Panu Matilainen [Thu, 9 Oct 2008 12:43:00 +0000 (15:43 +0300)]
Mark rpmdb internals as internals, including the "mergesort" gem
Panu Matilainen [Thu, 9 Oct 2008 12:37:59 +0000 (15:37 +0300)]
Mark psm functions as internal
Panu Matilainen [Thu, 9 Oct 2008 12:35:36 +0000 (15:35 +0300)]
Mark various fsm* bits internal
- build code uses fsmSetup() and fsmTearDown(), those need to remain
visible for now
Panu Matilainen [Thu, 9 Oct 2008 12:32:04 +0000 (15:32 +0300)]
Mark uidToUname() and friends as internal
Panu Matilainen [Thu, 9 Oct 2008 12:21:15 +0000 (15:21 +0300)]
Add RPM_GNUC_INTERNAL macro
- allows portable use of __attribute__((visibility("hidden")))
Panu Matilainen [Wed, 8 Oct 2008 15:26:58 +0000 (18:26 +0300)]
Unknown tag is an error
Jindrich Novy [Tue, 7 Oct 2008 20:06:28 +0000 (22:06 +0200)]
make %find_lang --with-kde work (rhbz#466009)
- patch from Rex Dieter
Panu Matilainen [Mon, 6 Oct 2008 15:50:30 +0000 (18:50 +0300)]
Malformed query query items (pkid's and such) are errors
Panu Matilainen [Mon, 6 Oct 2008 15:49:33 +0000 (18:49 +0300)]
Make "not an rpm or manifest" message an error, not just notice
Panu Matilainen [Mon, 6 Oct 2008 15:41:09 +0000 (18:41 +0300)]
Consistent log level for "unverifiable signature" message
- all the other ones in package are RPMLOG_ERR
Panu Matilainen [Sat, 4 Oct 2008 12:16:38 +0000 (15:16 +0300)]
Failure to rebuild database is worth at least a warning
Panu Matilainen [Sat, 4 Oct 2008 08:19:26 +0000 (11:19 +0300)]
Restore permissions of a signed rpm to that of the original (rhbz#465586)
Panu Matilainen [Fri, 3 Oct 2008 10:00:33 +0000 (13:00 +0300)]
Similar semantics for <lua> scriptlets
- missing <lua> support is always an error
- otherwise permit non-pre scriptlets to fail
Panu Matilainen [Fri, 3 Oct 2008 09:49:29 +0000 (12:49 +0300)]
Always perform callback notify of lua scriptlet errors too
Panu Matilainen [Fri, 3 Oct 2008 09:13:21 +0000 (12:13 +0300)]
Remove unnecessary {} block + indentation, no functional changes
Panu Matilainen [Fri, 3 Oct 2008 09:04:19 +0000 (12:04 +0300)]
Filter out error exit codes from non-pre scriptlets (rhbz#216221, #rhbz465409)
- typos and thinkos in scriptlets often cause duplicates to be left around
for no good reason
- treat non-zero exit codes from scriptlets as just warnings instead of
errors
- ... except that %pre and %preun are sometimes used for preventing
install/uninstall of a package, preserve the traditional behavior for them
- abuse rpmtsNotify() "total" value for signaling warning/error to callback
Panu Matilainen [Fri, 3 Oct 2008 08:41:41 +0000 (11:41 +0300)]
Emit RPMCALLBACK_SCRIPT_ERROR notification on all scriptlet failures
- previously the fork() failures and such would not be notified at all
- discard the not-so-useful differing exit codes (signal, exit status and
whatnot) from commit
179ebc1745b51eb9821dd783c9df475ac539532d
Panu Matilainen [Fri, 3 Oct 2008 08:38:08 +0000 (11:38 +0300)]
Assume failure in psm runScript()
Panu Matilainen [Fri, 3 Oct 2008 08:34:41 +0000 (11:34 +0300)]
Log message on all scriptlet failures.
Panu Matilainen [Thu, 2 Oct 2008 09:43:02 +0000 (12:43 +0300)]
Updated German translation (Fabian Affolter)
Panu Matilainen [Wed, 1 Oct 2008 14:05:35 +0000 (17:05 +0300)]
Kick out yet another ancient artifact
- --usecatalog hasn't done anything in 10 years if ever afaict...
Panu Matilainen [Wed, 1 Oct 2008 13:28:47 +0000 (16:28 +0300)]
We have easier ways to catenate strings...
Panu Matilainen [Wed, 1 Oct 2008 07:06:27 +0000 (10:06 +0300)]
Tweaks to how ISA-macros are generated (rhbz#464754)
- for ppc and sparc, move the check for 64bit version first and use a
wildcards to cover more field automatically
Jindrich Novy [Thu, 25 Sep 2008 17:44:29 +0000 (19:44 +0200)]
Use safer xrealloc() instead of realloc()
Jindrich Novy [Thu, 25 Sep 2008 17:32:27 +0000 (19:32 +0200)]
Avoid segfault in rpmLeadNew() if calloc() fails
Jindrich Novy [Thu, 25 Sep 2008 17:30:47 +0000 (19:30 +0200)]
Use safer xstrdup() instead of strdup() wherever possible
Panu Matilainen [Thu, 25 Sep 2008 16:16:14 +0000 (19:16 +0300)]
Fix up fuzz in Finnish translation
Panu Matilainen [Thu, 25 Sep 2008 16:10:02 +0000 (19:10 +0300)]
Extra ')' in error message
Panu Matilainen [Thu, 25 Sep 2008 11:58:05 +0000 (14:58 +0300)]
Don't go onto wild zombie rampage if python callback tracebacks (rhbz#463447)
- If a callback tracebacks, the python program that called us in the first
place is as good as dead already. Clean up what we can, throw an error
message and just die, no good is going to come out of blindly continuing.
Jindrich Novy [Wed, 24 Sep 2008 17:49:09 +0000 (19:49 +0200)]
Update Brasilian Portuguese translation (rhbz#463748)
Jindrich Novy [Wed, 24 Sep 2008 16:39:43 +0000 (18:39 +0200)]
Don't treat %patch numberless if -P parameter is present
Panu Matilainen [Mon, 22 Sep 2008 11:30:37 +0000 (14:30 +0300)]
Kick out useless and ancient junk genCpioListAndHeader()
- scareMem is not used for anything here
- xmalloc() + xcalloc() cannot return NULL, don't bother checking
Jindrich Novy [Mon, 22 Sep 2008 10:10:36 +0000 (12:10 +0200)]
Fix error messages in processScriptFiles()
Jindrich Novy [Thu, 18 Sep 2008 13:18:53 +0000 (15:18 +0200)]
Drop rpmfcFileDep() to simplify things
Jindrich Novy [Thu, 18 Sep 2008 12:57:52 +0000 (14:57 +0200)]
Move type regression test back to asserts
- no need to bother with message translations in this case
Jindrich Novy [Thu, 18 Sep 2008 12:56:15 +0000 (14:56 +0200)]
Use proper codes for logging via rpmlog().
Panu Matilainen [Thu, 18 Sep 2008 11:31:47 +0000 (14:31 +0300)]
Force error return if headerLoad() fails
Panu Matilainen [Thu, 18 Sep 2008 10:24:25 +0000 (13:24 +0300)]
Document manifest behavior for install options too (rhbz#461354)
Panu Matilainen [Thu, 18 Sep 2008 10:05:19 +0000 (13:05 +0300)]
Kick out references to "Red Hat Linux" out of the manpages, duh :)
- RHL hasn't existed in years and rpm isn't specific to RH anyway...
Panu Matilainen [Thu, 18 Sep 2008 09:38:04 +0000 (12:38 +0300)]
Update man-page wrt supported query formatter types
Panu Matilainen [Thu, 18 Sep 2008 09:36:31 +0000 (12:36 +0300)]
Don't segfault on unknown format specifiers
Panu Matilainen [Thu, 18 Sep 2008 09:15:43 +0000 (12:15 +0300)]
Add error message on query of unsupported files (rhbz#461362)
Panu Matilainen [Thu, 18 Sep 2008 08:39:42 +0000 (11:39 +0300)]
Remove pointless non-NULL check
- rpmfiNew() cannot return NULL, and even if it did, we'd be out of here
by the way of rpmfiFC(fi) returning zero before this...
Panu Matilainen [Thu, 18 Sep 2008 08:29:58 +0000 (11:29 +0300)]
Save some screen estate by initializing on declaration..
Panu Matilainen [Thu, 18 Sep 2008 07:26:39 +0000 (10:26 +0300)]
More informative output on ambiguous erase (rhbz#164743)
Panu Matilainen [Thu, 18 Sep 2008 07:04:51 +0000 (10:04 +0300)]
Fix rpmErase() exit code when arch is specified (rhbz#462631)
- essentially the same bug as rhbz#124016 but for erase code
- rpmdbIteratorCount() isn't realiable on RPMDBI_LABEL, we need to walk
the iterator to know if there were actual matches
Panu Matilainen [Tue, 16 Sep 2008 09:08:18 +0000 (12:08 +0300)]
Kick out double definitions of --(no)dirtokens options
- mistakenly added in commit
2f7c913723baa8135201fa930678d9e3d4138f2c
Panu Matilainen [Tue, 16 Sep 2008 09:03:26 +0000 (12:03 +0300)]
Resurrect --nodirtokens build option
- this gets somewhat hysterical: we create the filelist as uncompressed,
then compress it for rpmfi which only understands compressed filelist,
and if nodirtokens is used, expand the list again after initializing
rpmfi for cpio/fsm...
Jindrich Novy [Mon, 15 Sep 2008 14:42:01 +0000 (16:42 +0200)]
Avoid asserts and output error instead (rhbz#448939, rhbz#456100)
Jindrich Novy [Mon, 15 Sep 2008 14:37:32 +0000 (16:37 +0200)]
Use proper return codes in rpmfc.c
Jindrich Novy [Mon, 15 Sep 2008 14:35:30 +0000 (16:35 +0200)]
Use proper return codes in files.c
- don't go on if processPackageFiles() or rpmfcGenerateDepends() fails
Panu Matilainen [Fri, 12 Sep 2008 12:22:02 +0000 (15:22 +0300)]
headerAddEntry() is deprecated, use headerPutString() instead
Panu Matilainen [Fri, 12 Sep 2008 11:56:41 +0000 (14:56 +0300)]
Make sure RPMTAG_GROUP is always present in packages
- Many things (including LSB and various 3rd party applications) expect
group tag to be present even if they don't actually use it for anything.
rpmbuild doesn't consider it mandatory in specs anymore though, so stick
"Unspecified" in there to avoid breaking things in case spec didn't set it.
Panu Matilainen [Fri, 12 Sep 2008 08:38:07 +0000 (11:38 +0300)]
Fix indentation
Panu Matilainen [Thu, 11 Sep 2008 13:56:50 +0000 (16:56 +0300)]
Macro argument handling regression regression (rhbz#461180)
- commit
f366011c42e4cf13d3acdd1af45127f0a1baec3a is too simplistic and
broke other stuff in turn...
Jindrich Novy [Wed, 10 Sep 2008 13:04:57 +0000 (15:04 +0200)]
Don't generate broken cpio in case of hardlink pointing on softlink
- thanks to Pixel
Jindrich Novy [Sat, 6 Sep 2008 13:30:40 +0000 (15:30 +0200)]
Fail hard if patch isn't found (rhbz#461347)
Jindrich Novy [Thu, 4 Sep 2008 04:34:47 +0000 (06:34 +0200)]
Add Dutch translation to LINGUAS
Jindrich Novy [Thu, 4 Sep 2008 04:34:11 +0000 (06:34 +0200)]
Add initial Dutch translation (rhbz#461060)
Panu Matilainen [Tue, 2 Sep 2008 12:25:17 +0000 (15:25 +0300)]
Updated German translation from Fabian Affolter
Panu Matilainen [Tue, 2 Sep 2008 11:33:55 +0000 (14:33 +0300)]
--repackage is gone, remove references from French manual too
Panu Matilainen [Tue, 2 Sep 2008 11:28:35 +0000 (14:28 +0300)]
Updated French translation from man-pages-fr
Panu Matilainen [Tue, 2 Sep 2008 07:18:41 +0000 (10:18 +0300)]
Fix --dump usage description in Japanese manpage translation.
Patch from Masahiro Matsuya.
Panu Matilainen [Wed, 20 Aug 2008 09:01:38 +0000 (12:01 +0300)]
Avoid dumb casts
- switch-case on rpmTag enum makes gcc go crazy without a default case
Jindrich Novy [Mon, 1 Sep 2008 12:35:34 +0000 (14:35 +0200)]
Create directory structure for rpmbuild prior to build if it doesn't exist
- creates all needed directories that rpmbuild uses for input and output
Jindrich Novy [Mon, 1 Sep 2008 12:31:33 +0000 (14:31 +0200)]
Create _topdir if it doesn't exist when installing SRPM
- otherwise creating _sourcedir and _specdir fails
Jindrich Novy [Tue, 26 Aug 2008 17:03:25 +0000 (19:03 +0200)]
Fix type in format string to avoid segfault (rhbz#460146)
Jindrich Novy [Tue, 26 Aug 2008 11:25:21 +0000 (13:25 +0200)]
Allow noarch subpackages
- inspired by patch from Altlinux
Panu Matilainen [Tue, 19 Aug 2008 10:45:20 +0000 (13:45 +0300)]
Improve INSTALL documentation wrt NSRP/NSS requirements
Panu Matilainen [Tue, 19 Aug 2008 10:33:21 +0000 (13:33 +0300)]
Use standard auto*foo checks for NSS
- upstream NSPR/NSS doesn't ship pkg-config files (unfortunately)
- hardcoded guesses are always wrong for somebody, expect builder to
pass in appropriate CPPFLAGS etc
Panu Matilainen [Thu, 14 Aug 2008 12:02:52 +0000 (15:02 +0300)]
Fix segfault on zero-length %policy file
Panu Matilainen [Thu, 14 Aug 2008 11:28:12 +0000 (14:28 +0300)]
Fix segfault on zero-length rpmrc
Jindrich Novy [Thu, 14 Aug 2008 11:08:09 +0000 (13:08 +0200)]
Accept numerical characters in macro expansion (related to rhbz#456103)
Panu Matilainen [Thu, 14 Aug 2008 06:57:03 +0000 (09:57 +0300)]
Don't fail build on too large but excluded files
Panu Matilainen [Wed, 13 Aug 2008 06:43:41 +0000 (09:43 +0300)]
Unbreak payload size tag generation on ppc (rhbz#458817)
Panu Matilainen [Mon, 11 Aug 2008 12:39:47 +0000 (15:39 +0300)]
Macroize configure command/path in %configure (rhbz#458644)
- add + use %{_configure} macro defaulting to ./configure, can be overridden
for VPATH builds etc
Panu Matilainen [Fri, 8 Aug 2008 10:53:57 +0000 (13:53 +0300)]
Add parenthesis to make gcc happy
Panu Matilainen [Fri, 8 Aug 2008 10:53:39 +0000 (13:53 +0300)]
Missing include
Panu Matilainen [Fri, 8 Aug 2008 10:06:28 +0000 (13:06 +0300)]
Remember to free up resources from matchpathcon_init()
Panu Matilainen [Fri, 8 Aug 2008 10:04:10 +0000 (13:04 +0300)]
Fix check for nocontext flag
Panu Matilainen [Fri, 8 Aug 2008 09:49:41 +0000 (12:49 +0300)]
Plug memleak from %pretrans
- te->fi needs to be freed if we replace it...
Panu Matilainen [Fri, 8 Aug 2008 06:57:41 +0000 (09:57 +0300)]
Add key to transaction keyring on rpmtsImportPubkey() (rhbz#457360)
- still not 100% identical to pre-keyring behavior but enough for the
common case
- only add keys we don't already have (rhbz#68290)
Panu Matilainen [Fri, 8 Aug 2008 06:32:05 +0000 (09:32 +0300)]
Use rpmPubkey + methods for pubkey header generation
- moves some of the lowlevel pubkey fiddling into rpmio where it belongs
Panu Matilainen [Fri, 8 Aug 2008 06:26:30 +0000 (09:26 +0300)]
Add couple of new rpmPubkey() methods
- rpmPubkeyDig() to retrieve parsed OpenPGP parameters of key
- rpmPubkeyBase64() to return base64 encoding of the key
Panu Matilainen [Fri, 8 Aug 2008 06:14:53 +0000 (09:14 +0300)]
Split gpg-pubkey header generation out of rpmtsImportPubkey()
Panu Matilainen [Thu, 7 Aug 2008 15:44:03 +0000 (18:44 +0300)]
Permit i18nstring in headerPutString()
Jindrich Novy [Thu, 7 Aug 2008 12:50:49 +0000 (14:50 +0200)]
Don't allow multiple definition of PatchX or SourceX (rhbz#458261, rhbz#458260)
Panu Matilainen [Wed, 6 Aug 2008 09:27:25 +0000 (12:27 +0300)]
Clarify %buildroot in macros
- it's no longer really optional..
Panu Matilainen [Wed, 6 Aug 2008 09:24:34 +0000 (12:24 +0300)]
Simplify spec buildroot hysteria
- BuildRoot: in spec is now really ignored
- parseSpec() uses default %{buildroot} unless overridden
- BuildRoot is forced for all builds
Panu Matilainen [Wed, 6 Aug 2008 05:38:57 +0000 (08:38 +0300)]
Add %sparc macro for supported Sparc architectures (rhbz#455972)
Panu Matilainen [Tue, 5 Aug 2008 12:32:49 +0000 (15:32 +0300)]
Increment exit code on file not found errors too (#446202)
Panu Matilainen [Mon, 4 Aug 2008 12:46:18 +0000 (15:46 +0300)]
Bunch of keyring / pubkey refcounting fixes
Panu Matilainen [Sat, 26 Jul 2008 08:14:45 +0000 (11:14 +0300)]
Updated Italian translation from Francesco Tombolini (rhbz#453364)
Panu Matilainen [Sat, 26 Jul 2008 07:56:02 +0000 (10:56 +0300)]
Revert "plug memory leak in spec parsing code"
This reverts commit
ac3bd07ef2a11cbe9e7f7031786eebc8851c863b.
- It's not a leak, valueMakeString() doesn't make a copy of the argument
string and the free() causes invalid memory access
Panu Matilainen [Sat, 26 Jul 2008 06:57:09 +0000 (09:57 +0300)]
Fix typo in buildroot dir definition (rhbz#456708)
- it's _buildrootdir, not buildrootdir... this caused user overridden _topdir
not getting honored by buildroot
Florian Festi [Fri, 25 Jul 2008 10:28:46 +0000 (12:28 +0200)]
plug memory leak in spec parsing code
Florian Festi [Thu, 24 Jul 2008 15:38:11 +0000 (17:38 +0200)]
fix parsing of empty doublequoted strings in spec file (rhbz#456103)
off-by-one error
Panu Matilainen [Thu, 24 Jul 2008 08:17:50 +0000 (11:17 +0300)]
Dumb thinko breaking tar builds from relative paths (rhbz#456321)
Panu Matilainen [Sat, 19 Jul 2008 12:27:42 +0000 (15:27 +0300)]
Fix regression in %patch handling (rhbz#455872)
- popt doesn't think of "008" as an integer, let popt think it's a string
as we do our own conversion to number anyway
Panu Matilainen [Thu, 17 Jul 2008 12:50:35 +0000 (15:50 +0300)]
Add (and use) refcounting for rpmKeyring and rpmPubkey