|
From: | risc |
Subject: | Re: [help-GIFT] Swallowed Mails... have to resend |
Date: | Mon, 6 Nov 2006 14:14:16 -0600 |
User-agent: | Mutt/1.4.1i |
ARGH. need to remember to attach patches. Julia Longtin <address@hidden> On Mon, Nov 06, 2006 at 02:09:41PM -0600, address@hidden wrote: > Wolfgang, > > I've successfully reproduced the issue here, will get to work on it > immediately. > > That said, I've generated 24 patches as a result of getting gift to > compile on my AMD64 debian unstable box. > > Here are 19 of those patches (the other 5 have to do with embedded > perl ,and dont constitute a "complete fix".) > > I dont concider all of these commit worthy, i'll say that to start > with. specifically, I tried and suceeded at fixing three problems: > 64 bit specific linking errors, > 64 bit errors in general, and > ability to build where top_builddir != top_srcdir. > (EG: ./bootstrap-cvs.sh && mkdir build && cd build && ../configure && make) > > a lot of the builddir patches just go to show us where the include > path starts at top_srcdir. thats a deeper problem, i just included > top_srcdir, and dont concider that a "complete fix". it just happens > to work. > > the linking problems boil down to "use .la objects to link to, not .a" > > AND i'm not satisfied with my fix for the 64bit cast issue i showed > in an earlier mail: it works on 64 and 32, but.. feels Cish, not C++ish. > > SO, david, wolfgang, > > these patches are for your eyes. they're ugly. i dont like them, but > they make gift compile on my box. please show me a better way? :) > > Julia Longtin <address@hidden> > > On Sat, Nov 04, 2006 at 11:26:52AM +0100, Wolfgang Mueller wrote: > > Hi, David, Julia, > > I just noted that a mail of mine got "swallowed" last week. Apparently > > it did not reach the list. I tested the current feature extraction on my > > machine and got segfaults, as some array index in the current version of > > Gabor filtering get negative --> VERY big. Using GIFT with old features > > and current code worked on my machine after having fixed some file:/ vs. > > file:/// URL annoyances. > > > > I will resend the more detailed mails in the beginning of the next week. > > > > Cheers, > > Wolfgang > > -- > > Dr. Wolfgang Mueller > > LS Medieninformatik > > Universitaet Bamberg > > > > > > _______________________________________________ > > help-GIFT mailing list > > address@hidden > > http://lists.gnu.org/mailman/listinfo/help-gift
00-bootstrap-cvs_sh-support_aclocal_1_7.patch
Description: Text document
00-configure_in-cleanup_warning.patch
Description: Text document
01-configure_in-make_build_in_subdir.patch
Description: Text document
01-dtd_Makefile_am-make_build_in_subdir.patch
Description: Text document
02-libMRML_cc_CAttributeList_cc-fix_64bit_cast.patch
Description: Text document
03-FeatureExtraction_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-GIFTServer_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTAcDistanceMatrix_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTAcHierarchy_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTAcInvertedFile_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTAcPerl_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTAcURL2FTS_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTQuInvertedFile_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTQuHierarchy_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libGIFTQuPerl_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
03-libMRML_cc_Makefile_am-make_build_in_subdir.patch
Description: Text document
04-libGIFTAcDistanceMatrix_cc_Makefile_am-use_portable_library.patch
Description: Text document
04-libGIFTAcInvertedFile_cc_Makefile_am-use_portable_library.patch
Description: Text document
04-libGIFTQuInvertedFile_cc_Makefile_am-use_portable_library.patch
Description: Text document
[Prev in Thread] | Current Thread | [Next in Thread] |