Mon, 16 Mar 2015 13:25:19 -0500 crecord: fiddle with vertical whitespace
Matt Mackall <mpm@selenic.com> [Mon, 16 Mar 2015 13:25:19 -0500] rev 24317
crecord: fiddle with vertical whitespace
Mon, 16 Mar 2015 13:23:42 -0500 crecord: fix underbar style for orig_stdout
Matt Mackall <mpm@selenic.com> [Mon, 16 Mar 2015 13:23:42 -0500] rev 24316
crecord: fix underbar style for orig_stdout
Mon, 16 Mar 2015 13:22:37 -0500 crecord: use encoding.ucolwidth
Matt Mackall <mpm@selenic.com> [Mon, 16 Mar 2015 13:22:37 -0500] rev 24315
crecord: use encoding.ucolwidth
Mon, 16 Mar 2015 12:58:06 -0500 crecord: more import style
Matt Mackall <mpm@selenic.com> [Mon, 16 Mar 2015 12:58:06 -0500] rev 24314
crecord: more import style
Mon, 16 Mar 2015 12:55:21 -0500 crecord: fix some import style
Matt Mackall <mpm@selenic.com> [Mon, 16 Mar 2015 12:55:21 -0500] rev 24313
crecord: fix some import style
Fri, 13 Mar 2015 22:50:40 -0400 manifest.c: ensure realloc_if_full() returns 1 or 0
Matt Harbison <matt_harbison@yahoo.com> [Fri, 13 Mar 2015 22:50:40 -0400] rev 24312
manifest.c: ensure realloc_if_full() returns 1 or 0 This fixes an MSVC 2008 warning that I don't see with gcc 4.6.3-2: warning C4047: 'return' : 'bool' differs in levels of indirection from 'line *' More importantly, the truncation from pointer to 'unsigned char' would have returned 0 if self->lines pointed to an address divisible by 0xFF, which causes find_lines() to return MANIFEST_OOM. I was able to cause this to happen in a trivial program with the gcc compiler.
(0) -10000 -3000 -1000 -300 -100 -30 -10 -6 +6 +10 +30 +100 +300 +1000 +3000 +10000 tip