data_source.cpp

来自「ncbi源码」· C++ 代码 · 共 1,528 行 · 第 1/3 页

            if ( sub_info.IsSet() &&
                 sub_info.GetSet().GetClass() == CBioseq_set::eClass_parts ) {
                switch (level) {
                case CBioseq_CI_Base::eLevel_Mains:
                    // Skip parts
                    continue;
                case CBioseq_CI_Base::eLevel_Parts:
                    // Allow adding bioseqs from lower levels
                    local_level = CBioseq_CI_Base::eLevel_All;
                    break;
                default:
                    break;
                }
            }
            x_CollectBioseqs(sub_info, bioseqs, filter, local_level);
        }
    }
}


void CDataSource::Prefetch(CPrefetchToken_Impl& token)
{
#if !defined(NCBI_NO_THREADS)
    if (!m_PrefetchThread) {
        CFastMutexGuard guard(m_PrefetchLock);
        // Check againi
        if (!m_PrefetchThread) {
            m_PrefetchThread.Reset(new CPrefetchThread(*this));
            m_PrefetchThread->Run();
        }
    }
    _ASSERT(m_PrefetchThread);
    m_PrefetchThread->AddRequest(token);
#endif
}


void CDataSource::DebugDump(CDebugDumpContext /*ddc*/,
                            unsigned int /*depth*/) const
{
}


END_SCOPE(objects)
END_NCBI_SCOPE

/*
* ---------------------------------------------------------------------------
* $Log: data_source.cpp,v $
* Revision 1000.4  2004/06/01 19:23:11  gouriano
* PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.133
*
* Revision 1.133  2004/05/21 21:42:12  gorelenk
* Added PCH ncbi_pch.hpp
*
* Revision 1.132  2004/04/16 13:31:47  grichenk
* Added data pre-fetching functions.
*
* Revision 1.131  2004/03/31 17:08:07  vasilche
* Implemented ConvertSeqToSet and ConvertSetToSeq.
*
* Revision 1.130  2004/03/24 18:30:29  vasilche
* Fixed edit API.
* Every *_Info object has its own shallow copy of original object.
*
* Revision 1.129  2004/03/16 15:47:27  vasilche
* Added CBioseq_set_Handle and set of EditHandles
*
* Revision 1.128  2004/02/25 19:23:46  shomrat
* Use of entry_info instead of tse_info to allow collection of bioseqs on non-tse entries
*
* Revision 1.127  2004/02/03 19:02:17  vasilche
* Fixed broken 'dirty annot index' state after RemoveEntry().
*
* Revision 1.126  2004/02/02 15:27:20  dicuccio
* +<algorithm> for find
*
* Revision 1.125  2004/02/02 14:46:43  vasilche
* Several performance fixed - do not iterate whole tse set in CDataSource.
*
* Revision 1.124  2003/12/18 16:38:06  grichenk
* Added CScope::RemoveEntry()
*
* Revision 1.123  2003/11/26 17:55:57  vasilche
* Implemented ID2 split in ID1 cache.
* Fixed loading of splitted annotations.
*
* Revision 1.122  2003/10/27 16:47:12  vasilche
* Fixed error:
* src/objmgr/data_source.cpp", line 913: Fatal: Assertion failed: (it != tse_map.end() && it->first == id)
*
* Revision 1.121  2003/10/07 13:43:23  vasilche
* Added proper handling of named Seq-annots.
* Added feature search from named Seq-annots.
* Added configurable adaptive annotation search (default: gene, cds, mrna).
* Fixed selection of blobs for loading from GenBank.
* Added debug checks to CSeq_id_Mapper for easier finding lost CSeq_id_Handles.
* Fixed leaked split chunks annotation stubs.
* Moved some classes definitions in separate *.cpp files.
*
* Revision 1.120  2003/09/30 16:22:02  vasilche
* Updated internal object manager classes to be able to load ID2 data.
* SNP blobs are loaded as ID2 split blobs - readers convert them automatically.
* Scope caches results of requests for data to data loaders.
* Optimized CSeq_id_Handle for gis.
* Optimized bioseq lookup in scope.
* Reduced object allocations in annotation iterators.
* CScope is allowed to be destroyed before other objects using this scope are
* deleted (feature iterators, bioseq handles etc).
* Optimized lookup for matching Seq-ids in CSeq_id_Mapper.
* Added 'adaptive' option to objmgr_demo application.
*
* Revision 1.119  2003/09/05 17:29:40  grichenk
* Structurized Object Manager exceptions
*
* Revision 1.118  2003/09/03 20:00:02  grichenk
* Added sequence filtering by level (mains/parts/all)
*
* Revision 1.117  2003/08/14 20:05:19  vasilche
* Simple SNP features are stored as table internally.
* They are recreated when needed using CFeat_CI.
*
* Revision 1.116  2003/08/04 17:04:31  grichenk
* Added default data-source priority assignment.
* Added support for iterating all annotations from a
* seq-entry or seq-annot.
*
* Revision 1.115  2003/07/17 22:51:31  vasilche
* Fixed unused variables warnings.
*
* Revision 1.114  2003/07/17 20:07:56  vasilche
* Reduced memory usage by feature indexes.
* SNP data is loaded separately through PUBSEQ_OS.
* String compression for SNP data.
*
* Revision 1.113  2003/07/09 17:54:29  dicuccio
* Fixed uninitialized variables in CDataSource and CSeq_annot_Info
*
* Revision 1.112  2003/07/01 18:01:08  vasilche
* Added check for null m_SeqMap pointer.
*
* Revision 1.111  2003/06/24 14:35:01  vasilche
* Fixed compilation in debug mode.
*
* Revision 1.110  2003/06/24 14:25:18  vasilche
* Removed obsolete CTSE_Guard class.
* Used separate mutexes for bioseq and annot maps.
*
* Revision 1.109  2003/06/19 18:23:45  vasilche
* Added several CXxx_ScopeInfo classes for CScope related information.
* CBioseq_Handle now uses reference to CBioseq_ScopeInfo.
* Some fine tuning of locking in CScope.
*
* Revision 1.104  2003/05/20 15:44:37  vasilche
* Fixed interaction of CDataSource and CDataLoader in multithreaded app.
* Fixed some warnings on WorkShop.
* Added workaround for memory leak on WorkShop.
*
* Revision 1.103  2003/05/14 18:39:28  grichenk
* Simplified TSE caching and filtering in CScope, removed
* some obsolete members and functions.
*
* Revision 1.102  2003/05/12 19:18:29  vasilche
* Fixed locking of object manager classes in multi-threaded application.
*
* Revision 1.101  2003/05/06 18:54:09  grichenk
* Moved TSE filtering from CDataSource to CScope, changed
* some filtering rules (e.g. priority is now more important
* than scope history). Added more caches to CScope.
*
* Revision 1.100  2003/05/05 20:59:47  vasilche
* Use one static mutex for all instances of CTSE_LockingSet.
*
* Revision 1.99  2003/04/29 19:51:13  vasilche
* Fixed interaction of Data Loader garbage collector and TSE locking mechanism.
* Made some typedefs more consistent.
*
* Revision 1.98  2003/04/24 16:12:38  vasilche
* Object manager internal structures are splitted more straightforward.
* Removed excessive header dependencies.
*
* Revision 1.97  2003/04/14 21:32:18  grichenk
* Avoid passing CScope as an argument to CDataSource methods
*
* Revision 1.96  2003/03/24 21:26:45  grichenk
* Added support for CTSE_CI
*
* Revision 1.95  2003/03/21 21:08:53  grichenk
* Fixed TTSE_Lock initialization
*
* Revision 1.93  2003/03/21 19:22:51  grichenk
* Redesigned TSE locking, replaced CTSE_Lock with CRef<CTSE_Info>.
*
* Revision 1.92  2003/03/18 14:52:59  grichenk
* Removed obsolete methods, replaced seq-id with seq-id handle
* where possible. Added argument to limit annotations update to
* a single seq-entry.
*
* Revision 1.91  2003/03/12 20:09:34  grichenk
* Redistributed members between CBioseq_Handle, CBioseq_Info and CTSE_Info
*
* Revision 1.90  2003/03/11 15:51:06  kuznets
* iterate -> ITERATE
*
* Revision 1.89  2003/03/11 14:15:52  grichenk
* +Data-source priority
*
* Revision 1.88  2003/03/10 16:55:17  vasilche
* Cleaned SAnnotSelector structure.
* Added shortcut when features are limited to one TSE.
*
* Revision 1.87  2003/03/05 20:56:43  vasilche
* SAnnotSelector now holds all parameters of annotation iterators.
*
* Revision 1.86  2003/03/03 20:31:08  vasilche
* Removed obsolete method PopulateTSESet().
*
* Revision 1.85  2003/02/27 14:35:31  vasilche
* Splitted PopulateTSESet() by logically independent parts.
*
* Revision 1.84  2003/02/25 20:10:40  grichenk
* Reverted to single total-range index for annotations
*
* Revision 1.83  2003/02/24 18:57:22  vasilche
* Make feature gathering in one linear pass using CSeqMap iterator.
* Do not use feture index by sub locations.
* Sort features at the end of gathering in one vector.
* Extracted some internal structures and classes in separate header.
* Delay creation of mapped features.
*
* Revision 1.82  2003/02/24 14:51:11  grichenk
* Minor improvements in annot indexing
*
* Revision 1.81  2003/02/13 14:34:34  grichenk
* Renamed CAnnotObject -> CAnnotObject_Info
* + CSeq_annot_Info and CAnnotObject_Ref
* Moved some members of CAnnotObject to CSeq_annot_Info
* and CAnnotObject_Ref.
* Added feat/align/graph to CAnnotObject_Info map
* to CDataSource.
*
* Revision 1.80  2003/02/05 17:59:17  dicuccio
* Moved formerly private headers into include/objects/objmgr/impl
*
* Revision 1.79  2003/02/04 22:01:26  grichenk
* Fixed annotations loading/indexing order
*
* Revision 1.78  2003/02/04 21:46:32  grichenk
* Added map of annotations by intervals (the old one was
* by total ranges)
*
* Revision 1.77  2003/01/29 22:03:46  grichenk
* Use single static CSeq_id_Mapper instead of per-OM model.
*
* Revision 1.76  2003/01/29 17:45:02  vasilche
* Annotaions index is split by annotation/feature type.
*
* Revision 1.75  2003/01/22 20:11:54  vasilche
* Merged functionality of CSeqMapResolved_CI to CSeqMap_CI.
* CSeqMap_CI now supports resolution and iteration over sequence range.
* Added several caches to CScope.
* Optimized CSeqVector().
* Added serveral variants of CBioseqHandle::GetSeqVector().
* Tried to optimize annotations iterator (not much success).
* Rewritten CHandleRange and CHandleRangeMap classes to avoid sorting of list.
*
* Revision 1.74  2002/12/26 16:39:24  vasilche
* Object manager class CSeqMap rewritten.
*
* Revision 1.73  2002/12/20 20:54:24  grichenk
* Added optional location/product switch to CFeat_CI
*
* Revision 1.72  2002/12/19 20:16:39  grichenk
* Fixed locations on minus strand
*
* Revision 1.71  2002/12/06 15:36:00  grichenk
* Added overlap type for annot-iterators
*
* Revision 1.70  2002/11/08 22:15:51  grichenk
* Added methods for removing/replacing annotations
*
* Revision 1.69  2002/11/08 21:03:30  ucko
* CConstRef<> now requires an explicit constructor.
*
* Revision 1.68  2002/11/04 21:29:12  grichenk
* Fixed usage of const CRef<> and CRef<> constructor
*
* Revision 1.67  2002/10/18 19:12:40  grichenk
* Removed mutex pools, converted most static mutexes to non-static.
* Protected CSeqMap::x_Resolve() with mutex. Modified code to prevent
* dead-locks.
*
* Revision 1.66  2002/10/16 20:44:16  ucko
* MIPSpro: use #pragma instantiate rather than template<>, as the latter
* ended up giving rld errors for CMutexPool_Base<*>::sm_Pool.  (Sigh.)
*
* Revision 1.65  2002/10/02 17:58:23  grichenk
* Added sequence type filter to CBioseq_CI
*
* Revision 1.64  2002/09/30 20:01:19  grichenk
* Added methods to support CBioseq_CI
*
* Revision 1.63  2002/09/16 19:59:36  grichenk
* Fixed getting reference to a gap on minus strand
*
* Revision 1.62  2002/09/10 19:55:49  grichenk
* Throw exception when unable to create resolved seq-map
*
* Revision 1.61  2002/08/09 14:58:50  ucko
* Restrict template <> to MIPSpro for now, as it also leads to link
* errors with Compaq's compiler.  (Sigh.)
*
* Revision 1.60  2002/08/08 19:51:16  ucko
* Omit EMPTY_TEMPLATE for GCC and KCC, as it evidently leads to link errors(!)
*
* Revision 1.59  2002/08/08 14:28:00  ucko
* Add EMPTY_TEMPLATE to explicit instantiations.
*
* Revision 1.58  2002/08/07 18:22:48  ucko
* Explicitly instantiate CMutexPool_Base<{CDataSource,TTSESet}>::sm_Pool
*
* Revision 1.57  2002/07/25 15:01:51  grichenk
* Replaced non-const GetXXX() with SetXXX()
*
* Revision 1.56  2002/07/08 20:51:01  grichenk
* Moved log to the end of file
* Replaced static mutex (in CScope, CDataSource) with the mutex
* pool. Redesigned CDataSource data locking.
*
* Revision 1.55  2002/07/01 15:44:51  grichenk
* Fixed typos
*
* Revision 1.54  2002/07/01 15:40:58  grichenk
* Fixed 'tse_set' warning.
* Removed duplicate call to tse_set.begin().
* Fixed version resolving for annotation iterators.
* Fixed strstream bug in KCC.
*
* Revision 1.53  2002/06/28 17:30:42  grichenk
* Duplicate seq-id: ERR_POST() -> THROW1_TRACE() with the seq-id
* list.
* Fixed x_CleanupUnusedEntries().
* Fixed a bug with not found sequences.
* Do not copy bioseq data in GetBioseqCore().
*
* Revision 1.52  2002/06/21 15:12:15  grichenk
* Added resolving seq-id to the best version
*
* Revision 1.51  2002/06/12 14:39:53  grichenk
* Performance improvements
*
* Revision 1.50  2002/06/04 17:18:33  kimelman
* memory cleanup :  new/delete/Cref rearrangements
*
* Revision 1.49  2002/05/31 17:53:00  grichenk
* Optimized for better performance (CTSE_Info uses atomic counter,
* delayed annotations indexing, no location convertions in
* CAnnot_Types_CI if no references resolution is required etc.)
*
* Revision 1.48  2002/05/29 21:21:13  gouriano
* added debug dump
*
* Revision 1.47  2002/05/28 18:00:43  gouriano
* DebugDump added
*
* Revision 1.46  2002/05/24 14:57:12  grichenk
* SerialAssign<>() -> CSerialObject::Assign()
*
* Revision 1.45  2002/05/21 18:57:25  grichenk
* Fixed annotations dropping
*
* Revision 1.44  2002/05/21 18:40:50  grichenk
* Fixed annotations droppping
*
* Revision 1.43  2002/05/14 20:06:25  grichenk
* Improved CTSE_Info locking by CDataSource and CDataLoader
*
* Revision 1.42  2002/05/13 15:28:27  grichenk
* Fixed seqmap for virtual sequences
*
* Revision 1.41  2002/05/09 14:18:15  grichenk
* More TSE conflict resolving rules for annotations
*
* Revision 1.40  2002/05/06 03:28:46  vakatov
* OM/OM1 renaming
*
* Revision 1.39  2002/05/03 21:28:09  ucko
* Introduce T(Signed)SeqPos.
*
* Revision 1.38  2002/05/02 20:42:37  grichenk
* throw -> THROW1_TRACE
*
* Revision 1.37  2002/04/22 20:05:08  grichenk
* Fixed minor bug in GetSequence()
*
* Revision 1.36  2002/04/19 18:02:47  kimelman
* add verify to catch coredump
*
* Revision 1.35  2002/04/17 21:09:14  grichenk
* Fixed annotations loading
* +IsSynonym()
*
* Revision 1.34  2002/04/11 18:45:39  ucko
* Pull in extra headers to make KCC happy.
*
* Revision 1.33  2002/04/11 12:08:21  grichenk
* Fixed GetResolvedSeqMap() implementation
*
* Revision 1.32  2002/04/05 21:23:08  grichenk
* More duplicate id warnings fixed
*
* Revision 1.31  2002/04/05 20:27:52  grichenk
* Fixed duplicate identifier warning
*
* Revision 1.30  2002/04/04 21:33:13  grichenk
* Fixed GetSequence() for sequences with unresolved segments
*
* Revision 1.29  2002/04/03 18:06:47  grichenk
* Fixed segmented sequence bugs (invalid positioning of literals
* and gaps). Improved CSeqVector performance.
*
* Revision 1.27  2002/03/28 14:02:31  grichenk
* Added scope history checks to CDataSource::x_FindBestTSE()
*
* Revision 1.26  2002/03/22 17:24:12  gouriano
* loader-related fix in DropTSE()
*
* Revision 1.25  2002/03/21 21:39:48  grichenk
* garbage collector bugfix
*
* Revision 1.24  2002/03/20 21:24:59  gouriano
* *** empty log message ***
*
* Revision 1.23  2002/03/15 18:10:08  grichenk
* Removed CRef<CSeq_id> from CSeq_id_Handle, added
* key to seq-id map th CSeq_id_Mapper
*
* Revision 1.22  2002/03/11 21:10:13  grichenk
* +CDataLoader::ResolveConflict()
*
* Revision 1.21  2002/03/07 21:25:33  grichenk
* +GetSeq_annot() in annotation iterators
*
* Revision 1.20  2002/03/06 19:37:19  grichenk
* Fixed minor bugs and comments
*
* Revision 1.19  2002/03/05 18:44:55  grichenk
* +x_UpdateTSEStatus()
*
* Revision 1.18  2002/03/05 16:09:10  grichenk
* Added x_CleanupUnusedEntries()
*
* Revision 1.17  2002/03/04 15:09:27  grichenk
* Improved MT-safety. Added live/dead flag to CDataSource methods.
*
* Revision 1.16  2002/02/28 20:53:31  grichenk
* Implemented attaching segmented sequence data. Fixed minor bugs.
*
* Revision 1.15  2002/02/25 21:05:29  grichenk
* Removed seq-data references caching. Increased MT-safety. Fixed typos.
*
* Revision 1.14  2002/02/21 19:27:05  grichenk
* Rearranged includes. Added scope history. Added searching for the
* best seq-id match in data sources and scopes. Updated tests.
*
* Revision 1.13  2002/02/20 20:23:27  gouriano
* corrected FilterSeqid()
*
* Revision 1.12  2002/02/15 20:35:38  gouriano
* changed implementation of HandleRangeMap
*
* Revision 1.11  2002/02/12 19:41:42  grichenk
* Seq-id handles lock/unlock moved to CSeq_id_Handle 'ctors.
*
* Revision 1.10  2002/02/07 21:27:35  grichenk
* Redesigned CDataSource indexing: seq-id handle -> TSE -> seq/annot
*
* Revision 1.9  2002/02/06 21:46:11  gouriano
* *** empty log message ***
*
* Revision 1.8  2002/02/05 21:46:28  gouriano
* added FindSeqid function, minor tuneup in CSeq_id_mapper
*
* Revision 1.7  2002/01/30 22:09:28  gouriano
* changed CSeqMap interface
*
* Revision 1.6  2002/01/29 17:45:00  grichenk
* Added seq-id handles locking
*
* Revision 1.5  2002/01/28 19:44:49  gouriano
* changed the interface of BioseqHandle: two functions moved from Scope
*
* Revision 1.4  2002/01/23 21:59:31  grichenk
* Redesigned seq-id handles and mapper
*
* Revision 1.3  2002/01/18 15:56:23  gouriano
* changed TSeqMaps definition
*
* Revision 1.2  2002/01/16 16:25:57  gouriano
* restructured objmgr
*
* Revision 1.1  2002/01/11 19:06:18  gouriano
* restructured objmgr
*
*
* ===========================================================================
*/