aboutsummaryrefslogtreecommitdiff
path: root/fast-import.c
diff options
context:
space:
mode:
authorShawn O. Pearce <spearce@spearce.org>2006-08-14 02:50:18 -0400
committerShawn O. Pearce <spearce@spearce.org>2007-01-14 02:15:03 -0500
commit7111feede9c5905199ba48645fadc369faca5711 (patch)
treecf74fa5c51824b0293396788381df47c538daec9 /fast-import.c
parent463acbe1c60fc5009dac9d033df6c2b9c5037a91 (diff)
downloadgit-7111feede9c5905199ba48645fadc369faca5711.tar.gz
git-7111feede9c5905199ba48645fadc369faca5711.tar.xz
Implement blob ID validation in fast-import.
When accepting revision SHA1 IDs from the frontend verify the SHA1 actually refers to a blob and is known to exist. Its an error to use a SHA1 in a tree if the blob doesn't exist as this would cause git-fsck-objects to report a missing blob should the pack get closed without the blob being appended into it or a subsequent pack. So right now we'll just ask that the frontend "pre-declare" any blobs it wants to use in a tree before it can use them. Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Diffstat (limited to 'fast-import.c')
-rw-r--r--fast-import.c28
1 files changed, 24 insertions, 4 deletions
diff --git a/fast-import.c b/fast-import.c
index 4605b7469..95b84f57e 100644
--- a/fast-import.c
+++ b/fast-import.c
@@ -70,6 +70,7 @@ Format of STDIN stream:
struct object_entry
{
struct object_entry *next;
+ enum object_type type;
unsigned long offset;
unsigned char sha1[20];
};
@@ -528,6 +529,7 @@ static int store_object(
duplicate_count_by_type[type]++;
return 1;
}
+ e->type = type;
e->offset = pack_offset;
object_count++;
object_count_by_type[type]++;
@@ -713,7 +715,7 @@ static int tree_content_set(
}
if (!S_ISDIR(e->mode)) {
e->tree = new_tree_content(8);
- e->mode = 040000;
+ e->mode = S_IFDIR;
}
if (!e->tree)
load_tree(e);
@@ -732,7 +734,7 @@ static int tree_content_set(
t->entries[t->entry_count++] = e;
if (slash1) {
e->tree = new_tree_content(8);
- e->mode = 040000;
+ e->mode = S_IFDIR;
tree_content_set(e, slash1 + 1, sha1, mode);
} else {
e->tree = NULL;
@@ -948,16 +950,28 @@ static void load_branch(struct branch *b)
static void file_change_m(struct branch *b)
{
const char *path = read_path();
+ struct object_entry *oe;
char hexsha1[41];
unsigned char sha1[20];
+ char type[20];
yread(0, hexsha1, 40);
hexsha1[40] = 0;
if (get_sha1_hex(hexsha1, sha1))
die("Invalid sha1 %s for %s", hexsha1, path);
+ oe = find_object(sha1);
+ if (oe) {
+ if (oe->type != OBJ_BLOB)
+ die("%s is a %s not a blob (for %s)", hexsha1, type_names[oe->type], path);
+ } else {
+ if (sha1_object_info(sha1, type, NULL))
+ die("No blob %s for %s", hexsha1, path);
+ if (strcmp(blob_type, type))
+ die("%s is a %s not a blob (for %s)", hexsha1, type, path);
+ }
- tree_content_set(&b->branch_tree, path, sha1, 0100644);
+ tree_content_set(&b->branch_tree, path, sha1, S_IFREG | 0644);
}
static void file_change_d(struct branch *b)
@@ -986,6 +1000,10 @@ static void cmd_new_commit()
c = body + max_hdr_len;
yread(0, c, acmsglen);
+ /* oddly enough this is all that fsck-objects cares about */
+ if (memcmp(c, "author ", 7))
+ die("Invalid commit format on branch %s", name);
+
/* file_change* */
for (;;) {
unsigned char cmd;
@@ -1104,7 +1122,9 @@ int main(int argc, const char **argv)
fprintf(stderr, " tags : %10lu (%10lu duplicates)\n", object_count_by_type[OBJ_TAG], duplicate_count_by_type[OBJ_TAG]);
fprintf(stderr, "Total branches: %10lu\n", branch_count);
fprintf(stderr, "Total atoms: %10u\n", atom_cnt);
- fprintf(stderr, "Memory pools: %10lu MiB\n", total_allocd/(1024*1024));
+ fprintf(stderr, "Memory total: %10lu KiB\n", (total_allocd + alloc_count*sizeof(struct object_entry))/1024);
+ fprintf(stderr, " pools: %10lu KiB\n", total_allocd/1024);
+ fprintf(stderr, " objects: %10lu KiB\n", (alloc_count*sizeof(struct object_entry))/1024);
fprintf(stderr, "---------------------------------------------------\n");
stat(pack_name, &sb);