multiple times. If we're not careful
about which patterns overlap, we will output the same refs multiple
times.
For instance, consider the set of patterns 'refs/heads/a/*',
'refs/heads/a/b/c', and 'refs/tags/v1.0.0'. If we naïvely ran:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/heads/a/b/c", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
we would see 'refs/heads/a/b/c' (and everything underneath it) twice.
Instead, we want to partition the patterns into disjoint sets, where we
know that no ref will be matched by any two patterns in different sets.
In the above, these are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'}, and
- {'refs/tags/v1.0.0'}
Given one of these disjoint sets, what is a suitable pattern to pass to
'for_each_ref_in'? One approach is to compute the longest common prefix
over all elements in that disjoint set, and let the caller cull out the
refs they didn't want. Computing the longest prefix means that in most
cases, we won't match too many things the caller would like to ignore.
The longest common prefixes of the above are:
- {'refs/heads/a/*', 'refs/heads/a/b/c'} -> refs/heads/a/*
- {'refs/tags/v1.0.0'} -> refs/tags/v1.0.0
We instead invoke:
for_each_ref_in("refs/heads/a/*", ...);
for_each_ref_in("refs/tags/v1.0.0", ...);
Which provides us with the refs we were looking for with a minimal
amount of extra cruft, but never a duplicate of the ref we asked for.
Implemented here is an algorithm which accomplishes the above, which
works as follows:
1. Lexicographically sort the given list of patterns.
2. Initialize 'prefix' to the empty string, where our goal is to
build each element in the above set of longest common prefixes.
3. Consider each pattern in the given set, and emit 'prefix' if it
reaches the end of a pattern, or touches a wildcard character. The
end of a string is treated as if it precedes a wildcard. (Note that
there is some room for future work to detect that, e.g., 'a?b' and
'abc' are disjoint).
4. Otherwise, recurse on step (3) with the slice of the list
corresponding to our current prefix (i.e., the subset of patterns
that have our prefix as a literal string prefix.)
This algorithm is 'O(kn + n log(n))', where 'k' is max(len(pattern)) for
each pattern in the list, and 'n' is len(patterns).
By discovering this set of interesting patterns, we reduce the runtime
of multi-pattern 'git for-each-ref' (and other ref traversals) from
O(N) to O(n log(N)), where 'N' is the total number of packed references.
Running 'git for-each-ref refs/tags/a refs/tags/b' on a repository with
10,000,000 refs in 'refs/tags/huge-N', my best-of-five times drop from:
real 0m5.805s
user 0m5.188s
sys 0m0.468s
to:
real 0m0.001s
user 0m0.000s
sys 0m0.000s
On linux.git, the times to dig out two of the latest -rc tags drops from
0.002s to 0.001s, so the change on repositories with fewer tags is much
less noticeable.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-06-27 01:41:48 +03:00
|
|
|
string_list_clear(&prefixes, 0);
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-07-07 19:06:09 +03:00
|
|
|
/*
|
2020-03-30 17:04:11 +03:00
|
|
|
* Given a ref (oid, refname), check if the ref belongs to the array
|
|
|
|
* of oids. If the given ref is a tag, check if the given tag points
|
|
|
|
* at one of the oids in the given oid array.
|
2015-07-07 19:06:09 +03:00
|
|
|
* NEEDSWORK:
|
2020-07-29 06:33:28 +03:00
|
|
|
* 1. Only a single level of indirection is obtained, we might want to
|
2015-07-07 19:06:09 +03:00
|
|
|
* change this to account for multiple levels (e.g. annotated tags
|
|
|
|
* pointing to annotated tags pointing to a commit.)
|
|
|
|
* 2. As the refs are cached we might know what refname peels to without
|
|
|
|
* the need to parse the object via parse_object(). peel_ref() might be a
|
|
|
|
* more efficient alternative to obtain the pointee.
|
|
|
|
*/
|
2017-03-31 04:40:00 +03:00
|
|
|
static const struct object_id *match_points_at(struct oid_array *points_at,
|
2017-03-31 04:39:57 +03:00
|
|
|
const struct object_id *oid,
|
|
|
|
const char *refname)
|
2015-07-07 19:06:09 +03:00
|
|
|
{
|
2017-03-31 04:39:57 +03:00
|
|
|
const struct object_id *tagged_oid = NULL;
|
2015-07-07 19:06:09 +03:00
|
|
|
struct object *obj;
|
|
|
|
|
2017-03-31 04:40:00 +03:00
|
|
|
if (oid_array_lookup(points_at, oid) >= 0)
|
2017-03-31 04:39:57 +03:00
|
|
|
return oid;
|
2018-06-29 04:21:51 +03:00
|
|
|
obj = parse_object(the_repository, oid);
|
2015-07-07 19:06:09 +03:00
|
|
|
if (!obj)
|
|
|
|
die(_("malformed object at '%s'"), refname);
|
|
|
|
if (obj->type == OBJ_TAG)
|
2019-09-05 22:59:42 +03:00
|
|
|
tagged_oid = get_tagged_oid((struct tag *)obj);
|
2017-03-31 04:40:00 +03:00
|
|
|
if (tagged_oid && oid_array_lookup(points_at, tagged_oid) >= 0)
|
2017-03-31 04:39:57 +03:00
|
|
|
return tagged_oid;
|
2015-07-07 19:06:09 +03:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2018-04-06 21:59:26 +03:00
|
|
|
/*
|
|
|
|
* Allocate space for a new ref_array_item and copy the name and oid to it.
|
|
|
|
*
|
|
|
|
* Callers can then fill in other struct members at their leisure.
|
|
|
|
*/
|
2015-06-13 22:37:27 +03:00
|
|
|
static struct ref_array_item *new_ref_array_item(const char *refname,
|
2018-04-06 21:59:26 +03:00
|
|
|
const struct object_id *oid)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
2016-02-23 01:44:32 +03:00
|
|
|
struct ref_array_item *ref;
|
2018-04-06 21:59:26 +03:00
|
|
|
|
2016-02-23 01:44:32 +03:00
|
|
|
FLEX_ALLOC_STR(ref, refname, refname);
|
2018-04-06 21:58:32 +03:00
|
|
|
oidcpy(&ref->objectname, oid);
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
return ref;
|
|
|
|
}
|
|
|
|
|
2018-04-06 21:59:45 +03:00
|
|
|
struct ref_array_item *ref_array_push(struct ref_array *array,
|
|
|
|
const char *refname,
|
|
|
|
const struct object_id *oid)
|
|
|
|
{
|
|
|
|
struct ref_array_item *ref = new_ref_array_item(refname, oid);
|
|
|
|
|
|
|
|
ALLOC_GROW(array->items, array->nr + 1, array->alloc);
|
|
|
|
array->items[array->nr++] = ref;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
return ref;
|
|
|
|
}
|
|
|
|
|
2017-01-18 02:37:19 +03:00
|
|
|
static int ref_kind_from_refname(const char *refname)
|
2015-09-10 18:48:23 +03:00
|
|
|
{
|
|
|
|
unsigned int i;
|
|
|
|
|
|
|
|
static struct {
|
|
|
|
const char *prefix;
|
|
|
|
unsigned int kind;
|
|
|
|
} ref_kind[] = {
|
|
|
|
{ "refs/heads/" , FILTER_REFS_BRANCHES },
|
|
|
|
{ "refs/remotes/" , FILTER_REFS_REMOTES },
|
|
|
|
{ "refs/tags/", FILTER_REFS_TAGS}
|
|
|
|
};
|
|
|
|
|
2017-01-18 02:37:19 +03:00
|
|
|
if (!strcmp(refname, "HEAD"))
|
2015-09-10 18:48:23 +03:00
|
|
|
return FILTER_REFS_DETACHED_HEAD;
|
|
|
|
|
|
|
|
for (i = 0; i < ARRAY_SIZE(ref_kind); i++) {
|
|
|
|
if (starts_with(refname, ref_kind[i].prefix))
|
|
|
|
return ref_kind[i].kind;
|
|
|
|
}
|
|
|
|
|
|
|
|
return FILTER_REFS_OTHERS;
|
|
|
|
}
|
|
|
|
|
2017-01-18 02:37:19 +03:00
|
|
|
static int filter_ref_kind(struct ref_filter *filter, const char *refname)
|
|
|
|
{
|
|
|
|
if (filter->kind == FILTER_REFS_BRANCHES ||
|
|
|
|
filter->kind == FILTER_REFS_REMOTES ||
|
|
|
|
filter->kind == FILTER_REFS_TAGS)
|
|
|
|
return filter->kind;
|
|
|
|
return ref_kind_from_refname(refname);
|
|
|
|
}
|
|
|
|
|
2018-07-20 19:33:08 +03:00
|
|
|
struct ref_filter_cbdata {
|
|
|
|
struct ref_array *array;
|
|
|
|
struct ref_filter *filter;
|
|
|
|
struct contains_cache contains_cache;
|
|
|
|
struct contains_cache no_contains_cache;
|
|
|
|
};
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* A call-back given to for_each_ref(). Filter refs and keep them for
|
|
|
|
* later object processing.
|
|
|
|
*/
|
2015-06-13 22:37:28 +03:00
|
|
|
static int ref_filter_handler(const char *refname, const struct object_id *oid, int flag, void *cb_data)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
struct ref_filter_cbdata *ref_cbdata = cb_data;
|
2015-06-13 22:37:28 +03:00
|
|
|
struct ref_filter *filter = ref_cbdata->filter;
|
2015-06-13 22:37:27 +03:00
|
|
|
struct ref_array_item *ref;
|
2015-07-07 19:06:12 +03:00
|
|
|
struct commit *commit = NULL;
|
2015-09-10 18:48:23 +03:00
|
|
|
unsigned int kind;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
if (flag & REF_BAD_NAME) {
|
2016-02-27 09:42:04 +03:00
|
|
|
warning(_("ignoring ref with broken name %s"), refname);
|
2015-06-13 22:37:27 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-08-03 21:01:10 +03:00
|
|
|
if (flag & REF_ISBROKEN) {
|
2016-02-27 09:42:04 +03:00
|
|
|
warning(_("ignoring broken ref %s"), refname);
|
2015-08-03 21:01:10 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:23 +03:00
|
|
|
/* Obtain the current ref kind from filter_ref_kind() and ignore unwanted refs. */
|
|
|
|
kind = filter_ref_kind(filter, refname);
|
|
|
|
if (!(kind & filter->kind))
|
|
|
|
return 0;
|
|
|
|
|
2015-09-10 18:48:26 +03:00
|
|
|
if (!filter_pattern_match(filter, refname))
|
2015-06-13 22:37:27 +03:00
|
|
|
return 0;
|
|
|
|
|
2017-03-31 04:39:57 +03:00
|
|
|
if (filter->points_at.nr && !match_points_at(&filter->points_at, oid, refname))
|
2015-07-07 19:06:09 +03:00
|
|
|
return 0;
|
|
|
|
|
2015-07-07 19:06:12 +03:00
|
|
|
/*
|
|
|
|
* A merge filter is applied on refs pointing to commits. Hence
|
|
|
|
* obtain the commit using the 'oid' available and discard all
|
|
|
|
* non-commits early. The actual filtering is done later.
|
|
|
|
*/
|
2020-09-16 05:08:40 +03:00
|
|
|
if (filter->reachable_from || filter->unreachable_from ||
|
|
|
|
filter->with_commit || filter->no_commit || filter->verbose) {
|
|
|
|
commit = lookup_commit_reference_gently(the_repository, oid, 1);
|
2015-07-07 19:06:12 +03:00
|
|
|
if (!commit)
|
|
|
|
return 0;
|
ref-filter: add --no-contains option to tag/branch/for-each-ref
Change the tag, branch & for-each-ref commands to have a --no-contains
option in addition to their longstanding --contains options.
This allows for finding the last-good rollout tag given a known-bad
<commit>. Given a hypothetically bad commit cf5c7253e0, the git
version to revert to can be found with this hacky two-liner:
(git tag -l 'v[0-9]*'; git tag -l --contains cf5c7253e0 'v[0-9]*') |
sort | uniq -c | grep -E '^ *1 ' | awk '{print $2}' | tail -n 10
With this new --no-contains option the same can be achieved with:
git tag -l --no-contains cf5c7253e0 'v[0-9]*' | sort | tail -n 10
As the filtering machinery is shared between the tag, branch &
for-each-ref commands, implement this for those commands too. A
practical use for this with "branch" is e.g. finding branches which
were branched off between v2.8.0 and v2.10.0:
git branch --contains v2.8.0 --no-contains v2.10.0
The "describe" command also has a --contains option, but its semantics
are unrelated to what tag/branch/for-each-ref use --contains for. A
--no-contains option for "describe" wouldn't make any sense, other
than being exactly equivalent to not supplying --contains at all,
which would be confusing at best.
Add a --without option to "tag" as an alias for --no-contains, for
consistency with --with and --contains. The --with option is
undocumented, and possibly the only user of it is
Junio (<xmqqefy71iej.fsf@gitster.mtv.corp.google.com>). But it's
trivial to support, so let's do that.
The additions to the the test suite are inverse copies of the
corresponding --contains tests. With this change --no-contains for
tag, branch & for-each-ref is just as well tested as the existing
--contains option.
In addition to those tests, add a test for "tag" which asserts that
--no-contains won't find tree/blob tags, which is slightly
unintuitive, but consistent with how --contains works & is documented.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-03-24 21:40:57 +03:00
|
|
|
/* We perform the filtering for the '--contains' option... */
|
2015-07-07 19:06:16 +03:00
|
|
|
if (filter->with_commit &&
|
ref-filter: add --no-contains option to tag/branch/for-each-ref
Change the tag, branch & for-each-ref commands to have a --no-contains
option in addition to their longstanding --contains options.
This allows for finding the last-good rollout tag given a known-bad
<commit>. Given a hypothetically bad commit cf5c7253e0, the git
version to revert to can be found with this hacky two-liner:
(git tag -l 'v[0-9]*'; git tag -l --contains cf5c7253e0 'v[0-9]*') |
sort | uniq -c | grep -E '^ *1 ' | awk '{print $2}' | tail -n 10
With this new --no-contains option the same can be achieved with:
git tag -l --no-contains cf5c7253e0 'v[0-9]*' | sort | tail -n 10
As the filtering machinery is shared between the tag, branch &
for-each-ref commands, implement this for those commands too. A
practical use for this with "branch" is e.g. finding branches which
were branched off between v2.8.0 and v2.10.0:
git branch --contains v2.8.0 --no-contains v2.10.0
The "describe" command also has a --contains option, but its semantics
are unrelated to what tag/branch/for-each-ref use --contains for. A
--no-contains option for "describe" wouldn't make any sense, other
than being exactly equivalent to not supplying --contains at all,
which would be confusing at best.
Add a --without option to "tag" as an alias for --no-contains, for
consistency with --with and --contains. The --with option is
undocumented, and possibly the only user of it is
Junio (<xmqqefy71iej.fsf@gitster.mtv.corp.google.com>). But it's
trivial to support, so let's do that.
The additions to the the test suite are inverse copies of the
corresponding --contains tests. With this change --no-contains for
tag, branch & for-each-ref is just as well tested as the existing
--contains option.
In addition to those tests, add a test for "tag" which asserts that
--no-contains won't find tree/blob tags, which is slightly
unintuitive, but consistent with how --contains works & is documented.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-03-24 21:40:57 +03:00
|
|
|
!commit_contains(filter, commit, filter->with_commit, &ref_cbdata->contains_cache))
|
|
|
|
return 0;
|
|
|
|
/* ...or for the `--no-contains' option */
|
|
|
|
if (filter->no_commit &&
|
|
|
|
commit_contains(filter, commit, filter->no_commit, &ref_cbdata->no_contains_cache))
|
2015-07-07 19:06:16 +03:00
|
|
|
return 0;
|
2015-07-07 19:06:12 +03:00
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/*
|
|
|
|
* We do not open the object yet; sort may only need refname
|
|
|
|
* to do its job and the resulting list may yet to be pruned
|
|
|
|
* by maxcount logic.
|
|
|
|
*/
|
2018-04-06 21:59:45 +03:00
|
|
|
ref = ref_array_push(ref_cbdata->array, refname, oid);
|
2015-07-07 19:06:12 +03:00
|
|
|
ref->commit = commit;
|
2018-04-06 21:59:26 +03:00
|
|
|
ref->flag = flag;
|
2015-09-10 18:48:23 +03:00
|
|
|
ref->kind = kind;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Free memory allocated for a ref_array_item */
|
|
|
|
static void free_array_item(struct ref_array_item *item)
|
|
|
|
{
|
|
|
|
free((char *)item->symref);
|
2018-10-18 10:28:54 +03:00
|
|
|
if (item->value) {
|
2019-07-10 21:36:39 +03:00
|
|
|
int i;
|
|
|
|
for (i = 0; i < used_atom_cnt; i++)
|
|
|
|
free((char *)item->value[i].s);
|
2018-10-18 10:28:54 +03:00
|
|
|
free(item->value);
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
free(item);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Free all memory allocated for ref_array */
|
|
|
|
void ref_array_clear(struct ref_array *array)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < array->nr; i++)
|
|
|
|
free_array_item(array->items[i]);
|
2017-06-16 02:15:46 +03:00
|
|
|
FREE_AND_NULL(array->items);
|
2015-06-13 22:37:27 +03:00
|
|
|
array->nr = array->alloc = 0;
|
2019-07-10 21:36:39 +03:00
|
|
|
|
|
|
|
for (i = 0; i < used_atom_cnt; i++)
|
|
|
|
free((char *)used_atom[i].name);
|
|
|
|
FREE_AND_NULL(used_atom);
|
|
|
|
used_atom_cnt = 0;
|
|
|
|
|
2019-04-29 08:19:42 +03:00
|
|
|
if (ref_to_worktree_map.worktrees) {
|
2020-11-02 21:55:05 +03:00
|
|
|
hashmap_clear_and_free(&(ref_to_worktree_map.map),
|
2019-10-07 02:30:40 +03:00
|
|
|
struct ref_to_worktree_entry, ent);
|
2019-04-29 08:19:42 +03:00
|
|
|
free_worktrees(ref_to_worktree_map.worktrees);
|
|
|
|
ref_to_worktree_map.worktrees = NULL;
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
2020-09-19 00:58:41 +03:00
|
|
|
#define EXCLUDE_REACHED 0
|
|
|
|
#define INCLUDE_REACHED 1
|
|
|
|
static void reach_filter(struct ref_array *array,
|
|
|
|
struct commit_list *check_reachable,
|
|
|
|
int include_reached)
|
2015-07-07 19:06:12 +03:00
|
|
|
{
|
|
|
|
struct rev_info revs;
|
|
|
|
int i, old_nr;
|
2020-09-26 11:37:29 +03:00
|
|
|
struct commit **to_clear;
|
2020-09-19 00:58:41 +03:00
|
|
|
struct commit_list *cr;
|
2020-09-16 05:08:40 +03:00
|
|
|
|
2020-09-19 00:58:41 +03:00
|
|
|
if (!check_reachable)
|
2020-09-16 05:08:40 +03:00
|
|
|
return;
|
2015-07-07 19:06:12 +03:00
|
|
|
|
2020-09-26 11:37:29 +03:00
|
|
|
to_clear = xcalloc(sizeof(struct commit *), array->nr);
|
|
|
|
|
2018-09-21 18:57:38 +03:00
|
|
|
repo_init_revisions(the_repository, &revs, NULL);
|
2015-07-07 19:06:12 +03:00
|
|
|
|
|
|
|
for (i = 0; i < array->nr; i++) {
|
|
|
|
struct ref_array_item *item = array->items[i];
|
|
|
|
add_pending_object(&revs, &item->commit->object, item->refname);
|
|
|
|
to_clear[i] = item->commit;
|
|
|
|
}
|
|
|
|
|
2020-09-19 00:58:41 +03:00
|
|
|
for (cr = check_reachable; cr; cr = cr->next) {
|
|
|
|
struct commit *merge_commit = cr->item;
|
2020-09-16 05:08:40 +03:00
|
|
|
merge_commit->object.flags |= UNINTERESTING;
|
|
|
|
add_pending_object(&revs, &merge_commit->object, "");
|
|
|
|
}
|
2015-07-07 19:06:12 +03:00
|
|
|
|
|
|
|
revs.limited = 1;
|
|
|
|
if (prepare_revision_walk(&revs))
|
|
|
|
die(_("revision walk setup failed"));
|
|
|
|
|
|
|
|
old_nr = array->nr;
|
|
|
|
array->nr = 0;
|
|
|
|
|
|
|
|
for (i = 0; i < old_nr; i++) {
|
|
|
|
struct ref_array_item *item = array->items[i];
|
|
|
|
struct commit *commit = item->commit;
|
|
|
|
|
|
|
|
int is_merged = !!(commit->object.flags & UNINTERESTING);
|
|
|
|
|
2020-09-19 00:58:41 +03:00
|
|
|
if (is_merged == include_reached)
|
2015-07-07 19:06:12 +03:00
|
|
|
array->items[array->nr++] = array->items[i];
|
|
|
|
else
|
|
|
|
free_array_item(item);
|
|
|
|
}
|
|
|
|
|
2017-12-25 20:44:12 +03:00
|
|
|
clear_commit_marks_many(old_nr, to_clear, ALL_REV_FLAGS);
|
2020-09-16 05:08:40 +03:00
|
|
|
|
2020-09-19 00:58:41 +03:00
|
|
|
while (check_reachable) {
|
|
|
|
struct commit *merge_commit = pop_commit(&check_reachable);
|
2020-09-16 05:08:40 +03:00
|
|
|
clear_commit_marks(merge_commit, ALL_REV_FLAGS);
|
|
|
|
}
|
|
|
|
|
2015-07-07 19:06:12 +03:00
|
|
|
free(to_clear);
|
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:28 +03:00
|
|
|
/*
|
|
|
|
* API for filtering a set of refs. Based on the type of refs the user
|
|
|
|
* has requested, we iterate through those refs and apply filters
|
|
|
|
* as per the given ref_filter structure and finally store the
|
|
|
|
* filtered refs in the ref_array structure.
|
|
|
|
*/
|
|
|
|
int filter_refs(struct ref_array *array, struct ref_filter *filter, unsigned int type)
|
|
|
|
{
|
|
|
|
struct ref_filter_cbdata ref_cbdata;
|
2015-07-07 19:06:12 +03:00
|
|
|
int ret = 0;
|
2015-09-10 18:48:23 +03:00
|
|
|
unsigned int broken = 0;
|
2015-06-13 22:37:28 +03:00
|
|
|
|
|
|
|
ref_cbdata.array = array;
|
|
|
|
ref_cbdata.filter = filter;
|
|
|
|
|
2015-09-10 18:48:23 +03:00
|
|
|
if (type & FILTER_REFS_INCLUDE_BROKEN)
|
|
|
|
broken = 1;
|
|
|
|
filter->kind = type & FILTER_REFS_KIND_MASK;
|
|
|
|
|
2017-03-09 16:29:49 +03:00
|
|
|
init_contains_cache(&ref_cbdata.contains_cache);
|
ref-filter: add --no-contains option to tag/branch/for-each-ref
Change the tag, branch & for-each-ref commands to have a --no-contains
option in addition to their longstanding --contains options.
This allows for finding the last-good rollout tag given a known-bad
<commit>. Given a hypothetically bad commit cf5c7253e0, the git
version to revert to can be found with this hacky two-liner:
(git tag -l 'v[0-9]*'; git tag -l --contains cf5c7253e0 'v[0-9]*') |
sort | uniq -c | grep -E '^ *1 ' | awk '{print $2}' | tail -n 10
With this new --no-contains option the same can be achieved with:
git tag -l --no-contains cf5c7253e0 'v[0-9]*' | sort | tail -n 10
As the filtering machinery is shared between the tag, branch &
for-each-ref commands, implement this for those commands too. A
practical use for this with "branch" is e.g. finding branches which
were branched off between v2.8.0 and v2.10.0:
git branch --contains v2.8.0 --no-contains v2.10.0
The "describe" command also has a --contains option, but its semantics
are unrelated to what tag/branch/for-each-ref use --contains for. A
--no-contains option for "describe" wouldn't make any sense, other
than being exactly equivalent to not supplying --contains at all,
which would be confusing at best.
Add a --without option to "tag" as an alias for --no-contains, for
consistency with --with and --contains. The --with option is
undocumented, and possibly the only user of it is
Junio (<xmqqefy71iej.fsf@gitster.mtv.corp.google.com>). But it's
trivial to support, so let's do that.
The additions to the the test suite are inverse copies of the
corresponding --contains tests. With this change --no-contains for
tag, branch & for-each-ref is just as well tested as the existing
--contains option.
In addition to those tests, add a test for "tag" which asserts that
--no-contains won't find tree/blob tags, which is slightly
unintuitive, but consistent with how --contains works & is documented.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-03-24 21:40:57 +03:00
|
|
|
init_contains_cache(&ref_cbdata.no_contains_cache);
|
2017-03-09 16:29:49 +03:00
|
|
|
|
2015-07-07 19:06:12 +03:00
|
|
|
/* Simple per-ref filtering */
|
2015-09-10 18:48:23 +03:00
|
|
|
if (!filter->kind)
|
2015-06-13 22:37:28 +03:00
|
|
|
die("filter_refs: invalid type");
|
2015-09-10 18:48:23 +03:00
|
|
|
else {
|
|
|
|
/*
|
|
|
|
* For common cases where we need only branches or remotes or tags,
|
|
|
|
* we only iterate through those refs. If a mix of refs is needed,
|
|
|
|
* we iterate over all refs and filter out required refs with the help
|
|
|
|
* of filter_ref_kind().
|
|
|
|
*/
|
|
|
|
if (filter->kind == FILTER_REFS_BRANCHES)
|
|
|
|
ret = for_each_fullref_in("refs/heads/", ref_filter_handler, &ref_cbdata, broken);
|
|
|
|
else if (filter->kind == FILTER_REFS_REMOTES)
|
|
|
|
ret = for_each_fullref_in("refs/remotes/", ref_filter_handler, &ref_cbdata, broken);
|
|
|
|
else if (filter->kind == FILTER_REFS_TAGS)
|
|
|
|
ret = for_each_fullref_in("refs/tags/", ref_filter_handler, &ref_cbdata, broken);
|
|
|
|
else if (filter->kind & FILTER_REFS_ALL)
|
ref-filter: limit traversal to prefix
When we are matching refnames against a pattern, then we know that the
beginning of any refname that can match the pattern has to match the
part of the pattern up to the first glob character. For example, if
the pattern is `refs/heads/foo*bar`, then it can only match a
reference that has the prefix `refs/heads/foo`.
So pass that prefix to `for_each_fullref_in()`. This lets the ref code
avoid passing us the full set of refs, and in some cases avoid reading
them in the first place.
Note that this applies only when the `match_as_path` flag is set
(i.e., when `for-each-ref` is the caller), as the matching rules for
git-branch and git-tag are subtly different.
This could be generalized to the case of multiple patterns, but (a) it
probably doesn't come up that often, and (b) it is more awkward to
deal with multiple patterns (e.g., the patterns might not be
disjoint). So, since this is just an optimization, punt on the case of
multiple patterns.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-05-22 17:17:54 +03:00
|
|
|
ret = for_each_fullref_in_pattern(filter, ref_filter_handler, &ref_cbdata, broken);
|
2015-09-10 18:48:23 +03:00
|
|
|
if (!ret && (filter->kind & FILTER_REFS_DETACHED_HEAD))
|
|
|
|
head_ref(ref_filter_handler, &ref_cbdata);
|
|
|
|
}
|
|
|
|
|
2017-03-09 16:29:49 +03:00
|
|
|
clear_contains_cache(&ref_cbdata.contains_cache);
|
ref-filter: add --no-contains option to tag/branch/for-each-ref
Change the tag, branch & for-each-ref commands to have a --no-contains
option in addition to their longstanding --contains options.
This allows for finding the last-good rollout tag given a known-bad
<commit>. Given a hypothetically bad commit cf5c7253e0, the git
version to revert to can be found with this hacky two-liner:
(git tag -l 'v[0-9]*'; git tag -l --contains cf5c7253e0 'v[0-9]*') |
sort | uniq -c | grep -E '^ *1 ' | awk '{print $2}' | tail -n 10
With this new --no-contains option the same can be achieved with:
git tag -l --no-contains cf5c7253e0 'v[0-9]*' | sort | tail -n 10
As the filtering machinery is shared between the tag, branch &
for-each-ref commands, implement this for those commands too. A
practical use for this with "branch" is e.g. finding branches which
were branched off between v2.8.0 and v2.10.0:
git branch --contains v2.8.0 --no-contains v2.10.0
The "describe" command also has a --contains option, but its semantics
are unrelated to what tag/branch/for-each-ref use --contains for. A
--no-contains option for "describe" wouldn't make any sense, other
than being exactly equivalent to not supplying --contains at all,
which would be confusing at best.
Add a --without option to "tag" as an alias for --no-contains, for
consistency with --with and --contains. The --with option is
undocumented, and possibly the only user of it is
Junio (<xmqqefy71iej.fsf@gitster.mtv.corp.google.com>). But it's
trivial to support, so let's do that.
The additions to the the test suite are inverse copies of the
corresponding --contains tests. With this change --no-contains for
tag, branch & for-each-ref is just as well tested as the existing
--contains option.
In addition to those tests, add a test for "tag" which asserts that
--no-contains won't find tree/blob tags, which is slightly
unintuitive, but consistent with how --contains works & is documented.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-03-24 21:40:57 +03:00
|
|
|
clear_contains_cache(&ref_cbdata.no_contains_cache);
|
2015-07-07 19:06:12 +03:00
|
|
|
|
|
|
|
/* Filters that need revision walking */
|
2020-09-19 00:58:41 +03:00
|
|
|
reach_filter(array, filter->reachable_from, INCLUDE_REACHED);
|
|
|
|
reach_filter(array, filter->unreachable_from, EXCLUDE_REACHED);
|
2015-07-07 19:06:12 +03:00
|
|
|
|
|
|
|
return ret;
|
2015-06-13 22:37:28 +03:00
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
static int cmp_ref_sorting(struct ref_sorting *s, struct ref_array_item *a, struct ref_array_item *b)
|
|
|
|
{
|
|
|
|
struct atom_value *va, *vb;
|
|
|
|
int cmp;
|
2016-02-17 21:06:11 +03:00
|
|
|
cmp_type cmp_type = used_atom[s->atom].type;
|
2016-12-04 05:52:25 +03:00
|
|
|
int (*cmp_fn)(const char *, const char *);
|
2018-03-29 15:49:45 +03:00
|
|
|
struct strbuf err = STRBUF_INIT;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
if (get_ref_atom_value(a, s->atom, &va, &err))
|
|
|
|
die("%s", err.buf);
|
|
|
|
if (get_ref_atom_value(b, s->atom, &vb, &err))
|
|
|
|
die("%s", err.buf);
|
|
|
|
strbuf_release(&err);
|
2016-12-04 05:52:25 +03:00
|
|
|
cmp_fn = s->ignore_case ? strcasecmp : strcmp;
|
2015-09-10 18:48:25 +03:00
|
|
|
if (s->version)
|
|
|
|
cmp = versioncmp(va->s, vb->s);
|
|
|
|
else if (cmp_type == FIELD_STR)
|
2016-12-04 05:52:25 +03:00
|
|
|
cmp = cmp_fn(va->s, vb->s);
|
2015-09-10 18:48:25 +03:00
|
|
|
else {
|
2017-04-20 23:52:09 +03:00
|
|
|
if (va->value < vb->value)
|
2015-06-13 22:37:27 +03:00
|
|
|
cmp = -1;
|
2017-04-20 23:52:09 +03:00
|
|
|
else if (va->value == vb->value)
|
ref-filter: apply fallback refname sort only after all user sorts
Commit 9e468334b4 (ref-filter: fallback on alphabetical comparison,
2015-10-30) taught ref-filter's sort to fallback to comparing refnames.
But it did it at the wrong level, overriding the comparison result for a
single "--sort" key from the user, rather than after all sort keys have
been exhausted.
This worked correctly for a single "--sort" option, but not for multiple
ones. We'd break any ties in the first key with the refname and never
evaluate the second key at all.
To make matters even more interesting, we only applied this fallback
sometimes! For a field like "taggeremail" which requires a string
comparison, we'd truly return the result of strcmp(), even if it was 0.
But for numerical "value" fields like "taggerdate", we did apply the
fallback. And that's why our multiple-sort test missed this: it uses
taggeremail as the main comparison.
So let's start by adding a much more rigorous test. We'll have a set of
commits expressing every combination of two tagger emails, dates, and
refnames. Then we can confirm that our sort is applied with the correct
precedence, and we'll be hitting both the string and value comparators.
That does show the bug, and the fix is simple: moving the fallback to
the outer compare_refs() function, after all ref_sorting keys have been
exhausted.
Note that in the outer function we don't have an "ignore_case" flag, as
it's part of each individual ref_sorting element. It's debatable what
such a fallback should do, since we didn't use the user's keys to match.
But until now we have been trying to respect that flag, so the
least-invasive thing is to try to continue to do so. Since all callers
in the current code either set the flag for all keys or for none, we can
just pull the flag from the first key. In a hypothetical world where the
user really can flip the case-insensitivity of keys separately, we may
want to extend the code to distinguish that case from a blanket
"--ignore-case".
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-05-03 12:13:09 +03:00
|
|
|
cmp = 0;
|
2015-06-13 22:37:27 +03:00
|
|
|
else
|
|
|
|
cmp = 1;
|
|
|
|
}
|
2015-09-10 18:48:25 +03:00
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
return (s->reverse) ? -cmp : cmp;
|
|
|
|
}
|
|
|
|
|
2017-01-22 20:58:07 +03:00
|
|
|
static int compare_refs(const void *a_, const void *b_, void *ref_sorting)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
struct ref_array_item *a = *((struct ref_array_item **)a_);
|
|
|
|
struct ref_array_item *b = *((struct ref_array_item **)b_);
|
|
|
|
struct ref_sorting *s;
|
|
|
|
|
|
|
|
for (s = ref_sorting; s; s = s->next) {
|
|
|
|
int cmp = cmp_ref_sorting(s, a, b);
|
|
|
|
if (cmp)
|
|
|
|
return cmp;
|
|
|
|
}
|
ref-filter: apply fallback refname sort only after all user sorts
Commit 9e468334b4 (ref-filter: fallback on alphabetical comparison,
2015-10-30) taught ref-filter's sort to fallback to comparing refnames.
But it did it at the wrong level, overriding the comparison result for a
single "--sort" key from the user, rather than after all sort keys have
been exhausted.
This worked correctly for a single "--sort" option, but not for multiple
ones. We'd break any ties in the first key with the refname and never
evaluate the second key at all.
To make matters even more interesting, we only applied this fallback
sometimes! For a field like "taggeremail" which requires a string
comparison, we'd truly return the result of strcmp(), even if it was 0.
But for numerical "value" fields like "taggerdate", we did apply the
fallback. And that's why our multiple-sort test missed this: it uses
taggeremail as the main comparison.
So let's start by adding a much more rigorous test. We'll have a set of
commits expressing every combination of two tagger emails, dates, and
refnames. Then we can confirm that our sort is applied with the correct
precedence, and we'll be hitting both the string and value comparators.
That does show the bug, and the fix is simple: moving the fallback to
the outer compare_refs() function, after all ref_sorting keys have been
exhausted.
Note that in the outer function we don't have an "ignore_case" flag, as
it's part of each individual ref_sorting element. It's debatable what
such a fallback should do, since we didn't use the user's keys to match.
But until now we have been trying to respect that flag, so the
least-invasive thing is to try to continue to do so. Since all callers
in the current code either set the flag for all keys or for none, we can
just pull the flag from the first key. In a hypothetical world where the
user really can flip the case-insensitivity of keys separately, we may
want to extend the code to distinguish that case from a blanket
"--ignore-case".
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-05-03 12:13:09 +03:00
|
|
|
s = ref_sorting;
|
|
|
|
return s && s->ignore_case ?
|
|
|
|
strcasecmp(a->refname, b->refname) :
|
|
|
|
strcmp(a->refname, b->refname);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
ref-filter: apply --ignore-case to all sorting keys
All of the ref-filter users (for-each-ref, branch, and tag) take an
--ignore-case option which makes filtering and sorting case-insensitive.
However, this option was applied only to the first element of the
ref_sorting list. So:
git for-each-ref --ignore-case --sort=refname
would do what you expect, but:
git for-each-ref --ignore-case --sort=refname --sort=taggername
would sort the primary key (taggername) case-insensitively, but sort the
refname case-sensitively. We have two options here:
- teach callers to set ignore_case on the whole list
- replace the ref_sorting list with a struct that contains both the
list of sorting keys, as well as options that apply to _all_
keys
I went with the first one here, as it gives more flexibility if we later
want to let the users set the flag per-key (presumably through some
special syntax when defining the key; for now it's all or nothing
through --ignore-case).
The new test covers this by sorting on both tagger and subject
case-insensitively, which should compare "a" and "A" identically, but
still sort them before "b" and "B". We'll break ties by sorting on the
refname to give ourselves a stable output (this is actually supposed to
be done automatically, but there's another bug which will be fixed in
the next commit).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-05-03 12:11:57 +03:00
|
|
|
void ref_sorting_icase_all(struct ref_sorting *sorting, int flag)
|
|
|
|
{
|
|
|
|
for (; sorting; sorting = sorting->next)
|
|
|
|
sorting->ignore_case = !!flag;
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
void ref_array_sort(struct ref_sorting *sorting, struct ref_array *array)
|
|
|
|
{
|
2017-01-22 20:58:07 +03:00
|
|
|
QSORT_S(array->items, array->nr, compare_refs, sorting);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
|
2015-09-10 18:48:18 +03:00
|
|
|
static void append_literal(const char *cp, const char *ep, struct ref_formatting_state *state)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
2015-09-10 18:48:18 +03:00
|
|
|
struct strbuf *s = &state->stack->output;
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
while (*cp && (!ep || cp < ep)) {
|
|
|
|
if (*cp == '%') {
|
|
|
|
if (cp[1] == '%')
|
|
|
|
cp++;
|
|
|
|
else {
|
2016-09-03 18:59:20 +03:00
|
|
|
int ch = hex2chr(cp + 1);
|
2015-06-13 22:37:27 +03:00
|
|
|
if (0 <= ch) {
|
2015-09-10 18:48:18 +03:00
|
|
|
strbuf_addch(s, ch);
|
2015-06-13 22:37:27 +03:00
|
|
|
cp += 3;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2015-09-10 18:48:18 +03:00
|
|
|
strbuf_addch(s, *cp);
|
2015-06-13 22:37:27 +03:00
|
|
|
cp++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
int format_ref_array_item(struct ref_array_item *info,
|
2017-07-13 18:01:18 +03:00
|
|
|
const struct ref_format *format,
|
2018-03-29 15:49:45 +03:00
|
|
|
struct strbuf *final_buf,
|
|
|
|
struct strbuf *error_buf)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
const char *cp, *sp, *ep;
|
2015-09-10 18:48:18 +03:00
|
|
|
struct ref_formatting_state state = REF_FORMATTING_STATE_INIT;
|
|
|
|
|
2017-07-13 18:01:18 +03:00
|
|
|
state.quote_style = format->quote_style;
|
2015-09-10 18:48:18 +03:00
|
|
|
push_stack_element(&state.stack);
|
2015-06-13 22:37:27 +03:00
|
|
|
|
2017-07-13 18:01:18 +03:00
|
|
|
for (cp = format->format; *cp && (sp = find_next(cp)); cp = ep + 1) {
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value *atomv;
|
2018-03-29 15:49:45 +03:00
|
|
|
int pos;
|
2015-06-13 22:37:27 +03:00
|
|
|
|
|
|
|
ep = strchr(sp, ')');
|
|
|
|
if (cp < sp)
|
2015-09-10 18:48:18 +03:00
|
|
|
append_literal(cp, sp, &state);
|
2018-03-29 15:49:45 +03:00
|
|
|
pos = parse_ref_filter_atom(format, sp + 2, ep, error_buf);
|
2018-03-29 15:49:45 +03:00
|
|
|
if (pos < 0 || get_ref_atom_value(info, pos, &atomv, error_buf) ||
|
|
|
|
atomv->handler(atomv, &state, error_buf)) {
|
2018-03-29 15:49:45 +03:00
|
|
|
pop_stack_element(&state.stack);
|
|
|
|
return -1;
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
|
|
|
if (*cp) {
|
|
|
|
sp = cp + strlen(cp);
|
2015-09-10 18:48:18 +03:00
|
|
|
append_literal(cp, sp, &state);
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2017-07-13 18:02:30 +03:00
|
|
|
if (format->need_color_reset_at_eol) {
|
2015-06-13 22:37:27 +03:00
|
|
|
struct atom_value resetv;
|
2017-07-13 17:58:56 +03:00
|
|
|
resetv.s = GIT_COLOR_RESET;
|
2018-03-29 15:49:45 +03:00
|
|
|
if (append_atom(&resetv, &state, error_buf)) {
|
|
|
|
pop_stack_element(&state.stack);
|
|
|
|
return -1;
|
|
|
|
}
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2018-03-29 15:49:45 +03:00
|
|
|
if (state.stack->prev) {
|
|
|
|
pop_stack_element(&state.stack);
|
|
|
|
return strbuf_addf_ret(error_buf, -1, _("format: %%(end) atom missing"));
|
2015-06-13 22:37:27 +03:00
|
|
|
}
|
2017-01-10 11:49:39 +03:00
|
|
|
strbuf_addbuf(final_buf, &state.stack->output);
|
2015-09-10 18:48:18 +03:00
|
|
|
pop_stack_element(&state.stack);
|
2018-03-29 15:49:45 +03:00
|
|
|
return 0;
|
2017-01-10 11:49:39 +03:00
|
|
|
}
|
|
|
|
|
2017-07-13 18:01:18 +03:00
|
|
|
void show_ref_array_item(struct ref_array_item *info,
|
|
|
|
const struct ref_format *format)
|
2017-01-10 11:49:39 +03:00
|
|
|
{
|
|
|
|
struct strbuf final_buf = STRBUF_INIT;
|
2018-03-29 15:49:45 +03:00
|
|
|
struct strbuf error_buf = STRBUF_INIT;
|
2017-01-10 11:49:39 +03:00
|
|
|
|
2018-03-29 15:49:45 +03:00
|
|
|
if (format_ref_array_item(info, format, &final_buf, &error_buf))
|
|
|
|
die("%s", error_buf.buf);
|
2017-01-10 11:49:39 +03:00
|
|
|
fwrite(final_buf.buf, 1, final_buf.len, stdout);
|
2018-03-29 15:49:45 +03:00
|
|
|
strbuf_release(&error_buf);
|
2017-01-10 11:49:39 +03:00
|
|
|
strbuf_release(&final_buf);
|
2015-06-13 22:37:27 +03:00
|
|
|
putchar('\n');
|
|
|
|
}
|
|
|
|
|
2018-04-06 21:58:32 +03:00
|
|
|
void pretty_print_ref(const char *name, const struct object_id *oid,
|
2017-07-13 18:01:18 +03:00
|
|
|
const struct ref_format *format)
|
2017-01-18 02:37:19 +03:00
|
|
|
{
|
|
|
|
struct ref_array_item *ref_item;
|
2018-04-06 21:59:26 +03:00
|
|
|
ref_item = new_ref_array_item(name, oid);
|
2017-01-18 02:37:19 +03:00
|
|
|
ref_item->kind = ref_kind_from_refname(name);
|
2017-07-13 18:01:18 +03:00
|
|
|
show_ref_array_item(ref_item, format);
|
2017-01-18 02:37:19 +03:00
|
|
|
free_array_item(ref_item);
|
|
|
|
}
|
|
|
|
|
2017-07-13 18:02:58 +03:00
|
|
|
static int parse_sorting_atom(const char *atom)
|
|
|
|
{
|
2017-07-13 18:06:40 +03:00
|
|
|
/*
|
|
|
|
* This parses an atom using a dummy ref_format, since we don't
|
|
|
|
* actually care about the formatting details.
|
|
|
|
*/
|
|
|
|
struct ref_format dummy = REF_FORMAT_INIT;
|
2017-07-13 18:02:58 +03:00
|
|
|
const char *end = atom + strlen(atom);
|
2018-03-29 15:49:45 +03:00
|
|
|
struct strbuf err = STRBUF_INIT;
|
|
|
|
int res = parse_ref_filter_atom(&dummy, atom, end, &err);
|
|
|
|
if (res < 0)
|
|
|
|
die("%s", err.buf);
|
|
|
|
strbuf_release(&err);
|
|
|
|
return res;
|
2017-07-13 18:02:58 +03:00
|
|
|
}
|
|
|
|
|
2015-06-13 22:37:27 +03:00
|
|
|
/* If no sorting option is given, use refname to sort as default */
|
|
|
|
struct ref_sorting *ref_default_sorting(void)
|
|
|
|
{
|
|
|
|
static const char cstr_name[] = "refname";
|
|
|
|
|
|
|
|
struct ref_sorting *sorting = xcalloc(1, sizeof(*sorting));
|
|
|
|
|
|
|
|
sorting->next = NULL;
|
2017-07-13 18:02:58 +03:00
|
|
|
sorting->atom = parse_sorting_atom(cstr_name);
|
2015-06-13 22:37:27 +03:00
|
|
|
return sorting;
|
|
|
|
}
|
|
|
|
|
2017-07-13 18:02:44 +03:00
|
|
|
void parse_ref_sorting(struct ref_sorting **sorting_tail, const char *arg)
|
2015-06-13 22:37:27 +03:00
|
|
|
{
|
|
|
|
struct ref_sorting *s;
|
|
|
|
|
|
|
|
s = xcalloc(1, sizeof(*s));
|
|
|
|
s->next = *sorting_tail;
|
|
|
|
*sorting_tail = s;
|
|
|
|
|
|
|
|
if (*arg == '-') {
|
|
|
|
s->reverse = 1;
|
|
|
|
arg++;
|
|
|
|
}
|
2015-09-10 18:48:25 +03:00
|
|
|
if (skip_prefix(arg, "version:", &arg) ||
|
|
|
|
skip_prefix(arg, "v:", &arg))
|
|
|
|
s->version = 1;
|
2017-07-13 18:02:58 +03:00
|
|
|
s->atom = parse_sorting_atom(arg);
|
2017-07-13 18:02:44 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
int parse_opt_ref_sorting(const struct option *opt, const char *arg, int unset)
|
|
|
|
{
|
parse_opt_ref_sorting: always use with NONEG flag
The "--sort" parameter of for-each-ref, etc, does not handle negation,
and instead returns an error to the parse-options code. But neither
piece of code prints anything for the user, which may leave them
confused:
$ git for-each-ref --no-sort
$ echo $?
129
As the comment in the callback function notes, this probably should
clear the list, which would make it consistent with other list-like
options (i.e., anything that uses OPT_STRING_LIST currently).
Unfortunately that's a bit tricky due to the way the ref-filter code
works. But in the meantime, let's at least make the error a little less
confusing:
- switch to using PARSE_OPT_NONEG in the option definition, which will
cause the options code to produce a useful message
- since this was cut-and-pasted to four different spots, let's define
a single OPT_REF_SORT() macro that we can use everywhere
- the callback can use BUG_ON_OPT_NEG() to make sure the correct flags
are used (incidentally, this also satisfies -Wunused-parameters,
since we're now looking at "unset")
- expand the comment into a NEEDSWORK to make it clear that the
direction is right, but the details need to be worked out
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-03-20 23:22:15 +03:00
|
|
|
/*
|
|
|
|
* NEEDSWORK: We should probably clear the list in this case, but we've
|
|
|
|
* already munged the global used_atoms list, which would need to be
|
|
|
|
* undone.
|
|
|
|
*/
|
|
|
|
BUG_ON_OPT_NEG(unset);
|
|
|
|
|
2017-07-13 18:02:44 +03:00
|
|
|
parse_ref_sorting(opt->value, arg);
|
2015-06-13 22:37:27 +03:00
|
|
|
return 0;
|
|
|
|
}
|
2015-07-07 19:06:11 +03:00
|
|
|
|
|
|
|
int parse_opt_merge_filter(const struct option *opt, const char *arg, int unset)
|
|
|
|
{
|
|
|
|
struct ref_filter *rf = opt->value;
|
2017-05-07 01:10:09 +03:00
|
|
|
struct object_id oid;
|
2020-09-16 05:08:40 +03:00
|
|
|
struct commit *merge_commit;
|
2015-07-07 19:06:11 +03:00
|
|
|
|
assert NOARG/NONEG behavior of parse-options callbacks
When we define a parse-options callback, the flags we put in the option
struct must match what the callback expects. For example, a callback
which does not handle the "unset" parameter should only be used with
PARSE_OPT_NONEG. But since the callback and the option struct are not
defined next to each other, it's easy to get this wrong (as earlier
patches in this series show).
Fortunately, the compiler can help us here: compiling with
-Wunused-parameters can show us which callbacks ignore their "unset"
parameters (and likewise, ones that ignore "arg" expect to be triggered
with PARSE_OPT_NOARG).
But after we've inspected a callback and determined that all of its
callers use the right flags, what do we do next? We'd like to silence
the compiler warning, but do so in a way that will catch any wrong calls
in the future.
We can do that by actually checking those variables and asserting that
they match our expectations. Because this is such a common pattern,
we'll introduce some helper macros. The resulting messages aren't
as descriptive as we could make them, but the file/line information from
BUG() is enough to identify the problem (and anyway, the point is that
these should never be seen).
Each of the annotated callbacks in this patch triggers
-Wunused-parameters, and was manually inspected to make sure all callers
use the correct options (so none of these BUGs should be triggerable).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-05 09:45:42 +03:00
|
|
|
BUG_ON_OPT_NEG(unset);
|
|
|
|
|
2017-05-07 01:10:09 +03:00
|
|
|
if (get_oid(arg, &oid))
|
2015-07-07 19:06:11 +03:00
|
|
|
die(_("malformed object name %s"), arg);
|
|
|
|
|
2020-09-16 05:08:40 +03:00
|
|
|
merge_commit = lookup_commit_reference_gently(the_repository, &oid, 0);
|
|
|
|
|
|
|
|
if (!merge_commit)
|
2018-11-10 08:16:11 +03:00
|
|
|
return error(_("option `%s' must point to a commit"), opt->long_name);
|
2015-07-07 19:06:11 +03:00
|
|
|
|
2020-09-16 05:08:40 +03:00
|
|
|
if (starts_with(opt->long_name, "no"))
|
|
|
|
commit_list_insert(merge_commit, &rf->unreachable_from);
|
|
|
|
else
|
|
|
|
commit_list_insert(merge_commit, &rf->reachable_from);
|
|
|
|
|
2015-07-07 19:06:11 +03:00
|
|
|
return 0;
|
|
|
|
}
|