Panu Matilainen [Thu, 8 Apr 2010 06:10:13 +0000 (09:10 +0300)]
Eliminate dbiIndexNewItem()
- this is just simple struct with no special allocation, and contents
are memcpy()'ed so theres no need to malloc it
Panu Matilainen [Wed, 7 Apr 2010 10:23:03 +0000 (13:23 +0300)]
Change nextInstance() to permit retrieval too, use it for bitmap alloc
- The "was this verified already" bitmap uses header instance numbers
which are monotonically increasing and can be much much larger than
the number of actually installed headers. Grab the current instance
number for a better idea how much we'll at least need.
- Throw out the db stats based dbiNumKeys(), this doesn't have the
kind of performance penalty that stats have and suits our purposes
better anyway
Panu Matilainen [Wed, 7 Apr 2010 09:43:20 +0000 (12:43 +0300)]
Make rpmdbAdd() and rpmdbRemove() internal-only
- all additions and removals to the rpmdb need to go through the
transaction machinery, these are very low-level functions that
API users have no business messing with
Panu Matilainen [Wed, 7 Apr 2010 09:38:20 +0000 (12:38 +0300)]
Stash rpmdb struct definition out of sight into dbi.h
- not perhaps the ideal place for it but the backend level needs
access to the main db in several places anyway, and this leaves
rpmdb_internal.h clear for internal-only APIs
Panu Matilainen [Wed, 7 Apr 2010 09:28:02 +0000 (12:28 +0300)]
Unify rpmdbAdd/Remove() interfaces
- rpmdbRemove() takes now a header as argument too - we need both
the header number and the header itself there anyway, so might as
well use the header we already have instead of flipping through backwards
hoops to get to it
Panu Matilainen [Wed, 7 Apr 2010 09:12:51 +0000 (12:12 +0300)]
Eliminate the headerCheck() foo from rpmdbAdd/Remove()
- for rpmdbRemove() these have been completely unused, and for rpmdbAdd()
the higher levels have had more than one chance of verifying the
header if checking is enabled in the transaction set
Panu Matilainen [Wed, 7 Apr 2010 08:11:55 +0000 (11:11 +0300)]
Lift RPMDBI_PACKAGES handling from db add/remove to separate helper
- Remove vs add is rather similar, unify the copy-pasteish code
- Be a bit more careful about error returns from the primary db functions
Panu Matilainen [Wed, 7 Apr 2010 06:18:44 +0000 (09:18 +0300)]
Lift header instance number figuring out of rpmdbAdd() to helper function
Panu Matilainen [Wed, 7 Apr 2010 05:58:23 +0000 (08:58 +0300)]
Verify header sanity as the first thing in rpmdbAdd()
- avoid allocating header instance for something that we'd fail to add
- avoid possibly adding data to indexes even if adding the header
itself fails, duh
Panu Matilainen [Wed, 7 Apr 2010 05:35:14 +0000 (08:35 +0300)]
Move header INSTALLTID manipulation out of rpmdbAdd()
Panu Matilainen [Wed, 7 Apr 2010 04:48:33 +0000 (07:48 +0300)]
Remove unused require- and provideversion indexes
- For completely unused indexes, these are fairly expensive too
Panu Matilainen [Wed, 7 Apr 2010 03:55:15 +0000 (06:55 +0300)]
Eliminate redundant/bogus temporary variable shuffling
- This fetches the special key zero which is used for bookkeeping of
header "instance" numbers, the actual header to be stored is NOT
relevant here
- keyp, keylen, datap and datalen are just redundant fluff, manipulate
the key and data DBT directly
Panu Matilainen [Tue, 6 Apr 2010 10:13:14 +0000 (13:13 +0300)]
Remove unused _DBI defines
Panu Matilainen [Tue, 6 Apr 2010 10:09:34 +0000 (13:09 +0300)]
Lose unused fluff from openDatabase()
- As RPMDBI_PACKAGES is always the first index, the for-loop never
did anything else than open the Packages db and exit with success/fail.
For the same reason RPMDB_FLAG_MINIMAL didn't do anything at all here,
lose the unnecessary flag..
Panu Matilainen [Tue, 6 Apr 2010 10:02:58 +0000 (13:02 +0300)]
Eliminate static _rebuildinprogress hack, use per-db flag instead
- replace unused RPMDB_FLAG_CHROOT with RPMDB_FLAG_REBUILD and pass
around as necessary
Panu Matilainen [Tue, 6 Apr 2010 09:59:06 +0000 (12:59 +0300)]
Bury rpmdb flags inside rpmdb.c, not used by anything outside it
Panu Matilainen [Mon, 5 Apr 2010 15:09:57 +0000 (18:09 +0300)]
Unbreak --rebuilddb
- managed to bust up the rebuilddb-path generation once again in commit
890cea56c91ff6ff887a439c036d345715e6c5c2, doh
- make the intended logic more obviours
Yuri Chornoivan [Mon, 5 Apr 2010 07:22:31 +0000 (07:22 +0000)]
l10n: Updates to Ukrainian (uk) translation
Transmitted-via: Transifex (www.transifex.net)
Misha Shnurapet [Fri, 2 Apr 2010 12:55:37 +0000 (12:55 +0000)]
l10n: Updates to Russian (ru) translation
Transmitted-via: Transifex (www.transifex.net)
Panu Matilainen [Thu, 1 Apr 2010 15:19:49 +0000 (18:19 +0300)]
Nuke bogus comments + related NULL-checks
- db->_dbi cannot be NULL if db is not NULL, it's allocated at
newRpmDB() and freed in rpmdbClose() and nothing else touches it
Panu Matilainen [Thu, 1 Apr 2010 14:59:38 +0000 (17:59 +0300)]
Clean up rpmdb struct initialization
- lose the dumb template assignment and related defines
Panu Matilainen [Thu, 1 Apr 2010 14:49:07 +0000 (17:49 +0300)]
Lose unused static initialization foo
Panu Matilainen [Thu, 1 Apr 2010 13:58:12 +0000 (16:58 +0300)]
Clean up formatting in rpmdbCountPackages() (just cosmetics)
Panu Matilainen [Thu, 1 Apr 2010 13:29:38 +0000 (16:29 +0300)]
Remove unused rpmdbCloseDBI()
- this was used by the Depends temporary db hack, forcing it
to be exposed in the API... not anymore
Panu Matilainen [Thu, 1 Apr 2010 12:28:45 +0000 (15:28 +0300)]
Generate package database statistics on close & make use of it on open
- Turn dbiStat() into more useful: return the number of keys in the
index, hiding away the BDB internal access method stuff into the backend
- Force statistics gathering at Packages db close, take advantage of
that when its opened to get a fairly accurate count of packages for
initial "header verified" bitmap allocation. Previously DB_FAST_STAT
was used on open but it never returns anything when no stats have
been previously collected, hence the need for the expensive slow stat.
- The performance hit from stat generation is hardly worth it for
the bitmap allocation alone, but lets see if there are other uses...
- Also gets rid of dbi_stats member, this is not particularly useful
Panu Matilainen [Thu, 1 Apr 2010 10:54:39 +0000 (13:54 +0300)]
Lose "support" for truly ancient BDB versions
- BDB >= 4.3 is required now, and even thats several years old
Panu Matilainen [Thu, 1 Apr 2010 09:48:17 +0000 (12:48 +0300)]
Rename dbiOpen() + dbiOpenDB() to make their layer obvious from the name
- dbiOpen() in rpmdb.c operates on "rpmdb layer" so calling it
rpmdbOpenIndex(), dbiOpenDB() on the other hand is the lowest level
backend thing, which is now called dbiOpen() to be in line with
the other operations like dbiClose()
Panu Matilainen [Thu, 1 Apr 2010 09:34:59 +0000 (12:34 +0300)]
Pass rpmdb to dbiFindByLabel() / rpmdbiFindMatch() directly
- avoids having to go backwards from the dbi to rpmdb when we very well
have the rpmdb handle at hand here...
Panu Matilainen [Thu, 1 Apr 2010 09:25:04 +0000 (12:25 +0300)]
Eliminate dbi_rpmtag member from dbiIndex struct
- the backend doesn't care about the tag beyond initialization,
and the tag is only used for error messages from rpmdb layer which
already knows what the tag is
Panu Matilainen [Thu, 1 Apr 2010 09:01:28 +0000 (12:01 +0300)]
Define + use enum for primary/secondary index types + api to get it
- no functional changes, making the type more obvious than
"case 2*sizeof(int32_t):"
Panu Matilainen [Thu, 1 Apr 2010 08:47:15 +0000 (11:47 +0300)]
Rename the dbi access method member dbi_type -> dbi_dbtype
- make it more obvious what it is
Panu Matilainen [Thu, 1 Apr 2010 08:13:37 +0000 (11:13 +0300)]
Split dbiIndex out of rpmdb_internal.h into header of its own
- the dbi presents an internal api of its own, and deserves a separate
header (baby steps to making dbiIndex opaque outside the backend)
- move dbiVerify() to the backend where it belongs
- mark all the dbiFoo() functions as internal
Panu Matilainen [Thu, 1 Apr 2010 07:25:55 +0000 (10:25 +0300)]
Eliminate unnecessary argument to dbiCopen()
- txnid (which is unused...) is stored in dbi so dbiCopen() can already
get to it
Panu Matilainen [Thu, 1 Apr 2010 07:18:05 +0000 (10:18 +0300)]
Move dbi_no_dbsync logic to inside dbiSync()
- avoids having to check for it in every single caller
Panu Matilainen [Thu, 1 Apr 2010 07:03:21 +0000 (10:03 +0300)]
On header iteration format queries, dump the contents as-is
- This makes eg --xml query of packages give out the true contents
of the package, raw i18n strings and all, which is what you probably
want when examining the contents on this level. Regular queries still
run with all bells 'n whistless enabled.
Panu Matilainen [Thu, 1 Apr 2010 06:15:06 +0000 (09:15 +0300)]
Disable header i18n lookup in headerCopyTags() (RhBug:578299)
- without HEADERGET_RAW here, only one of the translations in header
(whatever happens to be "current" locale) gets copied
Panu Matilainen [Wed, 31 Mar 2010 09:17:00 +0000 (12:17 +0300)]
Remove BDB recno + queue access methods from configuration
- these rather limited access methods have never been used for anything,
rpm always used hash access and later btree to varying degree
Panu Matilainen [Tue, 30 Mar 2010 23:51:39 +0000 (02:51 +0300)]
Database metadata integer byte order is not user serviceable config
Panu Matilainen [Tue, 30 Mar 2010 23:48:03 +0000 (02:48 +0300)]
Lose dbi file configuration option
- index names are rpm tag names, period
- also we dont need to strdup() the tag name, rpmTagGetName() returns
pointers to const strings these days
Panu Matilainen [Tue, 30 Mar 2010 23:47:33 +0000 (02:47 +0300)]
Make dbiVerify() and dbiOpen() static, not needed outside rpmdb.c
Panu Matilainen [Tue, 30 Mar 2010 23:37:13 +0000 (02:37 +0300)]
Bury dbiIndexItem and dbiIndexSet structs inside rpmdb.c
Panu Matilainen [Tue, 30 Mar 2010 23:34:21 +0000 (02:34 +0300)]
Make dbiIndexSet operations static inside rpmdb.c
Panu Matilainen [Tue, 30 Mar 2010 23:16:54 +0000 (02:16 +0300)]
Lose the BDB access method configuration options
- In the ~10 years of their existance none of them have been used,
and besides users have no business messing with fundamental database
behavior such as whether duplicates are allowed or not
Panu Matilainen [Tue, 30 Mar 2010 23:04:21 +0000 (02:04 +0300)]
We really dont care about macro expanded per-db error prefix
Panu Matilainen [Tue, 30 Mar 2010 22:59:27 +0000 (01:59 +0300)]
Lose shmkey + shared BDB config options
- we dont want anything to do with System V IPC, really... BDB uses
file backed mmap when it needs, this suits us just fine
Panu Matilainen [Tue, 30 Mar 2010 22:55:06 +0000 (01:55 +0300)]
Remove "truncate" BDB config option
- Truncate is so dangerous that dbiOpenDB() discarded it even if
set in configuration. How useful is that?
Panu Matilainen [Tue, 30 Mar 2010 22:53:12 +0000 (01:53 +0300)]
We dont want to use DB_HOME environment, ever
- rpm has its own switches for overriding the db home when that's
necessary, DB_HOME it should use not
Panu Matilainen [Tue, 30 Mar 2010 22:45:26 +0000 (01:45 +0300)]
We dont have any temporary databases anymore, lose the config switches
- The last temporary db was Depends which is now replaced with an
in-memory hash, and the temporary dbs were troublesome with
chroot operations anyway, good riddance
Panu Matilainen [Tue, 30 Mar 2010 22:41:04 +0000 (01:41 +0300)]
Lose pointless per-db and per-dbi errfile and errcall pointers
- BDB uses stderr for errfile when not set, which is what we set it
to through various hoops. Just leave it alone.
- db_errcall has always been (intentionally) NULL to disable that output,
the errors get logged though cvtdberr() with slight filtering
Panu Matilainen [Tue, 30 Mar 2010 22:27:04 +0000 (01:27 +0300)]
Shred a pile of BDB config goo that's never been actually implemented
- with the exception subfile (used by now extinct sqlite backend)
none of this was ever even connected to the bdb configuration...
Panu Matilainen [Tue, 30 Mar 2010 22:12:15 +0000 (01:12 +0300)]
Shed the pointless per-dbi memory allocators
- we want to use our own allocators for exit-on-enomem behavior but
having per-index allocator configuration makes no sense whatsoever
Panu Matilainen [Tue, 30 Mar 2010 22:09:51 +0000 (01:09 +0300)]
Eliminate now unnecessary db_ndbi field, number of indexes is constant
Panu Matilainen [Tue, 30 Mar 2010 22:05:03 +0000 (01:05 +0300)]
Make rpmdb index list hard-wired
- We dont grow new indexes every other day, and especially this
is not activity that users need to be able to do
- Gets rid of the hysterical initialization and million can't happen
NULL-checks
Panu Matilainen [Tue, 30 Mar 2010 21:43:34 +0000 (00:43 +0300)]
Minor cleanup to rpmdbAdd()
- move dbiOpen() + check to more sensible place, curing the
busted indentation
- initialize at declaration save a few unnecessary lines
Panu Matilainen [Tue, 30 Mar 2010 21:40:23 +0000 (00:40 +0300)]
Add + use a helper function for retrieving single header at offset
Panu Matilainen [Tue, 30 Mar 2010 20:23:53 +0000 (23:23 +0300)]
Minor cleanups to rpmdbAdd() and rpmdbRemove()
Panu Matilainen [Tue, 30 Mar 2010 20:11:21 +0000 (23:11 +0300)]
Minor cleanup to rpmdbRebuild(), use rstreq() instead of hand-comparing
Panu Matilainen [Mon, 29 Mar 2010 15:36:21 +0000 (18:36 +0300)]
Hide rpmte open/close/error etc logic inside rpmteProcess()
- make a bunch of helper functions static now that they're not needed
elsewhere, rpmte is slowly becoming self-aware ;)
Panu Matilainen [Mon, 29 Mar 2010 14:14:55 +0000 (17:14 +0300)]
Move the little thats left of psm.h into rpmte_internal.h
- no functional changes, just shuffling declares around
Piotr Drąg [Mon, 29 Mar 2010 14:38:09 +0000 (14:38 +0000)]
l10n: Updates to Polish (pl) translation
Transmitted-via: Transifex (www.transifex.net)
Panu Matilainen [Mon, 29 Mar 2010 13:14:50 +0000 (16:14 +0300)]
Oops, scriptlet related error messages had dropped out of potfiles
Panu Matilainen [Mon, 29 Mar 2010 13:11:08 +0000 (16:11 +0300)]
Update translatios again to shed non-existent messages from recent churn
Panu Matilainen [Mon, 29 Mar 2010 13:05:32 +0000 (16:05 +0300)]
Lazy allocation of rpmds color array
- only requires and provides in transaction elements get colored,
dont allocate unused memory
Panu Matilainen [Mon, 29 Mar 2010 12:33:03 +0000 (15:33 +0300)]
Lose long since unused fdReadable() and fdWritable()
- these have been unused since rpm 4.6.0, and can be implemented
without access to rpmio internals too if somebody cares...
Panu Matilainen [Mon, 29 Mar 2010 12:27:43 +0000 (15:27 +0300)]
Lose long since unused url control structure and related functions
- these have been unused since rpm 4.6.0, and rpm is not in the
url business, rip.
Panu Matilainen [Mon, 29 Mar 2010 12:03:14 +0000 (15:03 +0300)]
Remove unused dependency set file reference count
- this has never been used for anything at all, there are better things
to use our memory for than unused arrays
Panu Matilainen [Mon, 29 Mar 2010 11:53:10 +0000 (14:53 +0300)]
Remove unused and non-sensible depedency set build-time
- Buildtime was never part of rpm version comparison nor should it be,
on rpm level the only sensible differentiator between two identical
dependencies is the dependency color if any. Other than that, random()
just as "appropriate" as build time.
Panu Matilainen [Mon, 29 Mar 2010 11:40:40 +0000 (14:40 +0300)]
headerNextTag() fits the checkForDuplicates() use-case better
- as we're only looking at tags, not their contents... avoid unnecessary
data shuffling and the code is a little simpler too
Panu Matilainen [Mon, 29 Mar 2010 07:59:58 +0000 (10:59 +0300)]
Unbreak cpio archive path generation
- fixes regression from commit
45e2d4e1d0260a9c07eb1daf0c68b82f602ad16f
- striplen was never set on build at all, and the psm part was unused
too - either unused since forever or possibly related to repackaging
Panu Matilainen [Mon, 29 Mar 2010 06:35:25 +0000 (09:35 +0300)]
Oops, being a bit too strict on dependency qualifiers
- Forgot how nasty hack parseBits() was... its abusing multilang
support and the multilang support passes "C" if no lang is specified,
which certainly isn't a valid qualifier. Make multilang vs tag with
optional qualifiers different types in PreambleRec, which allows
handling this more sanely
- Additionally build-dependencies dont take any qualifiers now
Panu Matilainen [Mon, 29 Mar 2010 05:58:27 +0000 (08:58 +0300)]
Teach rpm about pre-transaction dependencies
- A pre-transaction dependency is generally anything that must be
available at the start of the transaction, and cannot be resolved
by packages *in* the transaction. This lets %pretrans scriptlet
dependencies be expressed correctly, and could be also used for
other kinds of pre-conditions.
- rpmlib() dependencies are a special case of pre-trans dependencies
but leaving them handled separately as they cannot be provided by
anything in rpmdb either, whereas pretrans dependencies can.
Panu Matilainen [Mon, 29 Mar 2010 05:30:18 +0000 (08:30 +0300)]
Always fail build on unknown dependency qualifiers
- previously unknown qualifiers would go unnoticed if they happened
be last (including being the only qualifier), eg Requires(pre,junk)
or Requires(junk) would be cheerfully accepted but Requires(junk,pre) not.
Panu Matilainen [Mon, 29 Mar 2010 05:27:35 +0000 (08:27 +0300)]
Always initialize tagflags to RPMSENSE_ANY
Panu Matilainen [Mon, 29 Mar 2010 04:20:33 +0000 (07:20 +0300)]
Report all enabled dependency bits in deptype format extension
- dont make assumptions about which bits can be enabled simultaneously,
just dump 'em all
Panu Matilainen [Mon, 29 Mar 2010 03:48:23 +0000 (06:48 +0300)]
Dont bother translators with dependency lookup debug messages
Panu Matilainen [Mon, 29 Mar 2010 03:37:27 +0000 (06:37 +0300)]
Lose unused RPMSENSE_PATCHES
- this was never used by upstream at least, and the "patch rpms" this
refers to was abandoned by Suse in favor of deltarpm a long time ago...
Panu Matilainen [Mon, 29 Mar 2010 03:29:19 +0000 (06:29 +0300)]
Remove redundant assignments
- everything but buildrequires goes to pkg->header, no point having
this separately set in each case
Panu Matilainen [Mon, 29 Mar 2010 03:27:44 +0000 (06:27 +0300)]
Group the tag cases a bit more logically in parseRCPOT()
Panu Matilainen [Mon, 29 Mar 2010 03:17:30 +0000 (06:17 +0300)]
Drop "support" for per-scriptlet build-requires
- Rpm has never done anything useful with "BuildRequires(prep): foo"
style dependencies other than recorded them in src.rpms, nor is there much
point in properly supporting this in the future either
- Frees up four more bits from rpmsenseFlags for better uses
Panu Matilainen [Mon, 29 Mar 2010 02:59:54 +0000 (05:59 +0300)]
Free up three bits from rpmsenseFlags
- pass the wanted dependency type to addReqProv() as tagN argument
(previously unused, eh...) instead of taking up three precious bits
for internal-only purposes from a constrained bitfield
Panu Matilainen [Fri, 26 Mar 2010 15:07:00 +0000 (17:07 +0200)]
Add ds.Instance() method to python bindings
- the naming between header/rpmte/rpmds "instance" is wonderfully
inconsistent... oh well
Panu Matilainen [Fri, 26 Mar 2010 15:00:20 +0000 (17:00 +0200)]
Take advantage of rpmdsCurrent() in python bindings
Panu Matilainen [Fri, 26 Mar 2010 14:54:16 +0000 (16:54 +0200)]
Add public rpmdsCurrent() function for creating a single ds from iteration
- Permits extracting single dependencies from a set with instance
number inherited without exposing rpmdsSetInstance() kind of thing
- Also faster than calling rpmdsSingle(rpmdsTagN(ds), rpmdsN(ds)....)
where each call re-re-re-re-validates the indexes
Panu Matilainen [Fri, 26 Mar 2010 14:45:46 +0000 (16:45 +0200)]
Split single ds generation into helper function, clean up a bit
- allows nicer handling of ds instance without special cases
Panu Matilainen [Fri, 26 Mar 2010 14:14:52 +0000 (16:14 +0200)]
Simplify rpmcliTransaction() a bit
- No need for separate the install and erase cases here for problem
filter flags: we only check disk space for added packages so this
is just a no-op anyway
- Also no need to fiddle with eflags for ordering, we always want
ordering unless --noorder specified
Héctor Daniel Cabrera [Fri, 26 Mar 2010 13:51:00 +0000 (13:51 +0000)]
l10n: Updates to Spanish (Castilian) (es) translation
Transmitted-via: Transifex (www.transifex.net)
Panu Matilainen [Fri, 26 Mar 2010 13:28:25 +0000 (15:28 +0200)]
Oops, diskspace problem messed up from the argument shuffle
- fixes regression from commit
12392d93e2006ac63f918a154f29283a14fb98cf
Panu Matilainen [Fri, 26 Mar 2010 13:22:13 +0000 (15:22 +0200)]
Record originating header instance to all problems where relevant
- currently unused but allows tracking things like file conflicts
back to the originating header directly from the problem set
Panu Matilainen [Fri, 26 Mar 2010 13:11:57 +0000 (15:11 +0200)]
Take advantage of rpmdsInstance() in dependency problem reporting
- Other than having less arguments to pass here and there, doesn't
give any immediate advantage, but with the origin of installed
dependency stored in problem sets, it'd be possible to track back
a dependency problem back to the originating header
Panu Matilainen [Fri, 26 Mar 2010 12:56:14 +0000 (14:56 +0200)]
Record header instance in dependency sets + add API for retrieving
Panu Matilainen [Fri, 26 Mar 2010 12:18:45 +0000 (14:18 +0200)]
Merge checkPackageSet() into checkInstDeps()
- checkPackageSet() got refactored out of being useful standalone
while I wasn't looking :)
Panu Matilainen [Fri, 26 Mar 2010 11:36:36 +0000 (13:36 +0200)]
Avoid unnecessary argument now that problem pkgNEVRA is always te NEVRA
Panu Matilainen [Fri, 26 Mar 2010 11:20:17 +0000 (13:20 +0200)]
Change the problem data arrangement for dependency problems
- pkgNEVR in problems is now always the NEVR of the transaction element
triggering the problem, and altNEVR is the other affected package,
dependency string is stored in the problem string attribute
- no user visible changed, except for somebody crazy enough to try to
do something other than print the problem message strings
- we wouldn't really need to strdup() the pkgNEVR in problems now,
but leaving that alone for the moment...
Panu Matilainen [Fri, 26 Mar 2010 10:23:47 +0000 (12:23 +0200)]
Filter out duplicate problems when adding to element problem sets
- Problems associated with a transaction element are necessarily unique
to that element, so when filtered there we don't have to worry about
skipping dupes elsewhere like in merged sets. This can actually lead
to apparent duplicates in the current problem report output (eg in cases
where multiple packages provide the same dependency which would be
removed, like multilib packages), but this is only an artifact of
they way the problems are currently printed out.
- While this is still a dumb linear search, it can be several seconds
faster than the previous filtering in rpmpsPrint(), which is now
just a dumb convenience function.
Panu Matilainen [Fri, 26 Mar 2010 09:22:18 +0000 (11:22 +0200)]
Eliminate rpmpsAppend() from the API
- This was only a convenience function that shouldn't really have been
exported to begin with, and is not necessary at all - bury it
inside rpmte.c for now to hide lazy problem set allocation
- In the other news of the day... According to git, this happens to be
the 10000th commit to the master branch of this codebase.
Happy anniversary to rpm :P
Panu Matilainen [Fri, 26 Mar 2010 08:41:10 +0000 (10:41 +0200)]
Push problem creation argument sanitation to the level below
- rpmpsAppend() and rpmProblemCreate() have no use for fn/dn either,
its just stored in one string internally
- this is kinda unnecessary API break but these aren't used outside
rpm so it doesn't make much difference
Panu Matilainen [Fri, 26 Mar 2010 08:27:39 +0000 (10:27 +0200)]
Sanitize rpmteAddProblem() arguments
- lump fn/dn into one, in no case both are needed
- move altNEVR earlier to lump the generic attribute foo last
Panu Matilainen [Fri, 26 Mar 2010 07:31:00 +0000 (09:31 +0200)]
Use the new problem set iterator where it makes more sense
Panu Matilainen [Fri, 26 Mar 2010 07:21:06 +0000 (09:21 +0200)]
Use rpmpsMerge() for collecting element problems in rpmtsProblems()
Panu Matilainen [Fri, 26 Mar 2010 07:20:05 +0000 (09:20 +0200)]
Add a simple and dumb rpmpsMerge() function to merge two problem sets