blast::SDataLoaderConfig
InitializeQueryDataLoaderConfiguration(bool query_is_protein,
CRef<blast::CLocalDbAdapter> db_adapter)
{
SDataLoaderConfig retval(query_is_protein);
retval.OptimizeForWholeLargeSequenceRetrieval();
/* Load the BLAST database into the data loader configuration for the query
* so that if the query sequence(s) are specified as seq-ids, these can be
* fetched from the BLAST database being searched */
if (db_adapter->IsBlastDb() && /* this is a BLAST database search */
retval.m_UseBlastDbs && /* the BLAST database data loader is requested */
(query_is_protein == db_adapter->IsProtein())) { /* the same database type is used for both queries and subjects */
// Make sure we don't add the same database more than once
vector<string> default_dbs;
NStr::Tokenize(retval.m_BlastDbName, " ", default_dbs);
if (default_dbs.size() &&
(find(default_dbs.begin(), default_dbs.end(),
db_adapter->GetDatabaseName()) == default_dbs.end())) {
CNcbiOstrstream oss;
oss << db_adapter->GetDatabaseName() << " " << retval.m_BlastDbName;
retval.m_BlastDbName = CNcbiOstrstreamToString(oss);
}
}
if (retval.m_UseBlastDbs) {
_TRACE("Initializing query data loader to '" << retval.m_BlastDbName
<< "' (" << (query_is_protein ? "protein" : "nucleotide")
<< " BLAST database)");
}
if (retval.m_UseGenbank) {
_TRACE("Initializing query data loader to use GenBank data loader");
}
return retval;
}
void CExportStrategy::x_Process_SearchDb(CRef<CSearchDatabase> & db)
{
if (db.Empty())
{
NCBI_THROW(CBlastException, eInvalidArgument,
"Empty reference for CSearchDatabase.");
}
if (db->GetDatabaseName().empty())
{
NCBI_THROW(CBlastException, eInvalidArgument,
"Error: No database specified");
}
// Set database Name
CRef<CBlast4_subject> subject_p(new CBlast4_subject);
subject_p->SetDatabase(db->GetDatabaseName());
m_QueueSearchRequest->SetSubject(*subject_p);
// Set Entrez Query Limitation
string entrez_query_limit = db->GetEntrezQueryLimitation();
if(!entrez_query_limit.empty())
{
CRef<CBlast4_parameter> p(new CBlast4_parameter);
p->SetName(CBlast4Field::GetName(eBlastOpt_EntrezQuery));
CRef<CBlast4_value> v(new CBlast4_value);
v->SetString().assign(entrez_query_limit);
p->SetValue(*v);
_ASSERT(CBlast4Field::Get(eBlastOpt_EntrezQuery).Match(*p));
m_QueueSearchRequest->SetProgram_options().Set().push_back(p);
}
// Set the GI List Limitation
const CSearchDatabase::TGiList& gi_list_limit = db->GetGiListLimitation();
if (!gi_list_limit.empty())
{
x_AddParameterToProgramOptions(CBlast4Field::Get(eBlastOpt_GiList), gi_list_limit);
}
// Set the negative GI list
const CSearchDatabase::TGiList& neg_gi_list = db->GetNegativeGiListLimitation();
if (!neg_gi_list.empty())
{
x_AddParameterToProgramOptions(CBlast4Field::Get(eBlastOpt_NegativeGiList), neg_gi_list);
}
// Set the filtering algorithms
int algo_id = db->GetFilteringAlgorithm();
if (algo_id != -1)
{
x_AddParameterToProgramOptions(CBlast4Field::Get(eBlastOpt_DbFilteringAlgorithmId), algo_id);
}
}
void CExportStrategy::x_Process_SearchDb(CRef<CSearchDatabase> & db)
{
if (db.Empty())
{
NCBI_THROW(CBlastException, eInvalidArgument,
"Empty reference for CSearchDatabase.");
}
if (db->GetDatabaseName().empty())
{
NCBI_THROW(CBlastException, eInvalidArgument,
"Error: No database specified");
}
// Set database Name
CRef<CBlast4_subject> subject_p(new CBlast4_subject);
subject_p->SetDatabase(db->GetDatabaseName());
m_QueueSearchRequest->SetSubject(*subject_p);
// Set Entrez Query Limitation
string entrez_query_limit = db->GetEntrezQueryLimitation();
if(!entrez_query_limit.empty())
{
CRef<CBlast4_parameter> p(new CBlast4_parameter);
p->SetName(CBlast4Field::GetName(eBlastOpt_EntrezQuery));
CRef<CBlast4_value> v(new CBlast4_value);
v->SetString().assign(entrez_query_limit);
p->SetValue(*v);
_ASSERT(CBlast4Field::Get(eBlastOpt_EntrezQuery).Match(*p));
m_QueueSearchRequest->SetProgram_options().Set().push_back(p);
}
// Set the GI List Limitation
const CSearchDatabase::TGiList& gi_list_limit = db->GetGiListLimitation();
if (!gi_list_limit.empty())
{
#ifdef NCBI_STRICT_GI
vector<int> int_gi_list;
ITERATE ( vector<TGi>, it, gi_list_limit ) {
int_gi_list.push_back(GI_TO( int, *it));
}