2005-06-06 22:31:29 +02:00
|
|
|
#include "refs.h"
|
|
|
|
#include "cache.h"
|
|
|
|
|
|
|
|
#include <errno.h>
|
|
|
|
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
struct ref_list {
|
|
|
|
struct ref_list *next;
|
2006-09-21 07:02:01 +02:00
|
|
|
unsigned char flag; /* ISSYMREF? ISPACKED? */
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
unsigned char sha1[20];
|
|
|
|
char name[FLEX_ARRAY];
|
|
|
|
};
|
|
|
|
|
|
|
|
static const char *parse_ref_line(char *line, unsigned char *sha1)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* 42: the answer to everything.
|
|
|
|
*
|
|
|
|
* In this case, it happens to be the answer to
|
|
|
|
* 40 (length of sha1 hex representation)
|
|
|
|
* +1 (space in between hex and name)
|
|
|
|
* +1 (newline at the end of the line)
|
|
|
|
*/
|
|
|
|
int len = strlen(line) - 42;
|
|
|
|
|
|
|
|
if (len <= 0)
|
|
|
|
return NULL;
|
|
|
|
if (get_sha1_hex(line, sha1) < 0)
|
|
|
|
return NULL;
|
|
|
|
if (!isspace(line[40]))
|
|
|
|
return NULL;
|
|
|
|
line += 41;
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
if (isspace(*line))
|
|
|
|
return NULL;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
if (line[len] != '\n')
|
|
|
|
return NULL;
|
|
|
|
line[len] = 0;
|
|
|
|
return line;
|
|
|
|
}
|
|
|
|
|
2006-09-21 07:02:01 +02:00
|
|
|
static struct ref_list *add_ref(const char *name, const unsigned char *sha1,
|
|
|
|
int flag, struct ref_list *list)
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
{
|
|
|
|
int len;
|
|
|
|
struct ref_list **p = &list, *entry;
|
|
|
|
|
|
|
|
/* Find the place to insert the ref into.. */
|
|
|
|
while ((entry = *p) != NULL) {
|
|
|
|
int cmp = strcmp(entry->name, name);
|
|
|
|
if (cmp > 0)
|
|
|
|
break;
|
|
|
|
|
|
|
|
/* Same as existing entry? */
|
|
|
|
if (!cmp)
|
|
|
|
return list;
|
|
|
|
p = &entry->next;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Allocate it and add it in.. */
|
|
|
|
len = strlen(name) + 1;
|
|
|
|
entry = xmalloc(sizeof(struct ref_list) + len);
|
|
|
|
hashcpy(entry->sha1, sha1);
|
|
|
|
memcpy(entry->name, name, len);
|
2006-09-21 07:02:01 +02:00
|
|
|
entry->flag = flag;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
entry->next = *p;
|
|
|
|
*p = entry;
|
|
|
|
return list;
|
|
|
|
}
|
|
|
|
|
2006-09-30 21:37:37 +02:00
|
|
|
/*
|
|
|
|
* Future: need to be in "struct repository"
|
|
|
|
* when doing a full libification.
|
|
|
|
*/
|
|
|
|
struct cached_refs {
|
|
|
|
char did_loose;
|
|
|
|
char did_packed;
|
|
|
|
struct ref_list *loose;
|
|
|
|
struct ref_list *packed;
|
|
|
|
} cached_refs;
|
|
|
|
|
|
|
|
static void free_ref_list(struct ref_list *list)
|
|
|
|
{
|
|
|
|
struct ref_list *next;
|
|
|
|
for ( ; list; list = next) {
|
|
|
|
next = list->next;
|
|
|
|
free(list);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void invalidate_cached_refs(void)
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
{
|
2006-09-30 21:37:37 +02:00
|
|
|
struct cached_refs *ca = &cached_refs;
|
|
|
|
|
|
|
|
if (ca->did_loose && ca->loose)
|
|
|
|
free_ref_list(ca->loose);
|
|
|
|
if (ca->did_packed && ca->packed)
|
|
|
|
free_ref_list(ca->packed);
|
|
|
|
ca->loose = ca->packed = NULL;
|
|
|
|
ca->did_loose = ca->did_packed = 0;
|
|
|
|
}
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
|
2006-09-30 21:37:37 +02:00
|
|
|
static struct ref_list *get_packed_refs(void)
|
|
|
|
{
|
|
|
|
if (!cached_refs.did_packed) {
|
|
|
|
struct ref_list *refs = NULL;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
FILE *f = fopen(git_path("packed-refs"), "r");
|
|
|
|
if (f) {
|
|
|
|
struct ref_list *list = NULL;
|
|
|
|
char refline[PATH_MAX];
|
|
|
|
while (fgets(refline, sizeof(refline), f)) {
|
|
|
|
unsigned char sha1[20];
|
|
|
|
const char *name = parse_ref_line(refline, sha1);
|
|
|
|
if (!name)
|
|
|
|
continue;
|
2006-09-21 07:02:01 +02:00
|
|
|
list = add_ref(name, sha1, REF_ISPACKED, list);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
|
|
|
fclose(f);
|
|
|
|
refs = list;
|
|
|
|
}
|
2006-09-30 21:37:37 +02:00
|
|
|
cached_refs.packed = refs;
|
|
|
|
cached_refs.did_packed = 1;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
2006-09-30 21:37:37 +02:00
|
|
|
return cached_refs.packed;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct ref_list *get_ref_dir(const char *base, struct ref_list *list)
|
|
|
|
{
|
|
|
|
DIR *dir = opendir(git_path("%s", base));
|
|
|
|
|
|
|
|
if (dir) {
|
|
|
|
struct dirent *de;
|
|
|
|
int baselen = strlen(base);
|
2006-09-12 05:17:35 +02:00
|
|
|
char *ref = xmalloc(baselen + 257);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
|
2006-09-12 05:17:35 +02:00
|
|
|
memcpy(ref, base, baselen);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
if (baselen && base[baselen-1] != '/')
|
2006-09-12 05:17:35 +02:00
|
|
|
ref[baselen++] = '/';
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
|
|
|
|
while ((de = readdir(dir)) != NULL) {
|
|
|
|
unsigned char sha1[20];
|
|
|
|
struct stat st;
|
2006-09-21 07:02:01 +02:00
|
|
|
int flag;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
int namelen;
|
|
|
|
|
|
|
|
if (de->d_name[0] == '.')
|
|
|
|
continue;
|
|
|
|
namelen = strlen(de->d_name);
|
|
|
|
if (namelen > 255)
|
|
|
|
continue;
|
|
|
|
if (has_extension(de->d_name, ".lock"))
|
|
|
|
continue;
|
2006-09-12 05:17:35 +02:00
|
|
|
memcpy(ref + baselen, de->d_name, namelen+1);
|
|
|
|
if (stat(git_path("%s", ref), &st) < 0)
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
continue;
|
|
|
|
if (S_ISDIR(st.st_mode)) {
|
2006-09-12 05:17:35 +02:00
|
|
|
list = get_ref_dir(ref, list);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
continue;
|
|
|
|
}
|
2006-09-21 07:02:01 +02:00
|
|
|
if (!resolve_ref(ref, sha1, 1, &flag)) {
|
2006-09-12 05:17:35 +02:00
|
|
|
error("%s points nowhere!", ref);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
continue;
|
|
|
|
}
|
2006-09-21 07:02:01 +02:00
|
|
|
list = add_ref(ref, sha1, flag, list);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
2006-09-12 05:17:35 +02:00
|
|
|
free(ref);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
closedir(dir);
|
|
|
|
}
|
|
|
|
return list;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct ref_list *get_loose_refs(void)
|
|
|
|
{
|
2006-09-30 21:37:37 +02:00
|
|
|
if (!cached_refs.did_loose) {
|
|
|
|
cached_refs.loose = get_ref_dir("refs", NULL);
|
|
|
|
cached_refs.did_loose = 1;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
2006-09-30 21:37:37 +02:00
|
|
|
return cached_refs.loose;
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
}
|
|
|
|
|
2005-09-25 18:59:37 +02:00
|
|
|
/* We allow "recursive" symbolic refs. Only within reason, though */
|
|
|
|
#define MAXDEPTH 5
|
|
|
|
|
2006-09-21 07:02:01 +02:00
|
|
|
const char *resolve_ref(const char *ref, unsigned char *sha1, int reading, int *flag)
|
2005-07-03 05:23:36 +02:00
|
|
|
{
|
2005-09-30 23:08:25 +02:00
|
|
|
int depth = MAXDEPTH, len;
|
|
|
|
char buffer[256];
|
2006-09-12 05:17:35 +02:00
|
|
|
static char ref_buffer[256];
|
2005-09-25 18:59:37 +02:00
|
|
|
|
2006-09-21 07:02:01 +02:00
|
|
|
if (flag)
|
|
|
|
*flag = 0;
|
|
|
|
|
2005-09-30 23:08:25 +02:00
|
|
|
for (;;) {
|
2006-09-12 05:17:35 +02:00
|
|
|
const char *path = git_path("%s", ref);
|
2005-09-30 23:08:25 +02:00
|
|
|
struct stat st;
|
|
|
|
char *buf;
|
|
|
|
int fd;
|
2005-07-03 05:23:36 +02:00
|
|
|
|
2005-09-30 23:08:25 +02:00
|
|
|
if (--depth < 0)
|
|
|
|
return NULL;
|
2005-09-25 18:59:37 +02:00
|
|
|
|
2005-09-30 23:08:25 +02:00
|
|
|
/* Special case: non-existing file.
|
|
|
|
* Not having the refs/heads/new-branch is OK
|
|
|
|
* if we are writing into it, so is .git/HEAD
|
|
|
|
* that points at refs/heads/master still to be
|
|
|
|
* born. It is NOT OK if we are resolving for
|
|
|
|
* reading.
|
|
|
|
*/
|
|
|
|
if (lstat(path, &st) < 0) {
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
struct ref_list *list = get_packed_refs();
|
|
|
|
while (list) {
|
|
|
|
if (!strcmp(ref, list->name)) {
|
|
|
|
hashcpy(sha1, list->sha1);
|
2006-09-21 07:02:01 +02:00
|
|
|
if (flag)
|
|
|
|
*flag |= REF_ISPACKED;
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
return ref;
|
|
|
|
}
|
|
|
|
list = list->next;
|
|
|
|
}
|
2005-09-30 23:08:25 +02:00
|
|
|
if (reading || errno != ENOENT)
|
|
|
|
return NULL;
|
2006-08-23 08:49:00 +02:00
|
|
|
hashclr(sha1);
|
2006-09-12 05:17:35 +02:00
|
|
|
return ref;
|
2005-09-30 23:08:25 +02:00
|
|
|
}
|
2005-09-25 18:59:37 +02:00
|
|
|
|
2005-09-30 23:08:25 +02:00
|
|
|
/* Follow "normalized" - ie "refs/.." symlinks by hand */
|
|
|
|
if (S_ISLNK(st.st_mode)) {
|
|
|
|
len = readlink(path, buffer, sizeof(buffer)-1);
|
|
|
|
if (len >= 5 && !memcmp("refs/", buffer, 5)) {
|
2006-09-12 05:17:35 +02:00
|
|
|
buffer[len] = 0;
|
|
|
|
strcpy(ref_buffer, buffer);
|
|
|
|
ref = ref_buffer;
|
2006-09-21 07:02:01 +02:00
|
|
|
if (flag)
|
|
|
|
*flag |= REF_ISSYMREF;
|
2005-09-30 23:08:25 +02:00
|
|
|
continue;
|
|
|
|
}
|
2005-09-25 18:59:37 +02:00
|
|
|
}
|
2005-09-30 23:08:25 +02:00
|
|
|
|
2006-10-02 19:23:53 +02:00
|
|
|
/* Is it a directory? */
|
|
|
|
if (S_ISDIR(st.st_mode)) {
|
|
|
|
errno = EISDIR;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2005-09-30 23:08:25 +02:00
|
|
|
/*
|
|
|
|
* Anything else, just open it and try to use it as
|
|
|
|
* a ref
|
|
|
|
*/
|
|
|
|
fd = open(path, O_RDONLY);
|
|
|
|
if (fd < 0)
|
|
|
|
return NULL;
|
|
|
|
len = read(fd, buffer, sizeof(buffer)-1);
|
|
|
|
close(fd);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Is it a symbolic ref?
|
|
|
|
*/
|
|
|
|
if (len < 4 || memcmp("ref:", buffer, 4))
|
|
|
|
break;
|
|
|
|
buf = buffer + 4;
|
|
|
|
len -= 4;
|
|
|
|
while (len && isspace(*buf))
|
|
|
|
buf++, len--;
|
|
|
|
while (len && isspace(buf[len-1]))
|
2006-09-12 05:17:35 +02:00
|
|
|
len--;
|
|
|
|
buf[len] = 0;
|
|
|
|
memcpy(ref_buffer, buf, len + 1);
|
|
|
|
ref = ref_buffer;
|
2006-09-21 07:02:01 +02:00
|
|
|
if (flag)
|
|
|
|
*flag |= REF_ISSYMREF;
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
2005-09-30 23:08:25 +02:00
|
|
|
if (len < 40 || get_sha1_hex(buffer, sha1))
|
|
|
|
return NULL;
|
2006-09-12 05:17:35 +02:00
|
|
|
return ref;
|
2005-09-30 23:08:25 +02:00
|
|
|
}
|
|
|
|
|
2006-09-12 05:17:35 +02:00
|
|
|
int create_symref(const char *ref_target, const char *refs_heads_master)
|
2005-09-30 23:26:57 +02:00
|
|
|
{
|
|
|
|
const char *lockpath;
|
|
|
|
char ref[1000];
|
|
|
|
int fd, len, written;
|
2006-09-12 05:17:35 +02:00
|
|
|
const char *git_HEAD = git_path("%s", ref_target);
|
2005-09-30 23:26:57 +02:00
|
|
|
|
2006-05-02 09:40:24 +02:00
|
|
|
#ifndef NO_SYMLINK_HEAD
|
|
|
|
if (prefer_symlink_refs) {
|
2005-11-15 19:24:19 +01:00
|
|
|
unlink(git_HEAD);
|
|
|
|
if (!symlink(refs_heads_master, git_HEAD))
|
|
|
|
return 0;
|
|
|
|
fprintf(stderr, "no symlink - falling back to symbolic ref\n");
|
|
|
|
}
|
2005-10-26 01:40:31 +02:00
|
|
|
#endif
|
|
|
|
|
2005-09-30 23:26:57 +02:00
|
|
|
len = snprintf(ref, sizeof(ref), "ref: %s\n", refs_heads_master);
|
|
|
|
if (sizeof(ref) <= len) {
|
|
|
|
error("refname too long: %s", refs_heads_master);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
lockpath = mkpath("%s.lock", git_HEAD);
|
|
|
|
fd = open(lockpath, O_CREAT | O_EXCL | O_WRONLY, 0666);
|
|
|
|
written = write(fd, ref, len);
|
|
|
|
close(fd);
|
|
|
|
if (written != len) {
|
|
|
|
unlink(lockpath);
|
|
|
|
error("Unable to write to %s", lockpath);
|
|
|
|
return -2;
|
|
|
|
}
|
|
|
|
if (rename(lockpath, git_HEAD) < 0) {
|
|
|
|
unlink(lockpath);
|
|
|
|
error("Unable to create %s", git_HEAD);
|
|
|
|
return -3;
|
|
|
|
}
|
2006-06-10 07:07:23 +02:00
|
|
|
if (adjust_shared_perm(git_HEAD)) {
|
|
|
|
unlink(lockpath);
|
|
|
|
error("Unable to fix permissions on %s", lockpath);
|
|
|
|
return -4;
|
|
|
|
}
|
2005-09-30 23:26:57 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-09-12 05:17:35 +02:00
|
|
|
int read_ref(const char *ref, unsigned char *sha1)
|
2005-09-30 23:08:25 +02:00
|
|
|
{
|
2006-09-21 07:02:01 +02:00
|
|
|
if (resolve_ref(ref, sha1, 1, NULL))
|
2005-09-30 23:08:25 +02:00
|
|
|
return 0;
|
|
|
|
return -1;
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 07:02:01 +02:00
|
|
|
static int do_for_each_ref(const char *base, each_ref_fn fn, int trim,
|
|
|
|
void *cb_data)
|
2005-07-03 05:23:36 +02:00
|
|
|
{
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
int retval;
|
|
|
|
struct ref_list *packed = get_packed_refs();
|
|
|
|
struct ref_list *loose = get_loose_refs();
|
2005-07-03 05:23:36 +02:00
|
|
|
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
while (packed && loose) {
|
|
|
|
struct ref_list *entry;
|
|
|
|
int cmp = strcmp(packed->name, loose->name);
|
|
|
|
if (!cmp) {
|
|
|
|
packed = packed->next;
|
|
|
|
continue;
|
2005-07-05 00:28:19 +02:00
|
|
|
}
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
if (cmp > 0) {
|
|
|
|
entry = loose;
|
|
|
|
loose = loose->next;
|
|
|
|
} else {
|
|
|
|
entry = packed;
|
|
|
|
packed = packed->next;
|
|
|
|
}
|
|
|
|
if (strncmp(base, entry->name, trim))
|
|
|
|
continue;
|
2006-09-12 05:10:15 +02:00
|
|
|
if (is_null_sha1(entry->sha1))
|
|
|
|
continue;
|
|
|
|
if (!has_sha1_file(entry->sha1)) {
|
|
|
|
error("%s does not point to a valid object!", entry->name);
|
|
|
|
continue;
|
|
|
|
}
|
2006-09-21 07:02:01 +02:00
|
|
|
retval = fn(entry->name + trim, entry->sha1,
|
|
|
|
entry->flag, cb_data);
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
if (retval)
|
|
|
|
return retval;
|
|
|
|
}
|
2005-07-03 05:23:36 +02:00
|
|
|
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
packed = packed ? packed : loose;
|
|
|
|
while (packed) {
|
|
|
|
if (!strncmp(base, packed->name, trim)) {
|
2006-09-21 07:02:01 +02:00
|
|
|
retval = fn(packed->name + trim, packed->sha1,
|
|
|
|
packed->flag, cb_data);
|
2005-07-03 05:23:36 +02:00
|
|
|
if (retval)
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
return retval;
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
packed = packed->next;
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
Start handling references internally as a sorted in-memory list
This also adds some very rudimentary support for the notion of packed
refs. HOWEVER! At this point it isn't used to actually look up a ref
yet, only for listing them (ie "for_each_ref()" and friends see the
packed refs, but none of the other single-ref lookup routines).
Note how we keep two separate lists: one for the loose refs, and one for
the packed refs we read. That's so that we can easily keep the two apart,
and read only one set or the other (and still always make sure that the
loose refs take precedence).
[ From this, it's not actually obvious why we'd keep the two separate
lists, but it's important to have the packed refs on their own list
later on, when I add support for looking up a single loose one.
For that case, we will want to read _just_ the packed refs in case the
single-ref lookup fails, yet we may end up needing the other list at
some point in the future, so keeping them separated is important ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-12 01:37:32 +02:00
|
|
|
return 0;
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
int head_ref(each_ref_fn fn, void *cb_data)
|
2005-07-05 20:31:32 +02:00
|
|
|
{
|
|
|
|
unsigned char sha1[20];
|
2006-09-21 07:02:01 +02:00
|
|
|
int flag;
|
|
|
|
|
|
|
|
if (resolve_ref("HEAD", sha1, 1, &flag))
|
|
|
|
return fn("HEAD", sha1, flag, cb_data);
|
2005-07-06 00:45:00 +02:00
|
|
|
return 0;
|
2005-07-05 20:31:32 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
int for_each_ref(each_ref_fn fn, void *cb_data)
|
2005-07-03 05:23:36 +02:00
|
|
|
{
|
2006-09-21 06:47:42 +02:00
|
|
|
return do_for_each_ref("refs/", fn, 0, cb_data);
|
2006-05-14 03:43:00 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
int for_each_tag_ref(each_ref_fn fn, void *cb_data)
|
2006-05-14 03:43:00 +02:00
|
|
|
{
|
2006-09-21 06:47:42 +02:00
|
|
|
return do_for_each_ref("refs/tags/", fn, 10, cb_data);
|
2006-05-14 03:43:00 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
int for_each_branch_ref(each_ref_fn fn, void *cb_data)
|
2006-05-14 03:43:00 +02:00
|
|
|
{
|
2006-09-21 06:47:42 +02:00
|
|
|
return do_for_each_ref("refs/heads/", fn, 11, cb_data);
|
2006-05-14 03:43:00 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
int for_each_remote_ref(each_ref_fn fn, void *cb_data)
|
2006-05-14 03:43:00 +02:00
|
|
|
{
|
2006-09-21 06:47:42 +02:00
|
|
|
return do_for_each_ref("refs/remotes/", fn, 13, cb_data);
|
2005-07-03 05:23:36 +02:00
|
|
|
}
|
|
|
|
|
2006-09-21 06:47:42 +02:00
|
|
|
/* NEEDSWORK: This is only used by ssh-upload and it should go; the
|
|
|
|
* caller should do resolve_ref or read_ref like everybody else. Or
|
|
|
|
* maybe everybody else should use get_ref_sha1() instead of doing
|
|
|
|
* read_ref().
|
|
|
|
*/
|
2005-06-06 22:31:29 +02:00
|
|
|
int get_ref_sha1(const char *ref, unsigned char *sha1)
|
|
|
|
{
|
|
|
|
if (check_ref_format(ref))
|
|
|
|
return -1;
|
2006-09-12 05:17:35 +02:00
|
|
|
return read_ref(mkpath("refs/%s", ref), sha1);
|
2005-06-06 22:31:29 +02:00
|
|
|
}
|
|
|
|
|
2005-10-14 03:57:39 +02:00
|
|
|
/*
|
|
|
|
* Make sure "ref" is something reasonable to have under ".git/refs/";
|
|
|
|
* We do not like it if:
|
|
|
|
*
|
|
|
|
* - any path component of it begins with ".", or
|
|
|
|
* - it has double dots "..", or
|
|
|
|
* - it has ASCII control character, "~", "^", ":" or SP, anywhere, or
|
|
|
|
* - it ends with a "/".
|
|
|
|
*/
|
|
|
|
|
|
|
|
static inline int bad_ref_char(int ch)
|
|
|
|
{
|
|
|
|
return (((unsigned) ch) <= ' ' ||
|
2005-12-16 03:03:59 +01:00
|
|
|
ch == '~' || ch == '^' || ch == ':' ||
|
|
|
|
/* 2.13 Pattern Matching Notation */
|
|
|
|
ch == '?' || ch == '*' || ch == '[');
|
2005-10-14 03:57:39 +02:00
|
|
|
}
|
|
|
|
|
2005-06-06 22:31:29 +02:00
|
|
|
int check_ref_format(const char *ref)
|
|
|
|
{
|
2005-10-14 03:57:39 +02:00
|
|
|
int ch, level;
|
|
|
|
const char *cp = ref;
|
|
|
|
|
|
|
|
level = 0;
|
|
|
|
while (1) {
|
|
|
|
while ((ch = *cp++) == '/')
|
|
|
|
; /* tolerate duplicated slashes */
|
|
|
|
if (!ch)
|
|
|
|
return -1; /* should not end with slashes */
|
|
|
|
|
|
|
|
/* we are at the beginning of the path component */
|
|
|
|
if (ch == '.' || bad_ref_char(ch))
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
/* scan the rest of the path component */
|
|
|
|
while ((ch = *cp++) != 0) {
|
|
|
|
if (bad_ref_char(ch))
|
|
|
|
return -1;
|
|
|
|
if (ch == '/')
|
|
|
|
break;
|
|
|
|
if (ch == '.' && *cp == '.')
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
level++;
|
|
|
|
if (!ch) {
|
|
|
|
if (level < 2)
|
|
|
|
return -1; /* at least of form "heads/blah" */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
2005-06-06 22:31:29 +02:00
|
|
|
}
|
|
|
|
|
2006-06-06 23:04:17 +02:00
|
|
|
static struct ref_lock *verify_lock(struct ref_lock *lock,
|
2006-05-17 11:55:02 +02:00
|
|
|
const unsigned char *old_sha1, int mustexist)
|
|
|
|
{
|
2006-09-21 07:02:01 +02:00
|
|
|
if (!resolve_ref(lock->ref_name, lock->old_sha1, mustexist, NULL)) {
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
error("Can't verify ref %s", lock->ref_name);
|
2006-05-17 11:55:02 +02:00
|
|
|
unlock_ref(lock);
|
|
|
|
return NULL;
|
|
|
|
}
|
2006-08-17 20:54:57 +02:00
|
|
|
if (hashcmp(lock->old_sha1, old_sha1)) {
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
error("Ref %s is at %s but expected %s", lock->ref_name,
|
2006-05-17 11:55:02 +02:00
|
|
|
sha1_to_hex(lock->old_sha1), sha1_to_hex(old_sha1));
|
|
|
|
unlock_ref(lock);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
return lock;
|
|
|
|
}
|
|
|
|
|
2006-09-30 11:25:30 +02:00
|
|
|
static int remove_empty_dir_recursive(char *path, int len)
|
|
|
|
{
|
|
|
|
DIR *dir = opendir(path);
|
|
|
|
struct dirent *e;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
if (!dir)
|
|
|
|
return -1;
|
|
|
|
if (path[len-1] != '/')
|
|
|
|
path[len++] = '/';
|
|
|
|
while ((e = readdir(dir)) != NULL) {
|
|
|
|
struct stat st;
|
|
|
|
int namlen;
|
|
|
|
if ((e->d_name[0] == '.') &&
|
|
|
|
((e->d_name[1] == 0) ||
|
|
|
|
((e->d_name[1] == '.') && e->d_name[2] == 0)))
|
|
|
|
continue; /* "." and ".." */
|
|
|
|
|
|
|
|
namlen = strlen(e->d_name);
|
|
|
|
if ((len + namlen < PATH_MAX) &&
|
|
|
|
strcpy(path + len, e->d_name) &&
|
|
|
|
!lstat(path, &st) &&
|
|
|
|
S_ISDIR(st.st_mode) &&
|
2006-10-01 14:36:49 +02:00
|
|
|
!remove_empty_dir_recursive(path, len + namlen))
|
2006-09-30 11:25:30 +02:00
|
|
|
continue; /* happy */
|
|
|
|
|
|
|
|
/* path too long, stat fails, or non-directory still exists */
|
|
|
|
ret = -1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
closedir(dir);
|
|
|
|
if (!ret) {
|
|
|
|
path[len] = 0;
|
|
|
|
ret = rmdir(path);
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int remove_empty_directories(char *file)
|
|
|
|
{
|
|
|
|
/* we want to create a file but there is a directory there;
|
|
|
|
* if that is an empty directory (or a directory that contains
|
|
|
|
* only empty directories), remove them.
|
|
|
|
*/
|
|
|
|
char path[PATH_MAX];
|
|
|
|
int len = strlen(file);
|
|
|
|
|
|
|
|
if (len >= PATH_MAX) /* path too long ;-) */
|
|
|
|
return -1;
|
|
|
|
strcpy(path, file);
|
|
|
|
return remove_empty_dir_recursive(path, len);
|
|
|
|
}
|
|
|
|
|
2006-10-01 00:02:00 +02:00
|
|
|
static struct ref_lock *lock_ref_sha1_basic(const char *ref, const unsigned char *old_sha1, int *flag)
|
2006-05-17 11:55:02 +02:00
|
|
|
{
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
char *ref_file;
|
2006-09-12 05:17:35 +02:00
|
|
|
const char *orig_ref = ref;
|
2006-05-17 11:55:02 +02:00
|
|
|
struct ref_lock *lock;
|
2006-05-19 09:29:05 +02:00
|
|
|
struct stat st;
|
2006-09-30 23:14:31 +02:00
|
|
|
int last_errno = 0;
|
2006-09-27 10:09:18 +02:00
|
|
|
int mustexist = (old_sha1 && !is_null_sha1(old_sha1));
|
2006-05-17 11:55:02 +02:00
|
|
|
|
|
|
|
lock = xcalloc(1, sizeof(struct ref_lock));
|
|
|
|
lock->lock_fd = -1;
|
|
|
|
|
2006-10-01 00:02:00 +02:00
|
|
|
ref = resolve_ref(ref, lock->old_sha1, mustexist, flag);
|
2006-09-30 11:25:30 +02:00
|
|
|
if (!ref && errno == EISDIR) {
|
|
|
|
/* we are trying to lock foo but we used to
|
|
|
|
* have foo/bar which now does not exist;
|
|
|
|
* it is normal for the empty directory 'foo'
|
|
|
|
* to remain.
|
|
|
|
*/
|
|
|
|
ref_file = git_path("%s", orig_ref);
|
2006-09-30 23:14:31 +02:00
|
|
|
if (remove_empty_directories(ref_file)) {
|
|
|
|
last_errno = errno;
|
|
|
|
error("there are still refs under '%s'", orig_ref);
|
|
|
|
goto error_return;
|
|
|
|
}
|
2006-10-01 00:02:00 +02:00
|
|
|
ref = resolve_ref(orig_ref, lock->old_sha1, mustexist, flag);
|
2006-09-30 11:25:30 +02:00
|
|
|
}
|
2006-09-12 05:17:35 +02:00
|
|
|
if (!ref) {
|
2006-09-30 23:14:31 +02:00
|
|
|
last_errno = errno;
|
2006-07-29 05:44:51 +02:00
|
|
|
error("unable to resolve reference %s: %s",
|
2006-09-12 05:17:35 +02:00
|
|
|
orig_ref, strerror(errno));
|
2006-09-30 23:14:31 +02:00
|
|
|
goto error_return;
|
2006-05-17 11:55:02 +02:00
|
|
|
}
|
2006-09-30 23:19:25 +02:00
|
|
|
if (is_null_sha1(lock->old_sha1)) {
|
|
|
|
/* The ref did not exist and we are creating it.
|
|
|
|
* Make sure there is no existing ref that is packed
|
|
|
|
* whose name begins with our refname, nor a ref whose
|
|
|
|
* name is a proper prefix of our refname.
|
|
|
|
*/
|
|
|
|
int namlen = strlen(ref); /* e.g. 'foo/bar' */
|
|
|
|
struct ref_list *list = get_packed_refs();
|
|
|
|
while (list) {
|
|
|
|
/* list->name could be 'foo' or 'foo/bar/baz' */
|
|
|
|
int len = strlen(list->name);
|
|
|
|
int cmplen = (namlen < len) ? namlen : len;
|
|
|
|
const char *lead = (namlen < len) ? list->name : ref;
|
|
|
|
|
|
|
|
if (!strncmp(ref, list->name, cmplen) &&
|
|
|
|
lead[cmplen] == '/') {
|
|
|
|
error("'%s' exists; cannot create '%s'",
|
|
|
|
list->name, ref);
|
|
|
|
goto error_return;
|
|
|
|
}
|
|
|
|
list = list->next;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-06-06 22:54:14 +02:00
|
|
|
lock->lk = xcalloc(1, sizeof(struct lock_file));
|
2006-05-17 11:55:02 +02:00
|
|
|
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
lock->ref_name = xstrdup(ref);
|
2006-09-12 05:17:35 +02:00
|
|
|
lock->log_file = xstrdup(git_path("logs/%s", ref));
|
2006-09-23 01:08:45 +02:00
|
|
|
ref_file = git_path("%s", ref);
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
lock->force_write = lstat(ref_file, &st) && errno == ENOENT;
|
2006-05-17 11:55:02 +02:00
|
|
|
|
2006-09-30 23:14:31 +02:00
|
|
|
if (safe_create_leading_directories(ref_file)) {
|
|
|
|
last_errno = errno;
|
|
|
|
error("unable to create directory for %s", ref_file);
|
|
|
|
goto error_return;
|
|
|
|
}
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
lock->lock_fd = hold_lock_file_for_update(lock->lk, ref_file, 1);
|
2006-05-17 11:55:02 +02:00
|
|
|
|
|
|
|
return old_sha1 ? verify_lock(lock, old_sha1, mustexist) : lock;
|
2006-09-30 23:14:31 +02:00
|
|
|
|
|
|
|
error_return:
|
|
|
|
unlock_ref(lock);
|
|
|
|
errno = last_errno;
|
|
|
|
return NULL;
|
2006-05-17 11:55:02 +02:00
|
|
|
}
|
|
|
|
|
2006-09-27 10:09:18 +02:00
|
|
|
struct ref_lock *lock_ref_sha1(const char *ref, const unsigned char *old_sha1)
|
2005-06-06 22:31:29 +02:00
|
|
|
{
|
2006-09-19 22:58:23 +02:00
|
|
|
char refpath[PATH_MAX];
|
2005-06-06 22:31:29 +02:00
|
|
|
if (check_ref_format(ref))
|
2006-05-17 11:55:02 +02:00
|
|
|
return NULL;
|
2006-09-19 22:58:23 +02:00
|
|
|
strcpy(refpath, mkpath("refs/%s", ref));
|
2006-10-01 00:02:00 +02:00
|
|
|
return lock_ref_sha1_basic(refpath, old_sha1, NULL);
|
2006-05-17 11:55:02 +02:00
|
|
|
}
|
|
|
|
|
2006-09-27 10:09:18 +02:00
|
|
|
struct ref_lock *lock_any_ref_for_update(const char *ref, const unsigned char *old_sha1)
|
2006-05-17 11:55:02 +02:00
|
|
|
{
|
2006-10-01 00:02:00 +02:00
|
|
|
return lock_ref_sha1_basic(ref, old_sha1, NULL);
|
|
|
|
}
|
|
|
|
|
2006-10-01 20:41:00 +02:00
|
|
|
static struct lock_file packlock;
|
|
|
|
|
2006-10-01 00:02:00 +02:00
|
|
|
static int repack_without_ref(const char *refname)
|
|
|
|
{
|
|
|
|
struct ref_list *list, *packed_ref_list;
|
|
|
|
int fd;
|
|
|
|
int found = 0;
|
|
|
|
|
|
|
|
packed_ref_list = get_packed_refs();
|
|
|
|
for (list = packed_ref_list; list; list = list->next) {
|
|
|
|
if (!strcmp(refname, list->name)) {
|
|
|
|
found = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!found)
|
|
|
|
return 0;
|
|
|
|
memset(&packlock, 0, sizeof(packlock));
|
|
|
|
fd = hold_lock_file_for_update(&packlock, git_path("packed-refs"), 0);
|
|
|
|
if (fd < 0)
|
|
|
|
return error("cannot delete '%s' from packed refs", refname);
|
|
|
|
|
|
|
|
for (list = packed_ref_list; list; list = list->next) {
|
|
|
|
char line[PATH_MAX + 100];
|
|
|
|
int len;
|
|
|
|
|
|
|
|
if (!strcmp(refname, list->name))
|
|
|
|
continue;
|
|
|
|
len = snprintf(line, sizeof(line), "%s %s\n",
|
|
|
|
sha1_to_hex(list->sha1), list->name);
|
|
|
|
/* this should not happen but just being defensive */
|
|
|
|
if (len > sizeof(line))
|
|
|
|
die("too long a refname '%s'", list->name);
|
|
|
|
write_or_die(fd, line, len);
|
|
|
|
}
|
|
|
|
return commit_lock_file(&packlock);
|
|
|
|
}
|
|
|
|
|
|
|
|
int delete_ref(const char *refname, unsigned char *sha1)
|
|
|
|
{
|
|
|
|
struct ref_lock *lock;
|
|
|
|
int err, i, ret = 0, flag = 0;
|
|
|
|
|
|
|
|
lock = lock_ref_sha1_basic(refname, sha1, &flag);
|
|
|
|
if (!lock)
|
|
|
|
return 1;
|
|
|
|
if (!(flag & REF_ISPACKED)) {
|
|
|
|
/* loose */
|
|
|
|
i = strlen(lock->lk->filename) - 5; /* .lock */
|
|
|
|
lock->lk->filename[i] = 0;
|
|
|
|
err = unlink(lock->lk->filename);
|
|
|
|
if (err) {
|
|
|
|
ret = 1;
|
|
|
|
error("unlink(%s) failed: %s",
|
|
|
|
lock->lk->filename, strerror(errno));
|
|
|
|
}
|
|
|
|
lock->lk->filename[i] = '.';
|
|
|
|
}
|
|
|
|
/* removing the loose one could have resurrected an earlier
|
|
|
|
* packed one. Also, if it was not loose we need to repack
|
|
|
|
* without it.
|
|
|
|
*/
|
|
|
|
ret |= repack_without_ref(refname);
|
|
|
|
|
|
|
|
err = unlink(lock->log_file);
|
|
|
|
if (err && errno != ENOENT)
|
|
|
|
fprintf(stderr, "warning: unlink(%s) failed: %s",
|
|
|
|
lock->log_file, strerror(errno));
|
|
|
|
invalidate_cached_refs();
|
|
|
|
unlock_ref(lock);
|
|
|
|
return ret;
|
2006-05-17 11:55:02 +02:00
|
|
|
}
|
|
|
|
|
2006-06-06 23:04:17 +02:00
|
|
|
void unlock_ref(struct ref_lock *lock)
|
2006-05-17 11:55:02 +02:00
|
|
|
{
|
|
|
|
if (lock->lock_fd >= 0) {
|
|
|
|
close(lock->lock_fd);
|
2006-06-06 22:54:14 +02:00
|
|
|
/* Do not free lock->lk -- atexit() still looks at them */
|
|
|
|
if (lock->lk)
|
|
|
|
rollback_lock_file(lock->lk);
|
2006-05-17 11:55:02 +02:00
|
|
|
}
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
free(lock->ref_name);
|
2006-08-28 06:19:39 +02:00
|
|
|
free(lock->log_file);
|
2006-05-17 11:55:02 +02:00
|
|
|
free(lock);
|
|
|
|
}
|
|
|
|
|
2006-05-17 11:55:40 +02:00
|
|
|
static int log_ref_write(struct ref_lock *lock,
|
|
|
|
const unsigned char *sha1, const char *msg)
|
|
|
|
{
|
|
|
|
int logfd, written, oflags = O_APPEND | O_WRONLY;
|
|
|
|
unsigned maxlen, len;
|
|
|
|
char *logrec;
|
2006-07-09 11:36:24 +02:00
|
|
|
const char *committer;
|
2006-05-17 11:55:40 +02:00
|
|
|
|
|
|
|
if (log_all_ref_updates) {
|
|
|
|
if (safe_create_leading_directories(lock->log_file) < 0)
|
|
|
|
return error("unable to create directory for %s",
|
|
|
|
lock->log_file);
|
|
|
|
oflags |= O_CREAT;
|
|
|
|
}
|
|
|
|
|
|
|
|
logfd = open(lock->log_file, oflags, 0666);
|
|
|
|
if (logfd < 0) {
|
|
|
|
if (!log_all_ref_updates && errno == ENOENT)
|
|
|
|
return 0;
|
|
|
|
return error("Unable to append to %s: %s",
|
|
|
|
lock->log_file, strerror(errno));
|
|
|
|
}
|
|
|
|
|
2006-07-09 11:36:24 +02:00
|
|
|
committer = git_committer_info(1);
|
2006-05-17 11:55:40 +02:00
|
|
|
if (msg) {
|
2006-07-09 11:36:24 +02:00
|
|
|
maxlen = strlen(committer) + strlen(msg) + 2*40 + 5;
|
2006-05-17 11:55:40 +02:00
|
|
|
logrec = xmalloc(maxlen);
|
|
|
|
len = snprintf(logrec, maxlen, "%s %s %s\t%s\n",
|
|
|
|
sha1_to_hex(lock->old_sha1),
|
|
|
|
sha1_to_hex(sha1),
|
2006-07-09 11:36:24 +02:00
|
|
|
committer,
|
2006-05-17 11:55:40 +02:00
|
|
|
msg);
|
2006-06-06 23:04:17 +02:00
|
|
|
}
|
|
|
|
else {
|
2006-07-09 11:36:24 +02:00
|
|
|
maxlen = strlen(committer) + 2*40 + 4;
|
2006-05-17 11:55:40 +02:00
|
|
|
logrec = xmalloc(maxlen);
|
|
|
|
len = snprintf(logrec, maxlen, "%s %s %s\n",
|
|
|
|
sha1_to_hex(lock->old_sha1),
|
|
|
|
sha1_to_hex(sha1),
|
2006-07-09 11:36:24 +02:00
|
|
|
committer);
|
2006-05-17 11:55:40 +02:00
|
|
|
}
|
|
|
|
written = len <= maxlen ? write(logfd, logrec, len) : -1;
|
|
|
|
free(logrec);
|
|
|
|
close(logfd);
|
|
|
|
if (written != len)
|
|
|
|
return error("Unable to append to %s", lock->log_file);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-05-17 11:55:02 +02:00
|
|
|
int write_ref_sha1(struct ref_lock *lock,
|
|
|
|
const unsigned char *sha1, const char *logmsg)
|
|
|
|
{
|
|
|
|
static char term = '\n';
|
|
|
|
|
|
|
|
if (!lock)
|
2005-06-06 22:31:29 +02:00
|
|
|
return -1;
|
2006-08-17 20:54:57 +02:00
|
|
|
if (!lock->force_write && !hashcmp(lock->old_sha1, sha1)) {
|
2006-05-17 11:55:02 +02:00
|
|
|
unlock_ref(lock);
|
|
|
|
return 0;
|
2005-06-06 22:31:29 +02:00
|
|
|
}
|
2006-05-17 11:55:02 +02:00
|
|
|
if (write(lock->lock_fd, sha1_to_hex(sha1), 40) != 40 ||
|
|
|
|
write(lock->lock_fd, &term, 1) != 1
|
|
|
|
|| close(lock->lock_fd) < 0) {
|
2006-06-06 22:54:14 +02:00
|
|
|
error("Couldn't write %s", lock->lk->filename);
|
2006-05-17 11:55:02 +02:00
|
|
|
unlock_ref(lock);
|
|
|
|
return -1;
|
|
|
|
}
|
2006-09-30 21:37:37 +02:00
|
|
|
invalidate_cached_refs();
|
2006-05-17 11:55:40 +02:00
|
|
|
if (log_ref_write(lock, sha1, logmsg) < 0) {
|
|
|
|
unlock_ref(lock);
|
|
|
|
return -1;
|
|
|
|
}
|
2006-06-06 22:54:14 +02:00
|
|
|
if (commit_lock_file(lock->lk)) {
|
Enable the packed refs file format
This actually "turns on" the packed ref file format, now that the
infrastructure to do so sanely exists (ie notably the change to make the
reference reading logic take refnames rather than pathnames to the loose
objects that no longer necessarily even exist).
In particular, when the ref lookup hits a refname that has no loose file
associated with it, it falls back on the packed-ref information. Also, the
ref-locking code, while still using a loose file for the locking itself
(and _creating_ a loose file for the new ref) no longer requires that the
old ref be in such an unpacked state.
Finally, this does a minimal hack to git-checkout.sh to rather than check
the ref-file directly, do a "git-rev-parse" on the "heads/$refname".
That's not really wonderful - we should rather really have a special
routine to verify the names as proper branch head names, but it is a
workable solution for now.
With this, I can literally do something like
git pack-refs
find .git/refs -type f -print0 | xargs -0 rm -f --
and the end result is a largely working repository (ie I've done two
commits - which creates _one_ unpacked ref file - done things like run
"gitk" and "git log" etc, and it all looks ok).
There are probably things missing, but I'm hoping that the missing things
are now of the "small and obvious" kind, and that somebody else might want
to start looking at this too. Hint hint ;)
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-14 19:14:47 +02:00
|
|
|
error("Couldn't set %s", lock->ref_name);
|
2006-05-17 11:55:02 +02:00
|
|
|
unlock_ref(lock);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
lock->lock_fd = -1;
|
|
|
|
unlock_ref(lock);
|
|
|
|
return 0;
|
2005-06-06 22:31:29 +02:00
|
|
|
}
|
2006-05-17 11:56:09 +02:00
|
|
|
|
2006-10-06 08:16:15 +02:00
|
|
|
int read_ref_at(const char *ref, unsigned long at_time, int cnt, unsigned char *sha1)
|
2006-05-17 11:56:09 +02:00
|
|
|
{
|
2006-05-19 09:28:19 +02:00
|
|
|
const char *logfile, *logdata, *logend, *rec, *lastgt, *lastrec;
|
2006-05-17 11:56:09 +02:00
|
|
|
char *tz_c;
|
|
|
|
int logfd, tz;
|
|
|
|
struct stat st;
|
|
|
|
unsigned long date;
|
2006-05-19 09:28:19 +02:00
|
|
|
unsigned char logged_sha1[20];
|
2006-05-17 11:56:09 +02:00
|
|
|
|
|
|
|
logfile = git_path("logs/%s", ref);
|
|
|
|
logfd = open(logfile, O_RDONLY, 0);
|
|
|
|
if (logfd < 0)
|
|
|
|
die("Unable to read log %s: %s", logfile, strerror(errno));
|
|
|
|
fstat(logfd, &st);
|
|
|
|
if (!st.st_size)
|
|
|
|
die("Log %s is empty.", logfile);
|
|
|
|
logdata = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, logfd, 0);
|
|
|
|
close(logfd);
|
|
|
|
|
2006-05-19 09:28:19 +02:00
|
|
|
lastrec = NULL;
|
2006-05-17 11:56:09 +02:00
|
|
|
rec = logend = logdata + st.st_size;
|
|
|
|
while (logdata < rec) {
|
|
|
|
if (logdata < rec && *(rec-1) == '\n')
|
|
|
|
rec--;
|
2006-05-19 09:28:19 +02:00
|
|
|
lastgt = NULL;
|
|
|
|
while (logdata < rec && *(rec-1) != '\n') {
|
2006-05-17 11:56:09 +02:00
|
|
|
rec--;
|
2006-05-19 09:28:19 +02:00
|
|
|
if (*rec == '>')
|
|
|
|
lastgt = rec;
|
|
|
|
}
|
|
|
|
if (!lastgt)
|
2006-05-17 11:56:09 +02:00
|
|
|
die("Log %s is corrupt.", logfile);
|
2006-05-19 09:28:19 +02:00
|
|
|
date = strtoul(lastgt + 1, &tz_c, 10);
|
2006-10-06 08:16:15 +02:00
|
|
|
if (date <= at_time || cnt == 0) {
|
2006-05-19 09:28:19 +02:00
|
|
|
if (lastrec) {
|
|
|
|
if (get_sha1_hex(lastrec, logged_sha1))
|
|
|
|
die("Log %s is corrupt.", logfile);
|
|
|
|
if (get_sha1_hex(rec + 41, sha1))
|
|
|
|
die("Log %s is corrupt.", logfile);
|
2006-08-17 20:54:57 +02:00
|
|
|
if (hashcmp(logged_sha1, sha1)) {
|
2006-05-19 09:28:19 +02:00
|
|
|
tz = strtoul(tz_c, NULL, 10);
|
|
|
|
fprintf(stderr,
|
|
|
|
"warning: Log %s has gap after %s.\n",
|
|
|
|
logfile, show_rfc2822_date(date, tz));
|
|
|
|
}
|
2006-06-06 23:04:17 +02:00
|
|
|
}
|
|
|
|
else if (date == at_time) {
|
2006-05-19 09:28:19 +02:00
|
|
|
if (get_sha1_hex(rec + 41, sha1))
|
|
|
|
die("Log %s is corrupt.", logfile);
|
2006-06-06 23:04:17 +02:00
|
|
|
}
|
|
|
|
else {
|
2006-05-19 09:28:19 +02:00
|
|
|
if (get_sha1_hex(rec + 41, logged_sha1))
|
|
|
|
die("Log %s is corrupt.", logfile);
|
2006-08-17 20:54:57 +02:00
|
|
|
if (hashcmp(logged_sha1, sha1)) {
|
2006-05-19 09:28:19 +02:00
|
|
|
tz = strtoul(tz_c, NULL, 10);
|
|
|
|
fprintf(stderr,
|
|
|
|
"warning: Log %s unexpectedly ended on %s.\n",
|
|
|
|
logfile, show_rfc2822_date(date, tz));
|
|
|
|
}
|
|
|
|
}
|
2006-05-17 11:56:09 +02:00
|
|
|
munmap((void*)logdata, st.st_size);
|
|
|
|
return 0;
|
|
|
|
}
|
2006-05-19 09:28:19 +02:00
|
|
|
lastrec = rec;
|
2006-10-06 08:16:15 +02:00
|
|
|
if (cnt > 0)
|
|
|
|
cnt--;
|
2006-05-17 11:56:09 +02:00
|
|
|
}
|
|
|
|
|
2006-05-19 09:28:19 +02:00
|
|
|
rec = logdata;
|
|
|
|
while (rec < logend && *rec != '>' && *rec != '\n')
|
|
|
|
rec++;
|
|
|
|
if (rec == logend || *rec == '\n')
|
2006-05-17 11:56:09 +02:00
|
|
|
die("Log %s is corrupt.", logfile);
|
2006-05-19 09:28:19 +02:00
|
|
|
date = strtoul(rec + 1, &tz_c, 10);
|
2006-05-17 11:56:09 +02:00
|
|
|
tz = strtoul(tz_c, NULL, 10);
|
|
|
|
if (get_sha1_hex(logdata, sha1))
|
|
|
|
die("Log %s is corrupt.", logfile);
|
|
|
|
munmap((void*)logdata, st.st_size);
|
|
|
|
fprintf(stderr, "warning: Log %s only goes back to %s.\n",
|
|
|
|
logfile, show_rfc2822_date(date, tz));
|
|
|
|
return 0;
|
|
|
|
}
|