Steve Lawrence [Mon, 21 Jun 2010 21:04:36 +0000 (17:04 -0400)]
Add new Collections preamble tag
This adds a new Collections preamble tag, which contains a space separated list
of collection names that the package belongs to.
Héctor Daniel Cabrera [Mon, 14 Jun 2010 15:45:49 +0000 (15:45 +0000)]
l10n: Updated Spanish (Castilian) (es) translation to 100%
New status: 724 messages complete with 0 fuzzies and 0 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Panu Matilainen [Fri, 11 Jun 2010 05:19:08 +0000 (08:19 +0300)]
Dont bother with selinux contexts in test transactions
- initializing contexts isn't exactly cheap, and test transactions
have no use for them
Panu Matilainen [Fri, 11 Jun 2010 05:17:12 +0000 (08:17 +0300)]
If there are no hardlinks, dont bother with s-bit and caps removal
Tadashi Jokagi [Mon, 7 Jun 2010 22:12:12 +0000 (22:12 +0000)]
l10n: Updated Japanese (ja) translation to 97%
New status: 707 messages complete with 9 fuzzies and 8 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Tadashi Jokagi [Mon, 7 Jun 2010 22:10:29 +0000 (22:10 +0000)]
l10n: Updated Japanese (ja) translation to 97%
New status: 705 messages complete with 10 fuzzies and 9 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Panu Matilainen [Mon, 7 Jun 2010 11:01:02 +0000 (14:01 +0300)]
Oops, bad format spec in Finnish translation
Panu Matilainen [Mon, 7 Jun 2010 10:14:52 +0000 (13:14 +0300)]
Eliminate fuzzies from Finnish translation
Panu Matilainen [Mon, 7 Jun 2010 09:51:09 +0000 (12:51 +0300)]
Update .gitignore a bit
Yuri Chornoivan [Sun, 6 Jun 2010 05:40:03 +0000 (05:40 +0000)]
l10n: Updated Ukrainian (uk) translation to 100%
New status: 724 messages complete with 0 fuzzies and 0 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Piotr Drąg [Thu, 3 Jun 2010 18:45:25 +0000 (18:45 +0000)]
l10n: Updated Polish (pl) translation to 100%
New status: 724 messages complete with 0 fuzzies and 0 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Panu Matilainen [Thu, 3 Jun 2010 13:22:09 +0000 (16:22 +0300)]
Fix "empty reply from server" curl-syndrome with url retrieval (RhBug:598988)
- We were adding a trailing newline to urlhelper command line passed
to execvp(), with the newline ending up in the URL passed to curl.
Many servers dont seem to mind the extra newline, but some do. Oh well...
Panu Matilainen [Thu, 3 Jun 2010 08:04:12 +0000 (11:04 +0300)]
Remove POSIX file capabilities from executables on erase/rename too
- Just like suid/sgid bits, hardlinks to executables with capabilities
set can "leak" permissions to old, potentially vulnerable versions
of binaries. Related to RhBug:598775.
Panu Matilainen [Wed, 2 Jun 2010 06:15:48 +0000 (09:15 +0300)]
Strip s-bits on upgrade too (RhBug:598775)
- Previously only actual package removal would clear s-bits from
s[ug]id files. This needs to be done on upgrade too to prevent
leaving vulnerable hardlinks of s[ug]id binaries around.
Tadashi Jokagi [Wed, 2 Jun 2010 00:01:13 +0000 (00:01 +0000)]
l10n: Updates to Japanese (ja) translation
New status: 664 messages complete with 47 fuzzies and 13 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Panu Matilainen [Tue, 1 Jun 2010 12:49:16 +0000 (15:49 +0300)]
Refactor rpmdb lookups + caching to a helper function + cosmetics
- no functional changes (supposedly ;) just cleaning things up a bit
Panu Matilainen [Tue, 1 Jun 2010 10:55:15 +0000 (13:55 +0300)]
On colored transactions, require matching arch for freshen (RhBug:553108)
- What we'd really want to compare here is the color or ISA of the packages
but color isn't realiable as eg -devel packages often aren't correctly
colored, and ISA isn't currently available in the header as a standalone
entry. Comparing arch prevents otherwise reasoable i386 -> i686 type
arch changes on freshen but at least it avoids the most pathological
issues.
Panu Matilainen [Tue, 1 Jun 2010 08:58:42 +0000 (11:58 +0300)]
Permit DOS line-endings in PGP armors (RhBug:532992)
- RFC-4880 doesn't requires unix-style line-endings, we shouldn't either.
This is probably still oversly strict as RFC-4880 appears to permit
any whitespace to follow armor headers but ... shrug.
Panu Matilainen [Tue, 1 Jun 2010 07:37:09 +0000 (10:37 +0300)]
Support many pubkeys in a single file in --import (RhBug:586827)
Panu Matilainen [Tue, 1 Jun 2010 07:15:10 +0000 (10:15 +0300)]
Don't return errors on trying to import duplicate keys
- On the principle of "no news is good news", if we already have the
keys to be imported, then everything is ok. The former behavior is
just confusing as witnessed in RhBug:462979
Panu Matilainen [Mon, 31 May 2010 15:00:33 +0000 (18:00 +0300)]
Nuke leftover variable from the chroot workout
Panu Matilainen [Mon, 31 May 2010 14:01:53 +0000 (17:01 +0300)]
Update translations to pick up chroot-related string changes
Panu Matilainen [Mon, 31 May 2010 11:07:46 +0000 (14:07 +0300)]
Use the new chroot API everywhere, eliminate old related goo
Panu Matilainen [Mon, 31 May 2010 10:11:47 +0000 (13:11 +0300)]
Add a simple API for dealing with chroot state changes
- Chroot is a process global state so it needs to be tracked globally.
A process can (in theory) have several transaction sets, each with
different roots (although only one can be active at any time), so
associating the chroot state with transaction set (as currently done)
is not right.
- "Reference count" chroot entering and exiting so callers dont need
to track the state changes individually when they need to go
in and out of chroot if not already done.
- This should probably go to librpmio eventually but as there are no
needs outside librpm currently, keeping this internal so we're free
to fiddle with the api if necessary
Panu Matilainen [Mon, 31 May 2010 10:28:58 +0000 (13:28 +0300)]
Eliminate unused no-op FSM_CHROOT stage
- the fsm thingie doesn't need to deal with chroot issues: except for
build where it doesn't matter, its only called from psm where the
chroot is already taken care of
Panu Matilainen [Fri, 28 May 2010 09:57:59 +0000 (12:57 +0300)]
Add internal helper for formatting file attribute strings
- Eliminates some copy-paste slopping. Verify code traditionally only
shows the first attribute, take care to keep that format to avoid
breaking scripts unnecessarily
Panu Matilainen [Fri, 28 May 2010 09:21:34 +0000 (12:21 +0300)]
Add header extension tag RPMTAG_FILESTATUS for file verification
- Permits basic file verification with just a headerGet(), with some
caveats: there's no way to control which attributes get verified,
and there's no filtering of mtime differences of shared files. Those
aside, rpm -q --qf "[%{filestates:vflags} %{filenames}\n] <args>" now
performs the same as "rpm -V --nodeps --noscripts <args>"
Panu Matilainen [Fri, 28 May 2010 09:18:26 +0000 (12:18 +0300)]
Add internal helper for formatting verify flags
Panu Matilainen [Fri, 28 May 2010 07:49:58 +0000 (10:49 +0300)]
Move shared file timestamp filtering out of rpmVerifyFile()
- rpmVerifyFile() reports things as they are in reality - if timestamps
on disk differ to header then it reports it. Filtering on the result
belongs to the code interpreting the results.
Panu Matilainen [Fri, 28 May 2010 07:27:01 +0000 (10:27 +0300)]
Dont crash on NULL ts to rpmtsInitIterator()
Panu Matilainen [Fri, 28 May 2010 06:57:09 +0000 (09:57 +0300)]
Only pass the info we actually care about to verifyHeader()
Panu Matilainen [Fri, 28 May 2010 06:51:51 +0000 (09:51 +0300)]
Eliminate unused argument to verifyDependencies()
Panu Matilainen [Fri, 28 May 2010 06:47:47 +0000 (09:47 +0300)]
Clean up verify script handling a bit
- test verifyscript existence in rpmVerifyScript() instead of caller,
making it safe+correct for calling on any header
- eliminate unused/unnecessary arguments, local variables and bogus leftover
comments
Panu Matilainen [Wed, 26 May 2010 10:25:39 +0000 (13:25 +0300)]
Handle test-transaction centrally on entry to psm
- psm in test-mode is a no-op anyway, so dont bother at all with it
- also removes the need to check for test mode on pre/posttrans scripts,
these are now fully covered by psm too
Panu Matilainen [Wed, 26 May 2010 08:30:15 +0000 (11:30 +0300)]
Eliminate rpmts from the scriptlet machinery
- pass scriptFd and selinux enabled status as arguments from psm level
- selinux status could be queried directly with is_selinux_enabled()
but that's a fairly expensive call which does all sorts of funny
things and probably doesn't work at all in the average chroot
- Lua-scripts dont currently honor scriptFd and have no use for selinux
but pass the info there too as an early step towards supporting
scriptFd with Lua
- makes rpmScriptRun() even more of an argument monster, some of this
should probably go into rpmScript struct...
Panu Matilainen [Wed, 26 May 2010 07:15:06 +0000 (10:15 +0300)]
Simplify chroot handling wrt scriptlets
- switch root if necessary on entry and exit to psm already, nothing
inside the psm needs access to outside chroot
- eliminate chroot handling from scriptlet machinery, dealing with
chroot is a job for higher levels
- Lua scriptlets can change our cwd, always ensure we return to previous
cwd after executing by saving and restoring the cwd
Panu Matilainen [Tue, 25 May 2010 12:38:15 +0000 (15:38 +0300)]
Move scriptlet statistics collection (back) inside psm
- removes one rpmts dependency from scriptlets
Jindrich Novy [Mon, 24 May 2010 07:39:54 +0000 (09:39 +0200)]
Put the macro evaluation in spec comments back.
- '#' at the beginning doesn't need to always represent comment in spec
but can be part of the embedded bash script (RhBug:594672 - comment 6)
Panu Matilainen [Fri, 21 May 2010 12:36:03 +0000 (15:36 +0300)]
Move the lone hashFunctionString() into misc.[ch], eliminate rpmhash.[ch]
- Besides there not being much point in having a separate source + header
for a small single function, this fixes build on case-insensitive
systems such as Mac OS X.
Jindrich Novy [Fri, 21 May 2010 11:12:02 +0000 (13:12 +0200)]
Do not evaluate macros in spec comments (RhBug:594672)
Till Maas [Thu, 20 May 2010 08:35:16 +0000 (11:35 +0300)]
Search all locale dirs in find-lang.sh, not just those under share/ (#159)
Panu Matilainen [Wed, 19 May 2010 12:33:52 +0000 (15:33 +0300)]
Use pgpValString() for :pgpsig header formatting (RhBug:587755)
- instead of copy-slopping (and forgetting to do so whenever something
changes) just use the values we know from pgpVals..
Panu Matilainen [Wed, 19 May 2010 12:32:54 +0000 (15:32 +0300)]
Add public pgpValString() function + enum for the various types
- The equivalent used to be possible in 4.4.x era as the value tables
were all exported, but this way we need to export far less
and the implementation details stay internal.
Panu Matilainen [Wed, 19 May 2010 11:59:02 +0000 (14:59 +0300)]
Bury struct pgpValTbl_s definition inside rpmpgp.c
- as all the value tables are static inside rpmpgp.c, exposing this
is pointless (should've been unexported long time ago)
Panu Matilainen [Wed, 19 May 2010 07:12:43 +0000 (10:12 +0300)]
Handle non-existent dependency sets in python (RhBug:593553)
- rpmdsNew() returns NULL if the requested dependency type doesn't
exist in the header. The C-side API can handle NULL to all rpmds
"methods" and this is how librpm deals with non-existent sets rather
than waste memory on for empty ds structures. However the python side
wasn't expecting NULL for legal requests (but not setting error either)
and thus blowing up with SystemError exception.
- Raise TypeError on illegal arguments to rpm.ds constructor, and present
non-existent dependency sets as empty rpm.ds objects to python. This
lets python callers use iteration over ds items regardless of whether
the dependency actually exists or not. The alternative of returning
None (or raising exceptions) would break existing code for no
particularly good reason.
Panu Matilainen [Tue, 18 May 2010 07:39:22 +0000 (10:39 +0300)]
Fix return from chroot() on verify (RhBug:590588)
- a couple of important steps in chroot() in and out sequence missing,
causing "No such file or directory" whining on return from chroot()
unless cwd happened to be /
Roland McGrath [Tue, 11 May 2010 03:05:48 +0000 (20:05 -0700)]
Make debugedit -i recompute build ID only when any DWARF was actually changed.
Roland McGrath [Tue, 11 May 2010 00:53:07 +0000 (17:53 -0700)]
Whitespace fixups for tools/debugedit.c
Panu Matilainen [Fri, 7 May 2010 17:24:01 +0000 (20:24 +0300)]
Plug dumb memleak in rpmpsMerge()
Panu Matilainen [Fri, 7 May 2010 17:18:58 +0000 (20:18 +0300)]
Fix breakage from removedPackages changing to hash
- tsmem->delta is still needed for tsmem->order reallocation,
zero delta causes funny effects on larger package sets (test suite
doesnt trigger this)
Panu Matilainen [Fri, 7 May 2010 12:25:38 +0000 (15:25 +0300)]
Revert back to rpmdbRemove() taking instance number instead of actual header
- On package reinstall (ie --replacepkgs) we only know the old header
by the instance number, and grabbing the header from the db in
psm as a special case for that is more trouble than just doing what
we always did.
- Reverts commit
e16695e932e294ec600e673d38c41bde20931204 and unbreaks
the --replacepkgs test case
Panu Matilainen [Fri, 7 May 2010 11:31:25 +0000 (14:31 +0300)]
Convert rpmte internals to use the transaction pointer from element itself
Panu Matilainen [Fri, 7 May 2010 11:07:39 +0000 (14:07 +0300)]
Store pointer to transaction set in transaction elements
- unused now but allows various new things and more element "independency"
- eliminate the unused dboffset argument while at it
Florian Festi [Fri, 7 May 2010 11:06:40 +0000 (13:06 +0200)]
Fix compiler warning in rpmhash Resize()
Florian Festi [Thu, 6 May 2010 14:31:53 +0000 (16:31 +0200)]
Remove no longer needed function intcmp()
Florian Festi [Wed, 5 May 2010 10:49:16 +0000 (12:49 +0200)]
Convert tsmem->removedPackages to a hash table and remove rpmdbPruneIterator from the API
Florian Festi [Thu, 6 May 2010 13:30:01 +0000 (15:30 +0200)]
rpmhash: new Empty() function to remove all entries
Florian Festi [Wed, 5 May 2010 10:48:41 +0000 (12:48 +0200)]
rpmhash: Accessor functions for the size and usage
Florian Festi [Thu, 25 Mar 2010 11:06:22 +0000 (12:06 +0100)]
rpmhash: Grow when hash table gets too full
Add some statistics to be able to find out how full the hash is
Panu Matilainen [Thu, 6 May 2010 12:12:41 +0000 (15:12 +0300)]
Move header filecount retrieve from addTE() to rpmfsNew()
Panu Matilainen [Thu, 6 May 2010 11:55:38 +0000 (14:55 +0300)]
Tolerate NULL in rpmfsFree() and rpmfsFC()
- both are "can't happen" situations but easy to handle cleanly here...
Panu Matilainen [Thu, 6 May 2010 11:51:48 +0000 (14:51 +0300)]
Minor cleanup to rpmfsNew()
- use xcalloc() to ensure clean state, no need to manually do it all
- group non-conditional and conditional operations together
Panu Matilainen [Thu, 6 May 2010 11:14:16 +0000 (14:14 +0300)]
Move delTE() inline to rpmteFree()
- no point splitting this up, its just straightforward freeing of stuff
- dont bother with tonne of assigning everything to NULL, it gets
zeroed out by memset() anyway
Panu Matilainen [Thu, 6 May 2010 10:20:26 +0000 (13:20 +0300)]
Unexport rpmteNew() and rpmteFree()
- the only meaningful way to create transaction elements is by adding things
to a transaction through rpmtsAddFooElement(), there's no need to
expose these and restrict our ability to change the API when needed
Panu Matilainen [Thu, 6 May 2010 09:32:08 +0000 (12:32 +0300)]
Minor cleanup to rpmReSign()
- headerNew() cannot fail, dont bother checking
- initialize variables on declaration
Panu Matilainen [Thu, 6 May 2010 09:28:23 +0000 (12:28 +0300)]
Fix minor header iterator leak in rpmReSign()
- fix dumb double-initialization error introduced somewhere between
4.4 and 4.6, also while-loop is more natural here than for-loop
Panu Matilainen [Thu, 6 May 2010 09:16:20 +0000 (12:16 +0300)]
Clean up rpmcliImportPubkeys() a bit, no functional changes
- remove questionable rpmtsClean() call remnants from rpm 4.4.x days
when it used to clean the signature/digest data in ts but also
clean a pile of completely unrelated items
- rearrange the error code handling so we can get by with single place
of freeing the temp allocations
- move local variables to the scope where needed and eliminate unnecessary
ones
- remove redundant NULL check on argv, this is already done by caller
- remove unused qva argument (static function so no API implications)
Pavol Rusnak [Wed, 17 Mar 2010 15:02:46 +0000 (16:02 +0100)]
Don't process desktop files without Type=Application and Exec= line
Panu Matilainen [Tue, 4 May 2010 06:05:23 +0000 (09:05 +0300)]
Rename python module version symbol to __version__, add __version_info__
- This seems to be a de-facto standard in python land, as rpm.version
hasn't been in any released version we can still change this easily.
Suggestion from James Antill.
Panu Matilainen [Thu, 29 Apr 2010 12:47:59 +0000 (15:47 +0300)]
Doh, dont shadow the rpmtsVerifyDB() return value
Giulio Eulisse [Tue, 20 Apr 2010 09:42:10 +0000 (11:42 +0200)]
Fixes missing __fxstat64 symbol on mac.
Panu Matilainen [Wed, 28 Apr 2010 09:31:14 +0000 (12:31 +0300)]
Fix a few uninitialized variable cases found by clang-analyzer
Taylon Silmer [Tue, 27 Apr 2010 18:21:27 +0000 (18:21 +0000)]
l10n: Updates to Portuguese (Brazilian) (pt_BR) translation
New status: 721 messages complete with 1 fuzzy and 0 untranslated.
Transmitted-via: Transifex (www.transifex.net).
Panu Matilainen [Thu, 22 Apr 2010 10:48:58 +0000 (13:48 +0300)]
Protect rpmtsVerifyDB() with transaction lock
- BDB docs state that db->verify() doesn't perform any locking even
when used within an environment, and should only be used on files
that are not being modified by another thread of control. Grabbing
the transaction lock while verifying mostly ensures nobodys writing
to the db behind our back.
Panu Matilainen [Thu, 22 Apr 2010 09:49:03 +0000 (12:49 +0300)]
Ditch "fcntl_locking" bdb config option
- This is an undocumented BDB interface (only documented in the db.h header),
and its not something users should be messing around with
Panu Matilainen [Thu, 22 Apr 2010 09:23:24 +0000 (12:23 +0300)]
Make transaction lock path per-transaction
- Although it doesn't really happen in practise, rpm's API permits several
transactions with possibly differing roots within process lifetime.
Previously the lock path was calculated just once globally so we could
easily be locking in a completely wrong place (eg locking in a previously
accessed chroot when system rpmdb should be transaction-locked)
Panu Matilainen [Thu, 22 Apr 2010 08:01:27 +0000 (11:01 +0300)]
Further generalize the rpmlock interface
- Move transaction lock path handling into rpmts.c, export new low
level rpmlockAcquire() function to actually grab a lock
- Rename rpmtsFreeLock() to rpmlockFree() and return NULL in the
general rpm style
Panu Matilainen [Thu, 22 Apr 2010 07:44:38 +0000 (10:44 +0300)]
Generalize rpmlock_new() by moving transaction lock path calculation out
Panu Matilainen [Thu, 22 Apr 2010 07:37:51 +0000 (10:37 +0300)]
Make rpmlock type visible inside rpm, use instead of void*
Panu Matilainen [Wed, 21 Apr 2010 10:00:43 +0000 (13:00 +0300)]
Lose the useless header refcounting debug junk
- one more of these still left behind.. pff
Rakesh Pandit [Wed, 14 Apr 2010 11:14:52 +0000 (16:44 +0530)]
Eliminate unused "known problems" based filtering from rpmtsRun()
- This mechanism has never been used by anything at all, which suggests
its not a particularly useful feature
- Removing unused rpmpsTrim() from rpmps
- Marking okProbs parameter of rpmtsRun() function as unused to avoid
unnecessary breakage
Alexey Tourbin [Wed, 21 Apr 2010 07:10:15 +0000 (11:10 +0400)]
rpmdb.c (dbiOpen): fix dbix condition
In the loop, dbix is always initialized to 0, so checking (dbix < 0)
is uselss. Insetead, we sould check (dbix >= dbiTagsMax).
Panu Matilainen [Thu, 15 Apr 2010 07:24:03 +0000 (10:24 +0300)]
Only check the type of dependency requested in checkInstDeps()
- This gets called separately for requires, conflicts and obsoletes,
wth should it loop over conflicts when looking for requires?
Panu Matilainen [Wed, 14 Apr 2010 09:24:37 +0000 (12:24 +0300)]
Add a basic reinstall test to test-suite
- this currently fails due to commit
e16695e932e294ec600e673d38c41bde20931204,
on reinstall element header vs db header mismatch, adding this as
a reminder to fix it...
Panu Matilainen [Wed, 14 Apr 2010 06:22:07 +0000 (09:22 +0300)]
...but make sure the pre-requisite isn't also an erasure-dependency
- fixes eg Requires(pre,postun) case which must not be filtered out here
Panu Matilainen [Wed, 14 Apr 2010 06:05:55 +0000 (09:05 +0300)]
Pre-requisites of installed packages dont need to be verified (RhBug:223642)
Panu Matilainen [Tue, 13 Apr 2010 05:50:08 +0000 (08:50 +0300)]
Update a few tag descriptions + visibility
- RPMTAG_CAPABILITY and RPMTAG_RHNPLATFORM are truly obsolete and
should've been taken out a long time ago
Panu Matilainen [Tue, 13 Apr 2010 05:35:45 +0000 (08:35 +0300)]
Use correct tag for NOPATCH
- dumb copy-paste error from commit
a317eaadc3f7b9b4ec373719e7d60fb8b98e2c21
Panu Matilainen [Tue, 13 Apr 2010 05:32:49 +0000 (08:32 +0300)]
RPMTAG_NOSOURCE and NOPATCH aren't internal, they end up in nosrc headers
Michael Schroeder [Mon, 12 Apr 2010 10:12:46 +0000 (12:12 +0200)]
Gnu tar sometimes prints a "tar: Record size = 16" message, breaking
rpmbuild -ta.
Michael Schroeder [Mon, 12 Apr 2010 10:11:56 +0000 (12:11 +0200)]
Adapt the Makefile as rpmdb_svc can no longer be built.
Michael Schroeder [Mon, 12 Apr 2010 10:11:05 +0000 (12:11 +0200)]
New gcc versions comlpain about undefined behaviour...
Michael Schroeder [Mon, 12 Apr 2010 10:10:20 +0000 (12:10 +0200)]
readLineFromOFI may modify the fileStack, thus we have to re-set
ofi after calling it
Michael Schroeder [Mon, 12 Apr 2010 10:09:04 +0000 (12:09 +0200)]
Do not load keyring if signature checking is disabled.
Michael Schroeder [Mon, 12 Apr 2010 10:07:01 +0000 (12:07 +0200)]
Seems like a change was made to make %attr(-) go back to the
defattr setting. Unfortunatelly this broke %defattr(-).
Panu Matilainen [Fri, 9 Apr 2010 13:36:17 +0000 (16:36 +0300)]
Clean up dbiOpen() a bit
- eliminate now redundant rc-condition, move local variables to the
scope their used in
Panu Matilainen [Fri, 9 Apr 2010 12:52:29 +0000 (15:52 +0300)]
Move environment open "refcounting" to db_init() / db_fini()
Panu Matilainen [Fri, 9 Apr 2010 12:06:59 +0000 (15:06 +0300)]
And now lose db_use_env from rpmdb struct
- Environment is always used, shared if possible private otherwise
Panu Matilainen [Fri, 9 Apr 2010 11:53:31 +0000 (14:53 +0300)]
Simplify the dbenv->open() flags shuffle
- We always create an environment. If we dont have permissions to create
or join a shared environment, we use a private environment. Instead
of trying to figure out what to do beforehand, retry dbenv->open()
with different flags to see if it succeeds. This eliminates some
potential races when others might create/remove the environment
while we're pondering about appropriate flags.
- Lose the "create" bdb config option, this is something we always
want to decide internally.
- Remove "force" bdb config option, DB_FORCE is dbenv->remove() option
and its value clashes with DB_CREATE...
Panu Matilainen [Fri, 9 Apr 2010 08:16:37 +0000 (11:16 +0300)]
Move dbi_use_dbenv from dbi to rpmdb, this is a per-rpmdb setting