aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* - chdir "t"Pascal Rigaux2007-01-151-14/+16
| | | | | - remove temporary "headers/"
* work in directory "t"Pascal Rigaux2007-01-151-2/+3
|
* move test-rpm.spec into t/Pascal Rigaux2007-01-154-3/+4
|
* - use our own tmppath (Patrice Dumas)Pascal Rigaux2007-01-151-2/+5
| | | | | - remove temporary build directories
* perl-URPM.spec is no more bundled with dist, removing "rpm" and "srpm" targetsPascal Rigaux2007-01-151-15/+1
|
* better debug text1.54Pascal Rigaux2007-01-091-2/+2
|
* @chosen is always non empty, and @chosen == 1 is already handledPascal Rigaux2007-01-091-1/+1
|
* use a sort to make things clearPascal Rigaux2007-01-091-20/+15
|
* factorizePascal Rigaux2007-01-091-12/+10
|
* factorize code into _set_flag_installed_and_upgrade_if_no_newer()Pascal Rigaux2007-01-091-31/+21
|
* suggest a valid reason why a package was not foundPascal Rigaux2007-01-091-1/+1
|
* fix previous commitPascal Rigaux2007-01-091-4/+4
|
* create _id_to_name() and use itPascal Rigaux2007-01-091-12/+37
|
* allow toggling debugging and handling it through a callbackPascal Rigaux2007-01-091-14/+28
|
* - add missing files to the tarball to allow make testPascal Rigaux2007-01-081-1/+1
| | | | | - get rids of some "used of uninitialized value"
* updatePascal Rigaux2007-01-081-1/+1
|
* use SvTRUE() instead of SvIV() for boolean values 1.53Pascal Rigaux2007-01-081-7/+7
| | | | | (this gets rid of: Argument "" isn't numeric in subroutine entry at blib/lib/URPM/Build.pm line 68.)
* add to the tarball the files needed for "make test"Pascal Rigaux2007-01-081-0/+2
|
* 1.521.52Pascal Rigaux2006-12-042-1/+82
|
* - much stricter synthesis parsing. fail on first errorPascal Rigaux2006-12-044-15/+31
| | | | | - add test using a buggy synthesis that occured on kenobi
* correctly set "\0"Pascal Rigaux2006-12-041-0/+1
|
* remove temp filePascal Rigaux2006-12-041-1/+1
|
* correctly handle gzread error codePascal Rigaux2006-12-041-1/+4
|
* rpmReadPackageFile() can return ok but no header :-/Pascal Rigaux2006-12-011-2/+2
|
* rpmReadPackageFile() can return ok but no header :-/1.51Pascal Rigaux2006-12-012-2/+2
|
* (Trans_add) factorize size to allocateThierry Vignaud2006-11-291-1/+1
|
* (Trans_add) adjust allocated memoryThierry Vignaud2006-11-291-1/+1
|
* (Trans_add) remove guard, proper fix is now enoughThierry Vignaud2006-11-291-1/+0
|
* (Trans_add) fix segfault on ia32 (when one replace "struct foobar" byThierry Vignaud2006-11-291-0/+4
| | | | | "foobar*", he should expect sizeof() to *slightly* differ in results)
* 1.50 new release 1.50Pascal Rigaux2006-11-291-1/+1
|
* (Trans_add) fix segfault when using --excludepathThierry Vignaud2006-11-291-0/+1
|
* (Trans_add) fix the segfault when using --excludepath (introduced inThierry Vignaud2006-11-291-6/+5
| | | | | | r32435:32440). however, using --excludepath will still segfaults, but later, in rpmtsAddInstallElement() ...
* (Trans_add) fix segfaultThierry Vignaud2006-11-291-1/+1
|
* - default to strict-arch on 64bits (tvignaud)Pascal Rigaux2006-11-211-0/+92
| | | | | | | | - handle empty hdlist.cz/synthesis.cz (in build_hdlist, build_synthesis, parse_hdlist, parse_synthesis) - parse_rpms_build_headers: allow asking for packing (and so fixing an error in urpmi) - documentation & comments enhancement - release 1.48
* add some tests on empty header, buggy header, empty hdlistPascal Rigaux2006-11-171-1/+17
|
* at least one good header is needed for non compressed hdlist and non empty filePascal Rigaux2006-11-171-0/+2
|
* - parse_hdlist on a simple header was broken because pid was left unset,Pascal Rigaux2006-11-172-2/+14
| | | | | - add some tests for this
* do display on which pkgs the 2 sorted lists disagreePascal Rigaux2006-11-151-1/+3
|
* add a wrapper in perl around parse_hdlist and parse_synthesisPascal Rigaux2006-11-152-2/+19
|
* remove debug stuffPascal Rigaux2006-11-151-2/+0
|
* make the prototype of parse_synthesis more clear in the pod1.48Pascal Rigaux2006-11-151-4/+4
|
* parse_synthesis, parse_hdlist: handle gzip error status instead of relying onPascal Rigaux2006-11-151-9/+20
| | | | | | | | | wether we did read at least one header. This allow "empty" hdlist/synthesis. But it means it is getting stricter, and the headers added to {depslist} would need to be removed if an error occured. Alas i don't know how to do it (a simple splice) in XS. If no better solution, i'll create a wrapper function in perl.
* Bump version to 1.471.47Rafael Garcia-Suarez2006-10-162-1/+18
|
* Patch by Pixel to ignore self-obsoletesRafael Garcia-Suarez2006-10-161-0/+6
|
* Invert bogus checkRafael Garcia-Suarez2006-09-211-5/+5
|
* Add POD test, remove sole pod fragment in URPM::BuildRafael Garcia-Suarez2006-09-073-5/+8
|
* Bump version to 1.461.46Rafael Garcia-Suarez2006-09-062-1/+31
|
* Oops, remove debug codeRafael Garcia-Suarez2006-09-061-1/+0
|
* prefer kernel-source-stripped over kernel-sourceRafael Garcia-Suarez2006-09-061-0/+18
|
* Micro-optimisation, and make comments more explicitRafael Garcia-Suarez2006-09-051-4/+9
|
generated by cgit v1.2.1 (git 2.21.0) at 2024-11-25 23:28:14 +0000