# Re: [PATCH 3/3] mimic unpack-objects when --stdin is used with index-pack 3 messages from 2006-10-26 to 2006-10-26. Participants: Junio C Hamano, Nicolas Pitre. Thread: https://gitlist.dev/t/43091 ## Nicolas Pitre, 2006-10-26 03:31 Subject: [PATCH 3/3] mimic unpack-objects when --stdin is used with index-pack Message-ID: URL: https://gitlist.dev/e/Pine.LNX.4.64.0610252330320.12418%40xanadu.home ``` It appears that git-unpack-objects writes the last part of the input buffer to stdout after the pack has been parsed. This looks a bit suspicious since the last fill() might have filled the buffer up to the 4096 byte limit and more data might still be pending on stdin, but since this is about being a drop-in replacement for unpack-objects let's simply duplicate the same behavior for now. Signed-off-by: Nicolas Pitre --- index-pack.c | 17 +++++++++++++++-- 1 files changed, 15 insertions(+), 2 deletions(-) diff --git a/index-pack.c b/index-pack.c index 2046b37..7f7dc5d 100644 --- a/index-pack.c +++ b/index-pack.c @@ -456,6 +456,7 @@ static void parse_pack_objects(unsigned SHA1_Final(sha1, &input_ctx); if (hashcmp(fill(20), sha1)) die("pack is corrupted (SHA1 mismatch)"); + use(20); /* If input_fd is a file, we should have reached its end now. */ if (fstat(input_fd, &st)) @@ -765,6 +766,18 @@ static void final(const char *final_pack if (err) die("error while closing pack file: %s", strerror(errno)); chmod(curr_pack_name, 0444); + + /* + * Let's just mimic git-unpack-objects here and write + * the last part of the buffer to stdout. + */ + while (input_len) { + err = xwrite(1, input_buffer + input_offset, input_len); + if (err <= 0) + break; + input_len -= err; + input_offset += err; + } } if (final_pack_name != curr_pack_name) { @@ -863,7 +876,6 @@ int main(int argc, char **argv) nr_deltas - nr_resolved_deltas); } else { /* Flush remaining pack final 20-byte SHA1. */ - use(20); flush(); } free(deltas); @@ -872,7 +884,8 @@ int main(int argc, char **argv) free(objects); free(index_name_buf); - printf("%s\n", sha1_to_hex(sha1)); + if (!from_stdin) + printf("%s\n", sha1_to_hex(sha1)); return 0; } -- 1.4.3.3.g10cf-dirty ``` ## Junio C Hamano, 2006-10-26 07:55 Subject: Re: [PATCH 3/3] mimic unpack-objects when --stdin is used with index-pack Message-ID: <7vmz7jqzoo.fsf@assigned-by-dhcp.cox.net> URL: https://gitlist.dev/e/7vmz7jqzoo.fsf%40assigned-by-dhcp.cox.net In-Reply-To: ``` Nicolas Pitre writes: > It appears that git-unpack-objects writes the last part of the input > buffer to stdout after the pack has been parsed. This looks a bit > suspicious since the last fill() might have filled the buffer up to > the 4096 byte limit and more data might still be pending on stdin, > but since this is about being a drop-in replacement for unpack-objects > let's simply duplicate the same behavior for now. This seems to break t5300 when applied on top of everything else. The other two numbered patches are Ok. ``` ## Nicolas Pitre, 2006-10-26 13:46 Subject: Re: [PATCH 3/3] mimic unpack-objects when --stdin is used with index-pack Message-ID: URL: https://gitlist.dev/e/Pine.LNX.4.64.0610260940020.12418%40xanadu.home In-Reply-To: <7vmz7jqzoo.fsf@assigned-by-dhcp.cox.net> ``` On Thu, 26 Oct 2006, Junio C Hamano wrote: > Nicolas Pitre writes: > > > It appears that git-unpack-objects writes the last part of the input > > buffer to stdout after the pack has been parsed. This looks a bit > > suspicious since the last fill() might have filled the buffer up to > > the 4096 byte limit and more data might still be pending on stdin, > > but since this is about being a drop-in replacement for unpack-objects > > let's simply duplicate the same behavior for now. > > This seems to break t5300 when applied on top of everything > else. The other two numbered patches are Ok. Yeah.... actually the problem appeared to me in my sleep! (Worrisome isn't it?) So you just need to amend this patch with: diff --git a/index-pack.c b/index-pack.c index 7f7dc5d..5c747a6 100644 --- a/index-pack.c +++ b/index-pack.c @@ -461,7 +461,7 @@ static void parse_pack_objects(unsigned /* If input_fd is a file, we should have reached its end now. */ if (fstat(input_fd, &st)) die("cannot fstat packfile: %s", strerror(errno)); - if (S_ISREG(st.st_mode) && st.st_size != consumed_bytes + 20) + if (S_ISREG(st.st_mode) && st.st_size != consumed_bytes) die("pack has junk at the end"); if (!nr_deltas) ```