rc_t print_headers( const samdump_opts * opts, input_files * ifs )
{
rc_t rc = 0;
switch( opts->header_mode )
{
case hm_dump : rc = print_org_headers( opts, ifs ); break;
case hm_recalc : rc = print_headers_by_recalculating( opts, ifs ); break;
case hm_file : rc = print_headers_from_file( opts ); break;
case hm_none : break; /* to not let the compiler complain about not handled enum */
}
/* attach header comments from the commandline */
if ( rc == 0 && opts->hdr_comments != NULL )
{
uint32_t count;
rc = VNameListCount( opts->hdr_comments, &count );
if ( rc == 0 && count > 0 )
{
uint32_t i;
for ( i = 0; i < count && rc == 0; ++i )
{
const char * s;
rc = VNameListGet( opts->hdr_comments, i, &s );
if ( rc == 0 && s != NULL )
rc = KOutMsg( "@CO\t%s\n", s );
}
}
}
return rc;
}
static rc_t ref_walker_prepare( struct ref_walker * self )
{
rc_t rc = 0;
if ( self == NULL )
rc = RC( rcApp, rcNoTarg, rcConstructing, rcParam, rcNull );
else
{
uint32_t s_count = 0;
rc = VNameListCount ( self->sources, &s_count );
if ( rc == 0 && s_count > 0 )
{
uint32_t r_count = count_ref_regions( &self->regions );
if ( r_count == 0 )
{
uint32_t idx;
for ( idx = 0; idx < s_count && rc == 0; ++idx )
{
const char * name = NULL;
rc = VNameListGet ( self->sources, idx, &name );
if ( rc == 0 && name != NULL )
rc = ref_walker_prepare_1_src( self, name );
}
}
}
check_ref_regions( &self->regions );
self->prepared = ( rc == 0 );
}
return rc;
}
static rc_t print_spotgroups( VNamelist * spotgroups, bool from_stats )
{
uint32_t count;
rc_t rc = VNameListCount( spotgroups, &count );
if ( rc == 0 && count > 0 )
{
uint32_t i;
for ( i = 0; i < count && rc == 0; ++i )
{
const char * s;
rc = VNameListGet( spotgroups, i, &s );
if ( rc == 0 && s != NULL )
{
if ( from_stats )
{
if ( cmp_pchar( s, "default" ) != 0 )
rc = KOutMsg( "@RG\tID:%s\n", s );
}
else
{
rc = KOutMsg( "%s\n", s );
}
}
}
}
return rc;
}
static rc_t SaveToFile( struct KFile * f, const VNamelist * nl, const char * delim )
{
uint32_t count;
rc_t rc = VNameListCount ( nl, &count );
if ( rc == 0 && count > 0 )
{
uint32_t idx;
uint64_t pos = 0;
for ( idx = 0; idx < count && rc == 0; ++idx )
{
const char * s;
rc = VNameListGet ( nl, idx, &s );
if ( rc == 0 && s != NULL )
{
size_t num_writ;
rc = KFileWriteAll ( f, pos, s, string_size ( s ), &num_writ );
if ( rc == 0 )
{
pos += num_writ;
rc = KFileWriteAll ( f, pos, delim, string_size ( delim ), &num_writ );
if ( rc == 0 )
pos += num_writ;
}
}
}
if ( rc == 0 )
rc = KFileSetSize ( f, pos );
}
return rc;
}
LIB_EXPORT rc_t CC VNamelistRemoveAll( VNamelist *self )
{
rc_t rc = 0;
uint32_t count = 1;
while ( count > 0 && rc == 0 )
{
rc = VNameListCount ( self, &count );
if ( rc == 0 && count > 0 )
rc = VNamelistRemoveIdx( self, count - 1 );
}
return rc;
}
LIB_EXPORT rc_t CC VNamelistIndexOf( VNamelist *self, const char* s, uint32_t *found )
{
rc_t rc;
if ( self == NULL )
rc = RC ( rcCont, rcNamelist, rcSearching, rcSelf, rcNull );
else
{
if ( s == NULL )
rc = RC( rcCont, rcNamelist, rcSearching, rcString, rcNull );
else if ( s[0] == 0 )
rc = RC( rcCont, rcNamelist, rcSearching, rcString, rcEmpty );
else if ( found == NULL )
rc = RC( rcCont, rcNamelist, rcSearching, rcParam, rcNull );
else
{
uint32_t count;
rc = VNameListCount ( self, &count );
if ( rc == 0 )
{
uint32_t idx;
size_t n1 = string_size ( s );
for ( idx = 0; idx < count; ++idx )
{
const char * name;
rc = VNameListGet ( self, idx, &name );
if ( rc == 0 )
{
size_t n2 = string_size ( name );
if ( string_cmp ( s, n1, name, n2, (uint32_t) ( ( n1 < n2 ) ? n2 : n1 ) ) == 0 )
{
*found = idx;
return 0;
}
}
}
rc = RC( rcCont, rcNamelist, rcSearching, rcString, rcNotFound );
}
}
}
return rc;
}
static rc_t ref_walker_walk_ref_range( struct ref_walker * self, ref_walker_data * rwd )
{
ReferenceIterator * ref_iter;
rc_t rc = AlignMgrMakeReferenceIterator ( self->amgr, &ref_iter, &self->cb_block, self->min_mapq ); /* align/iterator.h */
if ( rc == 0 )
{
/* construct the reference iterator */
uint32_t idx, count;
uint32_t reflist_options = ref_walker_make_reflist_options( self ); /* above */
Vector cur_id_vector;
VectorInit ( &cur_id_vector, 0, 12 );
rc = VNameListCount ( self->sources, &count );
for ( idx = 0; idx < count && rc == 0; ++idx )
{
const char * src_name = NULL;
rc = VNameListGet ( self->sources, idx, &src_name );
if ( rc == 0 && src_name != NULL )
{
const VDatabase *db;
rc = VDBManagerOpenDBRead ( self->vmgr, &db, self->vschema, "%s", src_name );
if ( rc == 0 )
{
const ReferenceList * ref_list;
rc = ReferenceList_MakeDatabase( &ref_list, db, reflist_options, 0, NULL, 0 );
if ( rc == 0 )
{
const ReferenceObj * ref_obj;
rc = ReferenceList_Find( ref_list, &ref_obj, rwd->ref_name, string_size( rwd->ref_name ) );
if ( rc == 0 )
{
INSDC_coord_len len;
rc = ReferenceObj_SeqLength( ref_obj, &len );
if ( rc == 0 )
{
if ( rwd->ref_start == 0 )
rwd->ref_start = 1;
if ( ( rwd->ref_end == 0 )||( rwd->ref_end > len + 1 ) )
rwd->ref_end = ( len - rwd->ref_start ) + 1;
if ( self->primary_alignments )
rc = ref_walker_add_iterator( self, rwd->ref_name, rwd->ref_start, rwd->ref_end, src_name,
&cur_id_vector, db, ref_obj, ref_iter, TBL_PRIM, primary_align_ids );
if ( rc == 0 && self->secondary_alignments )
rc = ref_walker_add_iterator( self, rwd->ref_name, rwd->ref_start, rwd->ref_end, src_name,
&cur_id_vector, db, ref_obj, ref_iter, TBL_SEC, secondary_align_ids );
if ( rc == 0 && self->evidence_alignments )
rc = ref_walker_add_iterator( self, rwd->ref_name, rwd->ref_start, rwd->ref_end, src_name,
&cur_id_vector, db, ref_obj, ref_iter, TBL_EV, evidence_align_ids );
}
ReferenceObj_Release( ref_obj );
}
ReferenceList_Release( ref_list );
}
VDatabaseRelease( db );
}
}
}
if ( rc == 0 )
{
/* walk the reference iterator */
struct ReferenceObj const * ref_obj;
rc = ReferenceIteratorNextReference( ref_iter, NULL, NULL, &ref_obj );
if ( rc == 0 && ref_obj != NULL )
{
if ( self->use_seq_name )
rc = ReferenceObj_Name( ref_obj, &rwd->ref_name );
else
rc = ReferenceObj_SeqId( ref_obj, &rwd->ref_name );
if ( rc == 0 )
{
INSDC_coord_zero first_pos;
INSDC_coord_len len;
rc_t rc_w = ReferenceIteratorNextWindow ( ref_iter, &first_pos, &len );
while ( rc == 0 && rc_w == 0 )
{
rc_t rc_p = ReferenceIteratorNextPos ( ref_iter, !self->no_skip );
if ( rc_p == 0 )
{
rc = ReferenceIteratorPosition ( ref_iter, &rwd->pos, &rwd->depth, &rwd->bin_ref_base );
if ( rwd->depth > 0 && rc == 0 )
{
rc_t rc_sg = 0;
rwd->ascii_ref_base = _4na_to_ascii( rwd->bin_ref_base, false );
if ( self->on_enter_ref_pos != NULL )
rc = self->on_enter_ref_pos( rwd );
while ( rc_sg == 0 && rc == 0 )
{
rc_sg = ReferenceIteratorNextSpotGroup ( ref_iter, &rwd->spot_group, &rwd->spot_group_len );
if ( rc_sg == 0 )
{
rc_t rc_pr = 0;
if ( self->on_enter_spot_group != NULL )
rc = self->on_enter_spot_group( rwd );
while ( rc == 0 && rc_pr == 0 )
{
const PlacementRecord * rec;
rc_pr = ReferenceIteratorNextPlacement ( ref_iter, &rec );
if ( rc_pr == 0 && self->on_alignment != NULL )
rc = ref_walker_walk_alignment( self, ref_iter, rec, rwd );
}
if ( self->on_exit_spot_group != NULL )
rc = self->on_exit_spot_group( rwd );
}
}
if ( self->on_exit_ref_pos != NULL )
rc = self->on_exit_ref_pos( rwd );
}
rc = Quitting();
}
}
}
}
}
/* free cur_id_vector */
ReferenceIteratorRelease ( ref_iter );
}
return rc;
}
static rc_t print_headers_from_file( const samdump_opts * opts )
{
KDirectory * dir;
rc_t rc = KDirectoryNativeDir ( &dir );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant created native directory for file '$(t)'", "t=%s", opts->header_file ) );
}
else
{
const struct KFile * f;
rc = KDirectoryOpenFileRead ( dir, &f, "%s", opts->header_file );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant open file '$(t)'", "t=%s", opts->header_file ) );
}
else
{
VNamelist * headers;
rc = VNamelistMake ( &headers, 25 );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant create container for file '$(t)'", "t=%s", opts->header_file ) );
}
else
{
rc = LoadKFileToNameList( f, headers );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant load file '$(t)' into container", "t=%s", opts->header_file ) );
}
else
{
uint32_t count;
rc = VNameListCount ( headers, &count );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant get count for container of '$(t)'", "t=%s", opts->header_file ) );
}
else
{
uint32_t i;
for ( i = 0; i < count && rc == 0; ++i )
{
const char * line;
rc = VNameListGet ( headers, i, &line );
if ( rc != 0 )
{
(void)PLOGERR( klogErr, ( klogErr, rc, "cant get line #$(t) from container", "t=%u", i ) );
}
else
{
rc = KOutMsg( "%s\n", line );
}
}
}
}
VNamelistRelease ( headers );
}
KFileRelease ( f );
}
KDirectoryRelease ( dir );
}
return rc;
}
rc_t perform_cgi_request( struct cgi_request * request,
struct KStream ** reply_stream,
uint32_t * rslt_code )
{
rc_t rc;
if ( request == NULL || reply_stream == NULL || rslt_code == NULL )
rc = RC( rcVDB, rcNoTarg, rcConstructing, rcParam, rcNull );
else
{
ver_t http_vers = 0x01010000;
KClientHttpRequest * kns_req;
*reply_stream = NULL;
rc = KNSManagerMakeClientRequest( request->kns_mgr, &kns_req, http_vers, NULL, "%s", request->url );
if ( rc != 0 )
ErrMsg( "KNSManagerMakeClientRequest( '%s' ) -> %R", request->url, rc );
else
{
uint32_t count;
rc = VNameListCount( request->params, &count );
if ( rc != 0 )
ErrMsg( "VNameListCount() -> %R", rc );
else
{
uint32_t idx;
for ( idx = 0; idx < count && rc == 0; ++idx )
{
const char * s;
rc = VNameListGet( request->params, idx, &s );
if ( rc != 0 )
ErrMsg( "VNameListGet( #%d ) -> %R", idx, rc );
else
{
rc = KClientHttpRequestAddPostParam( kns_req, "%s", s );
if ( rc != 0 )
ErrMsg( "KClientHttpRequestAddPostParam( '%s' ) -> %R", s, rc );
}
}
if ( rc == 0 )
{
KClientHttpResult * kns_rslt;
rc = KClientHttpRequestPOST( kns_req, &kns_rslt );
if ( rc != 0 )
ErrMsg( "KClientHttpRequestPOST() -> %R", rc );
else
{
rc = KClientHttpResultStatus( kns_rslt, rslt_code, NULL, 0, NULL );
if ( rc != 0 )
ErrMsg( "KClientHttpResultStatus() -> %R", rc );
else
{
rc = KClientHttpResultGetInputStream( kns_rslt, reply_stream );
if ( rc != 0 )
ErrMsg( "KClientHttpResultGetInputStream() -> %R", rc );
}
KClientHttpResultRelease( kns_rslt );
}
}
}
KClientHttpRequestRelease( kns_req );
}
}
return rc;
}
void report_options( samdump_opts * opts )
{
rc_t rc;
uint32_t len;
KOutMsg( "dump unaligned reads : %s\n", opts->dump_unaligned_reads ? "YES" : "NO" );
KOutMsg( "dump unaligned only : %s\n", opts->dump_unaligned_only ? "YES" : "NO" );
KOutMsg( "dump prim. alignments : %s\n", opts->dump_primary_alignments ? "YES" : "NO" );
KOutMsg( "dump sec. alignments : %s\n", opts->dump_secondary_alignments ? "YES" : "NO" );
KOutMsg( "dump cga-tools-mode : %s\n", opts->dump_cga_tools_mode ? "YES" : "NO" );
KOutMsg( "dump cg-evidence : %s\n", opts->dump_cg_evidence ? "YES" : "NO" );
KOutMsg( "dump cg-ev-dnb : %s\n", opts->dump_cg_ev_dnb ? "YES" : "NO" );
KOutMsg( "dump cg-sam : %s\n", opts->dump_cg_sam ? "YES" : "NO" );
KOutMsg( "merge cg-cigar : %s\n", opts->merge_cg_cigar ? "YES" : "NO" );
KOutMsg( "use long cigar : %s\n", opts->use_long_cigar ? "YES" : "NO" );
KOutMsg( "print seqid : %s\n", opts->use_seqid_as_refname ? "YES" : "NO" );
KOutMsg( "print READ with '=' : %s\n", opts->print_matches_as_equal_sign ? "YES" : "NO" );
KOutMsg( "reverse unaligned rd : %s\n", opts->reverse_unaligned_reads ? "YES" : "NO" );
KOutMsg( "add spotgrp to qname : %s\n", opts->print_spot_group_in_name ? "YES" : "NO" );
KOutMsg( "print id in col. XI : %s\n", opts->print_alignment_id_in_column_xi ? "YES" : "NO" );
KOutMsg( "report matecache : %s\n", opts->report_cache ? "YES" : "NO" );
KOutMsg( "print cg-names : %s\n", opts->print_cg_names ? "YES" : "NO" );
switch( opts->header_mode )
{
case hm_none : KOutMsg( "header-mode : dont print\n" ); break;
case hm_recalc : KOutMsg( "header-mode : recalculate\n" ); break;
case hm_dump : KOutMsg( "header-mode : print meta-data\n" ); break;
default : KOutMsg( "header-mode : unknown\n" ); break;
}
switch( opts->cigar_treatment )
{
case ct_unchanged : KOutMsg( "cigar-treatment : unchanged\n" ); break;
case ct_cg_style : KOutMsg( "cigar-treatment : transform cg to 'cg-style\n" ); break;
case ct_cg_merge : KOutMsg( "cigar-treatment : transform cg to valid SAM\n" ); break;
default : KOutMsg( "cigar-treatment : unknow'\n" ); break;
}
switch( opts->output_compression )
{
case oc_none : KOutMsg( "output-compression : none\n" ); break;
case oc_gzip : KOutMsg( "output-compression : gzip\n" ); break;
case oc_bzip2 : KOutMsg( "output-compression : bzip2\n" ); break;
default : KOutMsg( "output-compression : unknown\n" ); break;
}
switch( opts->output_format )
{
case of_sam : KOutMsg( "output-format : SAM\n" ); break;
case of_fasta : KOutMsg( "output-format : FASTA\n" ); break;
case of_fastq : KOutMsg( "output-format : FASTQ\n" ); break;
default : KOutMsg( "output-format : unknown\n" ); break;
}
switch( opts->dump_mode )
{
case dm_one_ref_at_a_time : KOutMsg( "dump-mode : one ref at a time\n" ); break;
case dm_prepare_all_refs : KOutMsg( "dump-mode : prepare all refs\n" ); break;
default : KOutMsg( "dump-mode : unknown\n" ); break;
}
KOutMsg( "number of regions : %u\n", opts->region_count );
if ( opts->region_count > 0 )
foreach_reference( &opts->regions, report_reference_cb, NULL );
if ( opts->qname_prefix == NULL )
KOutMsg( "qname-prefix : NONE\n" );
else
KOutMsg( "qname-prefix : '%s'\n", opts->qname_prefix );
if ( opts->qual_quant == NULL )
KOutMsg( "quality-quantization : NONE\n" );
else
KOutMsg( "quality-quantization : '%s'\n", opts->qual_quant );
if ( opts->hdr_comments != NULL )
{
rc = VNameListCount( opts->hdr_comments, &len );
if ( rc == 0 && len > 0 )
{
uint32_t i;
KOutMsg( "header-comments-count : %u\n", len );
for ( i = 0; i < len; ++i )
{
const char * s;
rc = VNameListGet( opts->hdr_comments, i, &s );
if ( rc == 0 && s != NULL )
KOutMsg( "header-comment[ %u ] : '%s'\n", i, s );
}
}
}
if ( opts->input_files != NULL )
{
rc = VNameListCount( opts->input_files, &len );
if ( rc == 0 && len > 0 )
{
uint32_t i;
KOutMsg( "input-file-count : %u\n", len );
for ( i = 0; i < len; ++i )
{
const char * s;
rc = VNameListGet( opts->input_files, i, &s );
if ( rc == 0 && s != NULL )
KOutMsg( "input-file[ %u ] : '%s'\n", i, s );
}
}
}
len = VectorLength( &opts->mp_dist );
if ( len > 0 )
{
uint32_t i;
KOutMsg( "matepair-dist-filters : %u\n", len );
for ( i = 0; i < len; ++i )
{
range *r = VectorGet( &opts->mp_dist, i );
if ( r != NULL )
KOutMsg( "matepair-dist[ %u ] : '%u ... %u'\n", i, r->start, r->end );
}
}
KOutMsg( "mate-gap-cache-limit : %u\n", opts->mape_gap_cache_limit );
KOutMsg( "test-row limit : %u\n", opts->test_rows );
KOutMsg( "outputfile : %s\n", opts->outputfile );
KOutMsg( "outputbuffer-size : %u\n", opts->output_buffer_size );
KOutMsg( "cursor-cache-size : %u\n", opts->cursor_cache_size );
KOutMsg( "use min-mapq : %s\n", opts->use_min_mapq ? "YES" : "NO" );
KOutMsg( "min-mapq : %i\n", opts->min_mapq );
}
