2006-01-07 10:33:54 +01:00
|
|
|
#include "cache.h"
|
2007-08-09 22:42:50 +02:00
|
|
|
#include "cache-tree.h"
|
2005-04-18 20:39:48 +02:00
|
|
|
#include "tree.h"
|
|
|
|
#include "blob.h"
|
2005-09-05 08:03:51 +02:00
|
|
|
#include "commit.h"
|
|
|
|
#include "tag.h"
|
2006-05-29 21:16:12 +02:00
|
|
|
#include "tree-walk.h"
|
2005-04-18 20:39:48 +02:00
|
|
|
|
|
|
|
const char *tree_type = "tree";
|
|
|
|
|
2007-08-09 22:42:50 +02:00
|
|
|
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 01:42:37 +02:00
|
|
|
{
|
2005-11-26 18:38:20 +01:00
|
|
|
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 20:30:46 +02:00
|
|
|
ce = xcalloc(1, size);
|
2005-04-23 01:42:37 +02:00
|
|
|
|
|
|
|
ce->ce_mode = create_ce_mode(mode);
|
|
|
|
ce->ce_flags = create_ce_flags(baselen + len, stage);
|
|
|
|
memcpy(ce->name, base, baselen);
|
|
|
|
memcpy(ce->name + baselen, pathname, len+1);
|
2006-08-23 08:49:00 +02:00
|
|
|
hashcpy(ce->sha1, sha1);
|
2007-08-09 22:42:50 +02:00
|
|
|
return add_cache_entry(ce, opt);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int read_one_entry(const unsigned char *sha1, const char *base, int baselen, const char *pathname, unsigned mode, int stage)
|
|
|
|
{
|
|
|
|
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.
|
|
|
|
*/
|
|
|
|
static int read_one_entry_quick(const unsigned char *sha1, const char *base, int baselen, const char *pathname, unsigned mode, int stage)
|
|
|
|
{
|
|
|
|
return read_one_entry_opt(sha1, base, baselen, pathname, mode, stage,
|
|
|
|
ADD_CACHE_JUST_APPEND);
|
2005-04-23 01:42:37 +02:00
|
|
|
}
|
|
|
|
|
2005-07-14 20:39:27 +02:00
|
|
|
static int match_tree_entry(const char *base, int baselen, const char *path, unsigned int mode, const char **paths)
|
2005-07-14 20:26:31 +02:00
|
|
|
{
|
2005-07-14 20:39:27 +02:00
|
|
|
const char *match;
|
2005-07-14 20:26:31 +02:00
|
|
|
int pathlen;
|
|
|
|
|
|
|
|
if (!paths)
|
|
|
|
return 1;
|
|
|
|
pathlen = strlen(path);
|
|
|
|
while ((match = *paths++) != NULL) {
|
|
|
|
int matchlen = strlen(match);
|
|
|
|
|
|
|
|
if (baselen >= matchlen) {
|
|
|
|
/* If it doesn't match, move along... */
|
|
|
|
if (strncmp(base, match, matchlen))
|
|
|
|
continue;
|
|
|
|
/* The base is a subdirectory of a path which was specified. */
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Does the base match? */
|
|
|
|
if (strncmp(base, match, baselen))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
match += baselen;
|
|
|
|
matchlen -= baselen;
|
|
|
|
|
|
|
|
if (pathlen > matchlen)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (matchlen > pathlen) {
|
|
|
|
if (match[pathlen] != '/')
|
|
|
|
continue;
|
|
|
|
if (!S_ISDIR(mode))
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (strncmp(path, match, pathlen))
|
|
|
|
continue;
|
2005-07-14 20:39:27 +02:00
|
|
|
|
|
|
|
return 1;
|
2005-07-14 20:26:31 +02:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-01-26 07:13:36 +01:00
|
|
|
int read_tree_recursive(struct tree *tree,
|
2005-11-26 18:38:20 +01:00
|
|
|
const char *base, int baselen,
|
|
|
|
int stage, const char **match,
|
|
|
|
read_tree_fn_t fn)
|
2005-04-23 01:42:37 +02:00
|
|
|
{
|
2006-05-29 21:17:28 +02:00
|
|
|
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 18:45:45 +02:00
|
|
|
struct name_entry entry;
|
2006-05-29 21:17:28 +02:00
|
|
|
|
2006-01-26 07:13:36 +01:00
|
|
|
if (parse_tree(tree))
|
|
|
|
return -1;
|
2006-05-29 21:17:28 +02:00
|
|
|
|
2007-03-21 18:08:25 +01:00
|
|
|
init_tree_desc(&desc, tree->buffer, tree->size);
|
2006-05-29 21:17:28 +02:00
|
|
|
|
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 18:45:45 +02:00
|
|
|
while (tree_entry(&desc, &entry)) {
|
|
|
|
if (!match_tree_entry(base, baselen, entry.path, entry.mode, match))
|
2005-07-14 20:26:31 +02:00
|
|
|
continue;
|
|
|
|
|
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 18:45:45 +02:00
|
|
|
switch (fn(entry.sha1, base, baselen, entry.path, entry.mode, stage)) {
|
2005-11-26 18:38:20 +01:00
|
|
|
case 0:
|
|
|
|
continue;
|
|
|
|
case READ_TREE_RECURSIVE:
|
|
|
|
break;;
|
|
|
|
default:
|
|
|
|
return -1;
|
|
|
|
}
|
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 18:45:45 +02:00
|
|
|
if (S_ISDIR(entry.mode)) {
|
2005-04-23 01:42:37 +02:00
|
|
|
int retval;
|
2005-04-27 00:00:01 +02:00
|
|
|
char *newbase;
|
2007-03-21 18:07:46 +01:00
|
|
|
unsigned int pathlen = tree_entry_len(entry.path, entry.sha1);
|
2005-04-23 01:42:37 +02:00
|
|
|
|
2007-03-21 18:07:46 +01:00
|
|
|
newbase = xmalloc(baselen + 1 + pathlen);
|
2005-04-23 01:42:37 +02:00
|
|
|
memcpy(newbase, base, baselen);
|
2007-03-21 18:07:46 +01:00
|
|
|
memcpy(newbase + baselen, entry.path, pathlen);
|
|
|
|
newbase[baselen + pathlen] = '/';
|
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 18:45:45 +02:00
|
|
|
retval = read_tree_recursive(lookup_tree(entry.sha1),
|
2005-04-23 01:42:37 +02:00
|
|
|
newbase,
|
2007-03-21 18:07:46 +01:00
|
|
|
baselen + pathlen + 1,
|
2005-11-26 18:38:20 +01:00
|
|
|
stage, match, fn);
|
2005-04-23 01:42:37 +02:00
|
|
|
free(newbase);
|
|
|
|
if (retval)
|
|
|
|
return -1;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-08-09 22:42:50 +02:00
|
|
|
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_);
|
|
|
|
return cache_name_compare(ce1->name, ntohs(ce1->ce_flags),
|
|
|
|
ce2->name, ntohs(ce2->ce_flags));
|
|
|
|
}
|
|
|
|
|
2006-01-26 07:13:36 +01:00
|
|
|
int read_tree(struct tree *tree, int stage, const char **match)
|
2005-04-23 01:42:37 +02:00
|
|
|
{
|
2007-08-09 22:42:50 +02:00
|
|
|
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++) {
|
|
|
|
struct cache_entry *ce = active_cache[i];
|
|
|
|
if (ce_stage(ce) == stage)
|
|
|
|
fn = read_one_entry;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!fn)
|
|
|
|
fn = read_one_entry_quick;
|
|
|
|
err = read_tree_recursive(tree, "", 0, stage, match, fn);
|
|
|
|
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 01:42:37 +02:00
|
|
|
}
|
|
|
|
|
2005-06-03 17:05:39 +02:00
|
|
|
struct tree *lookup_tree(const unsigned char *sha1)
|
2005-04-18 20:39:48 +02:00
|
|
|
{
|
|
|
|
struct object *obj = lookup_object(sha1);
|
2007-04-17 07:11:43 +02:00
|
|
|
if (!obj)
|
|
|
|
return create_object(sha1, OBJ_TREE, alloc_tree_node());
|
2005-05-20 22:59:17 +02:00
|
|
|
if (!obj->type)
|
2006-07-12 05:45:31 +02:00
|
|
|
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 01:45:13 +02:00
|
|
|
error("Object %s is a %s, not a tree",
|
|
|
|
sha1_to_hex(sha1), typename(obj->type));
|
2005-04-18 20:39:48 +02:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return (struct tree *) obj;
|
|
|
|
}
|
|
|
|
|
2006-05-29 21:18:33 +02:00
|
|
|
int parse_tree_buffer(struct tree *item, void *buffer, unsigned long size)
|
|
|
|
{
|
2005-04-18 20:39:48 +02:00
|
|
|
if (item->object.parsed)
|
|
|
|
return 0;
|
|
|
|
item->object.parsed = 1;
|
2006-05-29 21:16:12 +02:00
|
|
|
item->buffer = buffer;
|
|
|
|
item->size = size;
|
|
|
|
|
2006-05-29 21:18:33 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-05-06 19:48:34 +02:00
|
|
|
int parse_tree(struct tree *item)
|
|
|
|
{
|
2007-02-26 20:55:59 +01:00
|
|
|
enum object_type type;
|
2005-05-06 19:48:34 +02:00
|
|
|
void *buffer;
|
|
|
|
unsigned long size;
|
|
|
|
|
|
|
|
if (item->object.parsed)
|
|
|
|
return 0;
|
2007-02-26 20:55:59 +01:00
|
|
|
buffer = read_sha1_file(item->object.sha1, &type, &size);
|
2005-05-06 19:48:34 +02:00
|
|
|
if (!buffer)
|
|
|
|
return error("Could not read %s",
|
|
|
|
sha1_to_hex(item->object.sha1));
|
2007-02-26 20:55:59 +01:00
|
|
|
if (type != OBJ_TREE) {
|
2005-05-06 19:48:34 +02:00
|
|
|
free(buffer);
|
|
|
|
return error("Object %s not a tree",
|
|
|
|
sha1_to_hex(item->object.sha1));
|
|
|
|
}
|
2006-05-29 21:16:12 +02:00
|
|
|
return parse_tree_buffer(item, buffer, size);
|
2005-05-06 19:48:34 +02:00
|
|
|
}
|
2005-09-05 08:03:51 +02:00
|
|
|
|
|
|
|
struct tree *parse_tree_indirect(const unsigned char *sha1)
|
|
|
|
{
|
|
|
|
struct object *obj = parse_object(sha1);
|
|
|
|
do {
|
|
|
|
if (!obj)
|
|
|
|
return NULL;
|
2006-07-12 05:45:31 +02:00
|
|
|
if (obj->type == OBJ_TREE)
|
2005-09-05 08:03:51 +02:00
|
|
|
return (struct tree *) obj;
|
2006-07-12 05:45:31 +02:00
|
|
|
else if (obj->type == OBJ_COMMIT)
|
2005-09-05 08:03:51 +02:00
|
|
|
obj = &(((struct commit *) obj)->tree->object);
|
2006-07-12 05:45:31 +02:00
|
|
|
else if (obj->type == OBJ_TAG)
|
2005-09-05 08:03:51 +02:00
|
|
|
obj = ((struct tag *) obj)->tagged;
|
|
|
|
else
|
|
|
|
return NULL;
|
|
|
|
if (!obj->parsed)
|
|
|
|
parse_object(obj->sha1);
|
|
|
|
} while (1);
|
|
|
|
}
|