On Dec 11, 2009, at 10:29 AM, Neil Conway wrote: > On Fri, Dec 11, 2009 at 10:26 AM, Sander Temme <[email protected]> wrote: >> Not sure How I'd vote given what's below and the fact that I haven no cycles >> to spend. >> >> Mac OS X 10.6.2 Snow Leopard: >> >> Darwin legadema.sandla.org 10.2.0 Darwin Kernel Version 10.2.0: Tue Nov 3 >> 10:37:10 PST 2009; root:xnu-1486.2.11~1/RELEASE_I386 i386 >> >> [scte...@legadema] test $ ./testall -v testfmt >> testfmt : \Line 52: expected <0>, but saw <%ld> >> FAILED 1 of 9 >> Failed Tests Total Fail Failed % >> =================================================== >> testfmt 9 1 11.11% > > Presumably this is the same problem I reported a few months ago: > > http://www.mail-archive.com/[email protected]/msg22133.html
Yes it is. I have a couple of questions. In r809745, Jimjag set the format tag for apr_off_t to "lld" on Snow Leopard. What was wrong with just using APR_INT64_T_FMT? Also, while "lld" is a possible outcome in several spots in configure.in, where in apr_snprintf.c is that actually supported? I see in apr_vformatter() starting on line 683 (trunk) that we walk the format string, and when we hit a tag, we pass the possibility of an 'l' modifier once and then move on to the actual argument. The logic around 816 doesn't help us in this case, because APR_INT64_T_FMT is "ld" on this platform. Unless my caffeine-starved brain is reading this wrong, should we put in a special case in the 'l' modifier (line 832) that catches 'll' and sets the variable type to IS_QUAD? Losing that Per OS tuning... section makes the testfmt test pass. Why is it needed? Thanks, S. -- [email protected] http://www.temme.net/sander/ PGP FP: 51B4 8727 466A 0BC3 69F4 B7B8 B2BE BC40 1529 24AF
smime.p7s
Description: S/MIME cryptographic signature
