2006-01-07 12:33:54 +03:00
|
|
|
#include "cache.h"
|
2007-08-10 00:42:50 +04:00
|
|
|
#include "cache-tree.h"
|
2005-04-18 22:39:48 +04:00
|
|
|
#include "tree.h"
|
|
|
|
#include "blob.h"
|
2005-09-05 10:03:51 +04:00
|
|
|
#include "commit.h"
|
|
|
|
#include "tag.h"
|
2006-05-29 23:16:12 +04:00
|
|
|
#include "tree-walk.h"
|
2005-04-18 22:39:48 +04:00
|
|
|
|
|
|
|
const char *tree_type = "tree";
|
|
|
|
|
2007-08-10 00:42:50 +04:00
|
|
|
static int read_one_entry_opt(const unsigned char *sha1, const char *base, int baselen, const char *pathname, unsigned mode, int stage, int opt)
|
2005-04-23 03:42:37 +04:00
|
|
|
{
|
2005-11-26 20:38:20 +03:00
|
|
|
int len;
|
|
|
|
unsigned int size;
|
|
|
|
struct cache_entry *ce;
|
|
|
|
|
|
|
|
if (S_ISDIR(mode))
|
|
|
|
return READ_TREE_RECURSIVE;
|
|
|
|
|
|
|
|
len = strlen(pathname);
|
|
|
|
size = cache_entry_size(baselen + len);
|
2006-04-03 22:30:46 +04:00
|
|
|
ce = xcalloc(1, size);
|
2005-04-23 03:42:37 +04:00
|
|
|
|
|
|
|
ce->ce_mode = create_ce_mode(mode);
|
2012-07-11 13:22:37 +04:00
|
|
|
ce->ce_flags = create_ce_flags(stage);
|
|
|
|
ce->ce_namelen = baselen + len;
|
2005-04-23 03:42:37 +04:00
|
|
|
memcpy(ce->name, base, baselen);
|
|
|
|
memcpy(ce->name + baselen, pathname, len+1);
|
2006-08-23 10:49:00 +04:00
|
|
|
hashcpy(ce->sha1, sha1);
|
2007-08-10 00:42:50 +04:00
|
|
|
return add_cache_entry(ce, opt);
|
|
|
|
}
|
|
|
|
|
2008-07-14 23:22:12 +04:00
|
|
|
static int read_one_entry(const unsigned char *sha1, const char *base, int baselen, const char *pathname, unsigned mode, int stage, void *context)
|
2007-08-10 00:42:50 +04:00
|
|
|
{
|
|
|
|
return read_one_entry_opt(sha1, base, baselen, pathname, mode, stage,
|
|
|
|
ADD_CACHE_OK_TO_ADD|ADD_CACHE_SKIP_DFCHECK);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This is used when the caller knows there is no existing entries at
|
|
|
|
* the stage that will conflict with the entry being added.
|
|
|
|
*/
|
2008-07-14 23:22:12 +04:00
|
|
|
static int read_one_entry_quick(const unsigned char *sha1, const char *base, int baselen, const char *pathname, unsigned mode, int stage, void *context)
|
2007-08-10 00:42:50 +04:00
|
|
|
{
|
|
|
|
return read_one_entry_opt(sha1, base, baselen, pathname, mode, stage,
|
|
|
|
ADD_CACHE_JUST_APPEND);
|
2005-04-23 03:42:37 +04:00
|
|
|
}
|
|
|
|
|
2011-03-25 12:34:18 +03:00
|
|
|
static int read_tree_1(struct tree *tree, struct strbuf *base,
|
|
|
|
int stage, struct pathspec *pathspec,
|
|
|
|
read_tree_fn_t fn, void *context)
|
2005-04-23 03:42:37 +04:00
|
|
|
{
|
2006-05-29 23:17:28 +04:00
|
|
|
struct tree_desc desc;
|
tree_entry(): new tree-walking helper function
This adds a "tree_entry()" function that combines the common operation of
doing a "tree_entry_extract()" + "update_tree_entry()".
It also has a simplified calling convention, designed for simple loops
that traverse over a whole tree: the arguments are pointers to the tree
descriptor and a name_entry structure to fill in, and it returns a boolean
"true" if there was an entry left to be gotten in the tree.
This allows tree traversal with
struct tree_desc desc;
struct name_entry entry;
desc.buf = tree->buffer;
desc.size = tree->size;
while (tree_entry(&desc, &entry) {
... use "entry.{path, sha1, mode, pathlen}" ...
}
which is not only shorter than writing it out in full, it's hopefully less
error prone too.
[ It's actually a tad faster too - we don't need to recalculate the entry
pathlength in both extract and update, but need to do it only once.
Also, some callers can avoid doing a "strlen()" on the result, since
it's returned as part of the name_entry structure.
However, by now we're talking just 1% speedup on "git-rev-list --objects
--all", and we're definitely at the point where tree walking is no
longer the issue any more. ]
NOTE! Not everybody wants to use this new helper function, since some of
the tree walkers very much on purpose do the descriptor update separately
from the entry extraction. So the "extract + update" sequence still
remains as the core sequence, this is just a simplified interface.
We should probably add a silly two-line inline helper function for
initializing the descriptor from the "struct tree" too, just to cut down
on the noise from that common "desc" initializer.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-05-30 20:45:45 +04:00
|
|
|
struct name_entry entry;
|
2011-03-25 12:34:18 +03:00
|
|
|
unsigned char sha1[20];
|
2011-10-24 10:36:10 +04:00
|
|
|
int len, oldlen = base->len;
|
|
|
|
enum interesting retval = entry_not_interesting;
|
2006-05-29 23:17:28 +04:00
|
|
|
|
2006-01-26 09:13:36 +03:00
|
|
|
if (parse_tree(tree))
|
|
|
|
return -1;
|
2006-05-29 23:17:28 +04:00
|
|
|
|
2007-03-21 20:08:25 +03:00
|
|
|
init_tree_desc(&desc, tree->buffer, tree->size);
|
2006-05-29 23:17:28 +04:00
|
|
|
|
tree_entry(): new tree-walking helper function
This adds a "tree_entry()" function that combines the common operation of
doing a "tree_entry_extract()" + "update_tree_entry()".
It also has a simplified calling convention, designed for simple loops
that traverse over a whole tree: the arguments are pointers to the tree
descriptor and a name_entry structure to fill in, and it returns a boolean
"true" if there was an entry left to be gotten in the tree.
This allows tree traversal with
struct tree_desc desc;
struct name_entry entry;
desc.buf = tree->buffer;
desc.size = tree->size;
while (tree_entry(&desc, &entry) {
... use "entry.{path, sha1, mode, pathlen}" ...
}
which is not only shorter than writing it out in full, it's hopefully less
error prone too.
[ It's actually a tad faster too - we don't need to recalculate the entry
pathlength in both extract and update, but need to do it only once.
Also, some callers can avoid doing a "strlen()" on the result, since
it's returned as part of the name_entry structure.
However, by now we're talking just 1% speedup on "git-rev-list --objects
--all", and we're definitely at the point where tree walking is no
longer the issue any more. ]
NOTE! Not everybody wants to use this new helper function, since some of
the tree walkers very much on purpose do the descriptor update separately
from the entry extraction. So the "extract + update" sequence still
remains as the core sequence, this is just a simplified interface.
We should probably add a silly two-line inline helper function for
initializing the descriptor from the "struct tree" too, just to cut down
on the noise from that common "desc" initializer.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-05-30 20:45:45 +04:00
|
|
|
while (tree_entry(&desc, &entry)) {
|
2011-10-24 10:36:10 +04:00
|
|
|
if (retval != all_entries_interesting) {
|
2011-03-25 12:34:18 +03:00
|
|
|
retval = tree_entry_interesting(&entry, base, 0, pathspec);
|
2011-10-24 10:36:10 +04:00
|
|
|
if (retval == all_entries_not_interesting)
|
2011-03-25 12:34:18 +03:00
|
|
|
break;
|
2011-10-24 10:36:10 +04:00
|
|
|
if (retval == entry_not_interesting)
|
2011-03-25 12:34:18 +03:00
|
|
|
continue;
|
|
|
|
}
|
2005-07-14 22:26:31 +04:00
|
|
|
|
2011-03-25 12:34:18 +03:00
|
|
|
switch (fn(entry.sha1, base->buf, base->len,
|
|
|
|
entry.path, entry.mode, stage, context)) {
|
2005-11-26 20:38:20 +03:00
|
|
|
case 0:
|
|
|
|
continue;
|
|
|
|
case READ_TREE_RECURSIVE:
|
2009-02-11 04:42:04 +03:00
|
|
|
break;
|
2005-11-26 20:38:20 +03:00
|
|
|
default:
|
|
|
|
return -1;
|
|
|
|
}
|
2009-01-25 03:52:05 +03:00
|
|
|
|
2011-03-25 12:34:18 +03:00
|
|
|
if (S_ISDIR(entry.mode))
|
|
|
|
hashcpy(sha1, entry.sha1);
|
|
|
|
else if (S_ISGITLINK(entry.mode)) {
|
|
|
|
struct commit *commit;
|
2009-01-25 03:52:05 +03:00
|
|
|
|
|
|
|
commit = lookup_commit(entry.sha1);
|
|
|
|
if (!commit)
|
2011-03-25 12:34:18 +03:00
|
|
|
die("Commit %s in submodule path %s%s not found",
|
|
|
|
sha1_to_hex(entry.sha1),
|
|
|
|
base->buf, entry.path);
|
2009-01-25 03:52:05 +03:00
|
|
|
|
|
|
|
if (parse_commit(commit))
|
2011-03-25 12:34:18 +03:00
|
|
|
die("Invalid commit %s in submodule path %s%s",
|
|
|
|
sha1_to_hex(entry.sha1),
|
|
|
|
base->buf, entry.path);
|
|
|
|
|
|
|
|
hashcpy(sha1, commit->tree->object.sha1);
|
2005-04-23 03:42:37 +04:00
|
|
|
}
|
2011-03-25 12:34:18 +03:00
|
|
|
else
|
|
|
|
continue;
|
|
|
|
|
2011-10-24 10:36:09 +04:00
|
|
|
len = tree_entry_len(&entry);
|
2011-03-25 12:34:18 +03:00
|
|
|
strbuf_add(base, entry.path, len);
|
|
|
|
strbuf_addch(base, '/');
|
|
|
|
retval = read_tree_1(lookup_tree(sha1),
|
|
|
|
base, stage, pathspec,
|
|
|
|
fn, context);
|
|
|
|
strbuf_setlen(base, oldlen);
|
|
|
|
if (retval)
|
|
|
|
return -1;
|
2005-04-23 03:42:37 +04:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-03-25 12:34:18 +03:00
|
|
|
int read_tree_recursive(struct tree *tree,
|
|
|
|
const char *base, int baselen,
|
2011-03-25 12:34:19 +03:00
|
|
|
int stage, struct pathspec *pathspec,
|
2011-03-25 12:34:18 +03:00
|
|
|
read_tree_fn_t fn, void *context)
|
|
|
|
{
|
|
|
|
struct strbuf sb = STRBUF_INIT;
|
2011-03-25 12:34:19 +03:00
|
|
|
int ret;
|
2011-03-25 12:34:18 +03:00
|
|
|
|
|
|
|
strbuf_add(&sb, base, baselen);
|
2011-03-25 12:34:19 +03:00
|
|
|
ret = read_tree_1(tree, &sb, stage, pathspec, fn, context);
|
2011-03-25 12:34:18 +03:00
|
|
|
strbuf_release(&sb);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2007-08-10 00:42:50 +04:00
|
|
|
static int cmp_cache_name_compare(const void *a_, const void *b_)
|
|
|
|
{
|
|
|
|
const struct cache_entry *ce1, *ce2;
|
|
|
|
|
|
|
|
ce1 = *((const struct cache_entry **)a_);
|
|
|
|
ce2 = *((const struct cache_entry **)b_);
|
2012-07-11 13:22:37 +04:00
|
|
|
return cache_name_stage_compare(ce1->name, ce1->ce_namelen, ce_stage(ce1),
|
|
|
|
ce2->name, ce2->ce_namelen, ce_stage(ce2));
|
2007-08-10 00:42:50 +04:00
|
|
|
}
|
|
|
|
|
2011-03-25 12:34:19 +03:00
|
|
|
int read_tree(struct tree *tree, int stage, struct pathspec *match)
|
2005-04-23 03:42:37 +04:00
|
|
|
{
|
2007-08-10 00:42:50 +04:00
|
|
|
read_tree_fn_t fn = NULL;
|
|
|
|
int i, err;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Currently the only existing callers of this function all
|
|
|
|
* call it with stage=1 and after making sure there is nothing
|
|
|
|
* at that stage; we could always use read_one_entry_quick().
|
|
|
|
*
|
|
|
|
* But when we decide to straighten out git-read-tree not to
|
|
|
|
* use unpack_trees() in some cases, this will probably start
|
|
|
|
* to matter.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* See if we have cache entry at the stage. If so,
|
|
|
|
* do it the original slow way, otherwise, append and then
|
|
|
|
* sort at the end.
|
|
|
|
*/
|
|
|
|
for (i = 0; !fn && i < active_nr; i++) {
|
Convert "struct cache_entry *" to "const ..." wherever possible
I attempted to make index_state->cache[] a "const struct cache_entry **"
to find out how existing entries in index are modified and where. The
question I have is what do we do if we really need to keep track of on-disk
changes in the index. The result is
- diff-lib.c: setting CE_UPTODATE
- name-hash.c: setting CE_HASHED
- preload-index.c, read-cache.c, unpack-trees.c and
builtin/update-index: obvious
- entry.c: write_entry() may refresh the checked out entry via
fill_stat_cache_info(). This causes "non-const struct cache_entry
*" in builtin/apply.c, builtin/checkout-index.c and
builtin/checkout.c
- builtin/ls-files.c: --with-tree changes stagemask and may set
CE_UPDATE
Of these, write_entry() and its call sites are probably most
interesting because it modifies on-disk info. But this is stat info
and can be retrieved via refresh, at least for porcelain
commands. Other just uses ce_flags for local purposes.
So, keeping track of "dirty" entries is just a matter of setting a
flag in index modification functions exposed by read-cache.c. Except
unpack-trees, the rest of the code base does not do anything funny
behind read-cache's back.
The actual patch is less valueable than the summary above. But if
anyone wants to re-identify the above sites. Applying this patch, then
this:
diff --git a/cache.h b/cache.h
index 430d021..1692891 100644
--- a/cache.h
+++ b/cache.h
@@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode)
#define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1)
struct index_state {
- struct cache_entry **cache;
+ const struct cache_entry **cache;
unsigned int version;
unsigned int cache_nr, cache_alloc, cache_changed;
struct string_list *resolve_undo;
will help quickly identify them without bogus warnings.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2013-07-09 19:29:00 +04:00
|
|
|
const struct cache_entry *ce = active_cache[i];
|
2007-08-10 00:42:50 +04:00
|
|
|
if (ce_stage(ce) == stage)
|
|
|
|
fn = read_one_entry;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!fn)
|
|
|
|
fn = read_one_entry_quick;
|
2008-07-14 23:22:12 +04:00
|
|
|
err = read_tree_recursive(tree, "", 0, stage, match, fn, NULL);
|
2007-08-10 00:42:50 +04:00
|
|
|
if (fn == read_one_entry || err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sort the cache entry -- we need to nuke the cache tree, though.
|
|
|
|
*/
|
|
|
|
cache_tree_free(&active_cache_tree);
|
|
|
|
qsort(active_cache, active_nr, sizeof(active_cache[0]),
|
|
|
|
cmp_cache_name_compare);
|
|
|
|
return 0;
|
2005-04-23 03:42:37 +04:00
|
|
|
}
|
|
|
|
|
2005-06-03 19:05:39 +04:00
|
|
|
struct tree *lookup_tree(const unsigned char *sha1)
|
2005-04-18 22:39:48 +04:00
|
|
|
{
|
|
|
|
struct object *obj = lookup_object(sha1);
|
2007-04-17 09:11:43 +04:00
|
|
|
if (!obj)
|
|
|
|
return create_object(sha1, OBJ_TREE, alloc_tree_node());
|
2005-05-21 00:59:17 +04:00
|
|
|
if (!obj->type)
|
2006-07-12 07:45:31 +04:00
|
|
|
obj->type = OBJ_TREE;
|
|
|
|
if (obj->type != OBJ_TREE) {
|
Shrink "struct object" a bit
This shrinks "struct object" by a small amount, by getting rid of the
"struct type *" pointer and replacing it with a 3-bit bitfield instead.
In addition, we merge the bitfields and the "flags" field, which
incidentally should also remove a useless 4-byte padding from the object
when in 64-bit mode.
Now, our "struct object" is still too damn large, but it's now less
obviously bloated, and of the remaining fields, only the "util" (which is
not used by most things) is clearly something that should be eventually
discarded.
This shrinks the "git-rev-list --all" memory use by about 2.5% on the
kernel archive (and, perhaps more importantly, on the larger mozilla
archive). That may not sound like much, but I suspect it's more on a
64-bit platform.
There are other remaining inefficiencies (the parent lists, for example,
probably have horrible malloc overhead), but this was pretty obvious.
Most of the patch is just changing the comparison of the "type" pointer
from one of the constant string pointers to the appropriate new TYPE_xxx
small integer constant.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-15 03:45:13 +04:00
|
|
|
error("Object %s is a %s, not a tree",
|
|
|
|
sha1_to_hex(sha1), typename(obj->type));
|
2005-04-18 22:39:48 +04:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return (struct tree *) obj;
|
|
|
|
}
|
|
|
|
|
2006-05-29 23:18:33 +04:00
|
|
|
int parse_tree_buffer(struct tree *item, void *buffer, unsigned long size)
|
|
|
|
{
|
2005-04-18 22:39:48 +04:00
|
|
|
if (item->object.parsed)
|
|
|
|
return 0;
|
|
|
|
item->object.parsed = 1;
|
2006-05-29 23:16:12 +04:00
|
|
|
item->buffer = buffer;
|
|
|
|
item->size = size;
|
|
|
|
|
2006-05-29 23:18:33 +04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-05-06 21:48:34 +04:00
|
|
|
int parse_tree(struct tree *item)
|
|
|
|
{
|
2007-02-26 22:55:59 +03:00
|
|
|
enum object_type type;
|
2005-05-06 21:48:34 +04:00
|
|
|
void *buffer;
|
|
|
|
unsigned long size;
|
|
|
|
|
|
|
|
if (item->object.parsed)
|
|
|
|
return 0;
|
2007-02-26 22:55:59 +03:00
|
|
|
buffer = read_sha1_file(item->object.sha1, &type, &size);
|
2005-05-06 21:48:34 +04:00
|
|
|
if (!buffer)
|
|
|
|
return error("Could not read %s",
|
|
|
|
sha1_to_hex(item->object.sha1));
|
2007-02-26 22:55:59 +03:00
|
|
|
if (type != OBJ_TREE) {
|
2005-05-06 21:48:34 +04:00
|
|
|
free(buffer);
|
|
|
|
return error("Object %s not a tree",
|
|
|
|
sha1_to_hex(item->object.sha1));
|
|
|
|
}
|
2006-05-29 23:16:12 +04:00
|
|
|
return parse_tree_buffer(item, buffer, size);
|
2005-05-06 21:48:34 +04:00
|
|
|
}
|
2005-09-05 10:03:51 +04:00
|
|
|
|
|
|
|
struct tree *parse_tree_indirect(const unsigned char *sha1)
|
|
|
|
{
|
|
|
|
struct object *obj = parse_object(sha1);
|
|
|
|
do {
|
|
|
|
if (!obj)
|
|
|
|
return NULL;
|
2006-07-12 07:45:31 +04:00
|
|
|
if (obj->type == OBJ_TREE)
|
2005-09-05 10:03:51 +04:00
|
|
|
return (struct tree *) obj;
|
2006-07-12 07:45:31 +04:00
|
|
|
else if (obj->type == OBJ_COMMIT)
|
2005-09-05 10:03:51 +04:00
|
|
|
obj = &(((struct commit *) obj)->tree->object);
|
2006-07-12 07:45:31 +04:00
|
|
|
else if (obj->type == OBJ_TAG)
|
2005-09-05 10:03:51 +04:00
|
|
|
obj = ((struct tag *) obj)->tagged;
|
|
|
|
else
|
|
|
|
return NULL;
|
|
|
|
if (!obj->parsed)
|
|
|
|
parse_object(obj->sha1);
|
|
|
|
} while (1);
|
|
|
|
}
|