static int select_visitor_region_node(GtNodeVisitor *nv, GtRegionNode *rn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *select_visitor;
gt_error_check(err);
select_visitor = select_visitor_cast(nv);
if (!gt_str_length(select_visitor->seqid) || /* no seqid was specified */
!gt_str_cmp(select_visitor->seqid, /* or seqids are equal */
gt_genome_node_get_seqid((GtGenomeNode*) rn))) {
if (select_visitor->contain_range.start != GT_UNDEF_ULONG) {
GtRange range = gt_genome_node_get_range((GtGenomeNode*) rn);
if (gt_range_overlap(&range, &select_visitor->contain_range)) {
/* an overlapping contain range was defined -> update range */
range.start = MAX(range.start, select_visitor->contain_range.start);
range.end = MIN(range.end, select_visitor->contain_range.end);
gt_genome_node_set_range((GtGenomeNode*) rn, &range);
gt_queue_add(select_visitor->node_buffer, rn);
}
else /* contain range does not overlap with <rn> range -> delete <rn> */
gt_genome_node_delete((GtGenomeNode*) rn);
}
else
gt_queue_add(select_visitor->node_buffer, rn);
}
else
gt_genome_node_delete((GtGenomeNode*) rn);
return 0;
}
void gt_desc_buffer_reset(GtDescBuffer *db)
{
GtUword laststartpos;
gt_assert(db);
if (!db->dirty) return;
if (gt_queue_size(db->startqueue) == 0) {
db->length = 0;
db->dirty = false;
return;
}
laststartpos = (GtUword) gt_queue_head(db->startqueue);
if (laststartpos != 0) {
laststartpos = (GtUword) gt_queue_get(db->startqueue);
db->length = db->length - laststartpos;
if (db->length >= laststartpos) {
/* strings overlap */
memmove(db->buf, db->buf + laststartpos, db->length * sizeof (char));
} else {
/* no overlap */
memcpy(db->buf, db->buf + laststartpos, db->length * sizeof (char));
}
gt_queue_add(db->startqueue, (void*) 0);
}
db->dirty = false;
}
static void infer_cds_visitor_test_data(GtQueue *queue)
{
GtError *error = gt_error_new();
const char *file = "data/gff3/grape-codons.gff3";
GtNodeStream *gff3in = gt_gff3_in_stream_new_unsorted(1, &file);
gt_gff3_in_stream_check_id_attributes((GtGFF3InStream *)gff3in);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream *)gff3in);
GtLogger *logger = gt_logger_new(true, "", stderr);
GtNodeStream *icv_stream = agn_infer_cds_stream_new(gff3in, NULL, logger);
GtArray *feats = gt_array_new( sizeof(GtFeatureNode *) );
GtNodeStream *arraystream = gt_array_out_stream_new(icv_stream, feats, error);
int pullresult = gt_node_stream_pull(arraystream, error);
if(pullresult == -1)
{
fprintf(stderr, "[AgnInferCDSVisitor::infer_cds_visitor_test_data] error "
"processing features: %s\n", gt_error_get(error));
}
gt_node_stream_delete(gff3in);
gt_node_stream_delete(icv_stream);
gt_node_stream_delete(arraystream);
gt_logger_delete(logger);
gt_array_sort(feats, (GtCompare)agn_genome_node_compare);
gt_array_reverse(feats);
while(gt_array_size(feats) > 0)
{
GtFeatureNode *fn = *(GtFeatureNode **)gt_array_pop(feats);
gt_queue_add(queue, fn);
}
gt_array_delete(feats);
gt_error_delete(error);
}
static int select_visitor_eof_node(GtNodeVisitor *nv, GtEOFNode *eofn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *select_visitor;
gt_error_check(err);
select_visitor = select_visitor_cast(nv);
gt_queue_add(select_visitor->node_buffer, eofn);
return 0;
}
static int add_ids_visitor_comment_node(GtNodeVisitor *nv, GtCommentNode *c,
GT_UNUSED GtError *err)
{
GtAddIDsVisitor *add_ids_visitor;
gt_error_check(err);
add_ids_visitor = add_ids_visitor_cast(nv);
gt_queue_add(add_ids_visitor->node_buffer, c);
return 0;
}
static int select_visitor_comment_node(GtNodeVisitor *nv, GtCommentNode *c,
GT_UNUSED GtError *err)
{
GtSelectVisitor *select_visitor;
gt_error_check(err);
select_visitor = select_visitor_cast(nv);
gt_queue_add(select_visitor->node_buffer, c);
return 0;
}
static int add_ids_visitor_meta_node(GtNodeVisitor *nv, GtMetaNode *mn,
GT_UNUSED GtError *err)
{
GtAddIDsVisitor *add_ids_visitor;
gt_error_check(err);
add_ids_visitor = add_ids_visitor_cast(nv);
gt_queue_add(add_ids_visitor->node_buffer, mn);
return 0;
}
static int add_ids_visitor_eof_node(GtNodeVisitor *nv, GtEOFNode *eofn,
GT_UNUSED GtError *err)
{
GtAddIDsVisitor *add_ids_visitor;
gt_error_check(err);
add_ids_visitor = add_ids_visitor_cast(nv);
gt_add_ids_visitor_finalize(nv);
gt_queue_add(add_ids_visitor->node_buffer, eofn);
return 0;
}
static int filter_stream_next(GtNodeStream *ns, GtGenomeNode **gn,
GtError *error)
{
AgnFilterStream *stream;
GtFeatureNode *fn;
int had_err;
gt_error_check(error);
stream = filter_stream_cast(ns);
if(gt_queue_size(stream->cache) > 0)
{
*gn = gt_queue_get(stream->cache);
return 0;
}
while(1)
{
had_err = gt_node_stream_next(stream->in_stream, gn, error);
if(had_err)
return had_err;
if(!*gn)
return 0;
fn = gt_feature_node_try_cast(*gn);
if(!fn)
return 0;
GtFeatureNode *current;
GtFeatureNodeIterator *iter = gt_feature_node_iterator_new(fn);
for(current = gt_feature_node_iterator_next(iter);
current != NULL;
current = gt_feature_node_iterator_next(iter))
{
const char *type = gt_feature_node_get_type(current);
bool keepfeature = false;
if(gt_hashmap_get(stream->typestokeep, type) != NULL)
keepfeature = true;
if(keepfeature)
{
gt_genome_node_ref((GtGenomeNode *)current);
gt_queue_add(stream->cache, current);
}
}
gt_feature_node_iterator_delete(iter);
gt_genome_node_delete((GtGenomeNode *)fn);
if(gt_queue_size(stream->cache) > 0)
{
*gn = gt_queue_get(stream->cache);
return 0;
}
}
return 0;
}
static int add_auto_sr_to_queue(GT_UNUSED void *key, void *value, void *data,
GT_UNUSED GtError *err)
{
AutomaticSequenceRegion *auto_sr = value;
GtQueue *genome_nodes = data;
GtGenomeNode *gf;
unsigned int i;
gt_error_check(err);
gt_assert(key && value && data);
if (gt_array_size(auto_sr->feature_nodes)) {
gt_queue_add(genome_nodes, auto_sr->sequence_region);
auto_sr->sequence_region = NULL;
for (i = 0; i < gt_array_size(auto_sr->feature_nodes); i++) {
gf = *(GtGenomeNode**) gt_array_get(auto_sr->feature_nodes, i);
gt_queue_add(genome_nodes, gf);
}
gt_array_reset(auto_sr->feature_nodes);
}
return 0;
}
static int add_ids_visitor_sequence_node(GtNodeVisitor *nv, GtSequenceNode *sn,
GT_UNUSED GtError *err)
{
GtAddIDsVisitor *add_ids_visitor;
gt_error_check(err);
add_ids_visitor = add_ids_visitor_cast(nv);
/* sequence nodes have to be at the end of a stream -> finalize first */
gt_add_ids_visitor_finalize(nv);
/* then add sequence node to buffer */
gt_queue_add(add_ids_visitor->node_buffer, sn);
return 0;
}
static int select_visitor_sequence_node(GtNodeVisitor *nv, GtSequenceNode *sn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *select_visitor;
gt_error_check(err);
select_visitor = select_visitor_cast(nv);
if (!gt_str_length(select_visitor->seqid) || /* no seqid was specified */
!gt_str_cmp(select_visitor->seqid, /* or seqids are equal */
gt_genome_node_get_seqid((GtGenomeNode*) sn))) {
gt_queue_add(select_visitor->node_buffer, sn);
}
else
gt_genome_node_delete((GtGenomeNode*) sn);
return 0;
}
GtDescBuffer* gt_desc_buffer_new(void)
{
GtDescBuffer *db = gt_malloc(sizeof *db);
db->buf = gt_calloc(GT_DESC_BUFFER_INIT_SIZE, sizeof (char));
db->length = 0;
db->maxlength = db->curlength = 0;
db->allocated = GT_DESC_BUFFER_INIT_SIZE;
db->finished = false;
db->dirty = true;
db->shorten = false;
db->seen_whitespace = false;
db->reference_count = 0;
db->startqueue = gt_queue_new();
gt_queue_add(db->startqueue, (void*) 0);
return db;
}
static int buffer_stream_next(GtNodeStream *ns, GtGenomeNode **gn, GtError *err)
{
GtBufferStream *bs;
gt_error_check(err);
bs = buffer_stream_cast(ns);
if (bs->buffering) {
int had_err = gt_node_stream_next(bs->in_stream, gn, err);
if (!had_err && *gn)
gt_queue_add(bs->node_buffer, gt_genome_node_ref(*gn));
return had_err;
}
else {
*gn = gt_queue_size(bs->node_buffer) ? gt_queue_get(bs->node_buffer) : NULL;
return 0;
}
}
static int snp_annotator_stream_process_current_gene(GtSNPAnnotatorStream *sas,
GtError *err)
{
int had_err = 0;
GtUword i;
GtUword nof_genes = gt_array_size(sas->cur_gene_set);
gt_error_check(err);
if (gt_queue_size(sas->snps) > 0) {
/* we need to process SNPs for a gene cluster*/
gt_assert(gt_queue_size(sas->outqueue) == 0);
for (i = 0; !had_err && i < nof_genes; i++) {
GtNodeVisitor *sav;
GtFeatureNode *gene;
gene = *(GtFeatureNode**) gt_array_get(sas->cur_gene_set, i);
sav = gt_snp_annotator_visitor_new(gene, sas->tt, sas->rmap, err);
if (!sav)
had_err = -1;
if (!had_err) {
if (i < nof_genes-1) {
had_err = gt_queue_iterate(sas->snps,
snp_annotator_stream_process_snp,
sav, err);
} else {
while (!had_err && gt_queue_size(sas->snps) > 0) {
GtFeatureNode *snp = (GtFeatureNode*) gt_queue_get(sas->snps);
had_err = gt_genome_node_accept((GtGenomeNode*) snp, sav, err);
gt_queue_add(sas->outqueue, snp);
gt_genome_node_delete((GtGenomeNode*) snp);
}
}
gt_node_visitor_delete(sav);
}
gt_genome_node_delete((GtGenomeNode*) gene);
}
} else {
/* no SNPs for this gene cluster, delete it */
for (i = 0; !had_err && i < nof_genes; i++) {
gt_genome_node_delete(*(GtGenomeNode**) gt_array_get(sas->cur_gene_set,
i));
}
}
gt_assert(gt_queue_size(sas->snps) == 0);
gt_array_reset(sas->cur_gene_set);
return had_err;
}
int gt_bed_parser_parse(GtBEDParser *bed_parser, GtQueue *genome_nodes,
const char *filename, GtError *err)
{
GtIO *bed_file;
int had_err;
gt_error_check(err);
gt_assert(bed_parser && genome_nodes);
bed_file = gt_io_new(filename, "r");
/* parse BED file */
had_err = parse_bed_file(bed_parser, bed_file, err);
/* process created region and feature nodes */
gt_region_node_builder_build(bed_parser->region_node_builder, genome_nodes);
gt_region_node_builder_reset(bed_parser->region_node_builder);
while (gt_queue_size(bed_parser->feature_nodes))
gt_queue_add(genome_nodes, gt_queue_get(bed_parser->feature_nodes));
gt_io_delete(bed_file);
return had_err;
}
void gt_orphanage_add(GtOrphanage *o, GtGenomeNode *orphan,
const char *orphan_id, GtStrArray *missing_parents)
{
const char *missing_parent;
GtUword i;
gt_assert(o && orphan);
gt_assert(gt_feature_node_get_attribute((GtFeatureNode*) orphan,
GT_GFF_PARENT));
gt_queue_add(o->orphans, orphan);
if (orphan_id && !gt_cstr_table_get(o->orphan_ids, orphan_id))
gt_cstr_table_add(o->orphan_ids, orphan_id);
if (missing_parents) {
for (i = 0; i < gt_str_array_size(missing_parents); i++) {
missing_parent = gt_str_array_get(missing_parents, i);
if (!gt_cstr_table_get(o->missing_parents, missing_parent))
gt_cstr_table_add(o->missing_parents, missing_parent);
}
}
}
void gt_desc_buffer_append_char(GtDescBuffer *db, char c)
{
gt_assert(db);
if (db->shorten) {
if (db->seen_whitespace)
return;
if (isspace(c)) {
db->seen_whitespace = true;
return;
}
}
if (db->finished) {
gt_queue_add(db->startqueue, (void*) (db->length));
db->finished = false;
}
if (db->length + 2 > db->allocated) {
db->buf = gt_dynalloc(db->buf, &db->allocated,
(db->length + 2) * sizeof (char));
}
db->curlength++;
db->buf[db->length++] = c;
}
void feature_in_stream_init(GtFeatureInStream *stream)
{
GtUword i;
GtError *error = gt_error_new();
stream->seqids = gt_feature_index_get_seqids(stream->fi, error);
stream->seqindex = 0;
for (i = 0; i < gt_str_array_size(stream->seqids); i++)
{
const char *seqid = gt_str_array_get(stream->seqids, i);
GtRange seqrange;
if (stream->useorig)
gt_feature_index_get_orig_range_for_seqid(stream->fi, &seqrange, seqid,
error);
else
gt_feature_index_get_range_for_seqid(stream->fi, &seqrange, seqid, error);
GtStr *seqstr = gt_str_new_cstr(seqid);
GtGenomeNode *rn = gt_region_node_new(seqstr, seqrange.start, seqrange.end);
gt_queue_add(stream->regioncache, rn);
gt_str_delete(seqstr);
}
gt_error_delete(error);
}
static int gff3_numsorted_out_stream_next(GtNodeStream *ns, GtGenomeNode **gn,
GtError *err)
{
GtGFF3NumsortedOutStream *gff3_out_stream;
int had_err = 0;
GtUword i = 0;
gt_error_check(err);
gff3_out_stream = gff3_numsorted_out_stream_cast(ns);
if (!gff3_out_stream->outqueue) {
gff3_out_stream->outqueue = gt_queue_new();
while (!(had_err =
gt_node_stream_next(gff3_out_stream->in_stream, gn, err))) {
if (!*gn) break;
gt_array_add(gff3_out_stream->buffer, *gn);
}
if (!had_err) {
gt_genome_nodes_sort_stable_with_func(gff3_out_stream->buffer,
(GtCompare) gt_genome_node_compare_numeric_seqids);
for (i = 0; !had_err && i < gt_array_size(gff3_out_stream->buffer); i++) {
GtGenomeNode *mygn = *(GtGenomeNode**)
gt_array_get(gff3_out_stream->buffer, i);
gt_queue_add(gff3_out_stream->outqueue, mygn);
}
}
}
if (gff3_out_stream->outqueue && !had_err) {
if (gt_queue_size(gff3_out_stream->outqueue) > 0) {
GtGenomeNode *mygn = (GtGenomeNode*)
gt_queue_get(gff3_out_stream->outqueue);
gt_assert(mygn);
had_err = gt_genome_node_accept(mygn, gff3_out_stream->gff3_visitor, err);
if (!had_err)
*gn = mygn;
}
}
return had_err;
}
static int add_ids_visitor_region_node(GtNodeVisitor *nv, GtRegionNode *rn,
GT_UNUSED GtError *err)
{
GtAddIDsVisitor *aiv;
const char *seqid;
int had_err = 0;
gt_error_check(err);
aiv = add_ids_visitor_cast(nv);
seqid = gt_str_get(gt_genome_node_get_seqid((GtGenomeNode*) rn));
if (gt_hashmap_get(aiv->undefined_sequence_regions, seqid)) {
gt_error_set(err, "genome feature with id \"%s\" has been defined before "
"the corresponding \"%s\" definition on line %u in file "
"\"%s\"", seqid, GT_GFF_SEQUENCE_REGION,
gt_genome_node_get_line_number((GtGenomeNode*) rn),
gt_genome_node_get_filename((GtGenomeNode*) rn));
had_err = -1;
}
if (!had_err) {
if (!gt_cstr_table_get(aiv->defined_seqids, seqid))
gt_cstr_table_add(aiv->defined_seqids, seqid);
gt_queue_add(aiv->node_buffer, rn);
}
return had_err;
}
int gt_gtf_parser_parse(GtGTFParser *parser, GtQueue *genome_nodes,
GtStr *filenamestr, GtFile *fpin, bool be_tolerant,
GtError *err)
{
GtStr *seqid_str, *source_str, *line_buffer;
char *line;
size_t line_length;
GtUword i, line_number = 0;
GtGenomeNode *gn;
GtRange range;
GtPhase phase_value;
GtStrand gt_strand_value;
GtSplitter *splitter, *attribute_splitter;
float score_value;
char *seqname,
*source,
*feature,
*start,
*end,
*score,
*strand,
*frame,
*attributes,
*token,
*gene_id,
*gene_name = NULL,
*transcript_id,
*transcript_name = NULL,
**tokens;
GtHashmap *transcript_id_hash; /* map from transcript id to array of genome
nodes */
GtArray *gt_genome_node_array;
ConstructionInfo cinfo;
GTF_feature_type gtf_feature_type;
GT_UNUSED bool gff_type_is_valid = false;
const char *type = NULL;
const char *filename;
bool score_is_defined;
int had_err = 0;
gt_assert(parser && genome_nodes);
gt_error_check(err);
filename = gt_str_get(filenamestr);
/* alloc */
line_buffer = gt_str_new();
splitter = gt_splitter_new(),
attribute_splitter = gt_splitter_new();
#define HANDLE_ERROR \
if (had_err) { \
if (be_tolerant) { \
fprintf(stderr, "skipping line: %s\n", gt_error_get(err)); \
gt_error_unset(err); \
gt_str_reset(line_buffer); \
had_err = 0; \
continue; \
} \
else { \
had_err = -1; \
break; \
} \
}
while (gt_str_read_next_line_generic(line_buffer, fpin) != EOF) {
line = gt_str_get(line_buffer);
line_length = gt_str_length(line_buffer);
line_number++;
had_err = 0;
if (line_length == 0) {
gt_warning("skipping blank line " GT_WU " in file \"%s\"", line_number,
filename);
}
else if (line[0] == '#') {
/* storing comment */
if (line_length >= 2 && line[1] == '#')
gn = gt_comment_node_new(line+2); /* store '##' line as '#' line */
else
gn = gt_comment_node_new(line+1);
gt_genome_node_set_origin(gn, filenamestr, line_number);
gt_queue_add(genome_nodes, gn);
}
else {
/* process tab delimited GTF line */
gt_splitter_reset(splitter);
gt_splitter_split(splitter, line, line_length, '\t');
if (gt_splitter_size(splitter) != 9UL) {
gt_error_set(err, "line " GT_WU " in file \"%s\" contains " GT_WU
" tab (\\t) " "separated fields instead of 9", line_number,
filename,
gt_splitter_size(splitter));
had_err = -1;
break;
}
tokens = gt_splitter_get_tokens(splitter);
seqname = tokens[0];
source = tokens[1];
feature = tokens[2];
start = tokens[3];
end = tokens[4];
score = tokens[5];
strand = tokens[6];
frame = tokens[7];
attributes = tokens[8];
/* parse feature */
if (GTF_feature_type_get(>f_feature_type, feature) == -1) {
/* we skip unknown features */
fprintf(stderr, "skipping line " GT_WU " in file \"%s\": unknown "
"feature: \"%s\"\n", line_number, filename, feature);
gt_str_reset(line_buffer);
continue;
}
/* translate into GFF3 feature type */
switch (gtf_feature_type) {
case GTF_CDS:
case GTF_stop_codon:
gff_type_is_valid = gt_type_checker_is_valid(parser->type_checker,
gt_ft_CDS);
type = gt_ft_CDS;
break;
case GTF_exon:
gff_type_is_valid = gt_type_checker_is_valid(parser->type_checker,
gt_ft_exon);
type = gt_ft_exon;
}
gt_assert(gff_type_is_valid);
/* parse the range */
had_err = gt_parse_range(&range, start, end, line_number, filename, err);
HANDLE_ERROR;
/* process seqname (we have to do it here because we need the range) */
gt_region_node_builder_add_region(parser->region_node_builder, seqname,
range);
/* parse the score */
had_err = gt_parse_score(&score_is_defined, &score_value, score,
line_number, filename, err);
HANDLE_ERROR;
/* parse the strand */
had_err = gt_parse_strand(>_strand_value, strand, line_number, filename,
err);
HANDLE_ERROR;
/* parse the frame */
had_err = gt_parse_phase(&phase_value, frame, line_number, filename, err);
HANDLE_ERROR;
/* parse the attributes */
gt_splitter_reset(attribute_splitter);
gene_id = NULL;
transcript_id = NULL;
gt_splitter_split(attribute_splitter, attributes, strlen(attributes),
';');
for (i = 0; i < gt_splitter_size(attribute_splitter); i++) {
token = gt_splitter_get_token(attribute_splitter, i);
/* skip leading blanks */
while (*token == ' ')
token++;
/* look for the two mandatory attributes */
if (strncmp(token, GENE_ID_ATTRIBUTE, strlen(GENE_ID_ATTRIBUTE)) == 0) {
if (strlen(token) + 2 < strlen(GENE_ID_ATTRIBUTE)) {
gt_error_set(err, "missing value to attribute \"%s\" on line "
GT_WU "in file \"%s\"", GENE_ID_ATTRIBUTE, line_number,
filename);
had_err = -1;
}
HANDLE_ERROR;
gene_id = token + strlen(GENE_ID_ATTRIBUTE) + 1;
}
else if (strncmp(token, TRANSCRIPT_ID_ATTRIBUTE,
strlen(TRANSCRIPT_ID_ATTRIBUTE)) == 0) {
if (strlen(token) + 2 < strlen(TRANSCRIPT_ID_ATTRIBUTE)) {
gt_error_set(err, "missing value to attribute \"%s\" on line "
GT_WU "in file \"%s\"", TRANSCRIPT_ID_ATTRIBUTE,
line_number, filename);
had_err = -1;
}
HANDLE_ERROR;
transcript_id = token + strlen(TRANSCRIPT_ID_ATTRIBUTE) + 1;
}
else if (strncmp(token, GENE_NAME_ATTRIBUTE,
strlen(GENE_NAME_ATTRIBUTE)) == 0) {
if (strlen(token) + 2 < strlen(GENE_NAME_ATTRIBUTE)) {
gt_error_set(err, "missing value to attribute \"%s\" on line "
GT_WU "in file \"%s\"", GENE_NAME_ATTRIBUTE,
line_number, filename);
had_err = -1;
}
HANDLE_ERROR;
gene_name = token + strlen(GENE_NAME_ATTRIBUTE) + 1;
/* for output we want to strip quotes */
if (*gene_name == '"')
gene_name++;
if (gene_name[strlen(gene_name)-1] == '"')
gene_name[strlen(gene_name)-1] = '\0';
}
else if (strncmp(token, TRANSCRIPT_NAME_ATTRIBUTE,
strlen(TRANSCRIPT_NAME_ATTRIBUTE)) == 0) {
if (strlen(token) + 2 < strlen(TRANSCRIPT_NAME_ATTRIBUTE)) {
gt_error_set(err, "missing value to attribute \"%s\" on line "
GT_WU "in file \"%s\"", TRANSCRIPT_NAME_ATTRIBUTE,
line_number, filename);
had_err = -1;
}
HANDLE_ERROR;
transcript_name = token + strlen(TRANSCRIPT_NAME_ATTRIBUTE) + 1;
/* for output we want to strip quotes */
if (*transcript_name == '"')
transcript_name++;
if (transcript_name[strlen(transcript_name)-1] == '"')
transcript_name[strlen(transcript_name)-1] = '\0';
}
}
/* check for the mandatory attributes */
if (!gene_id) {
gt_error_set(err, "missing attribute \"%s\" on line " GT_WU
" in file \"%s\"", GENE_ID_ATTRIBUTE, line_number,
filename);
had_err = -1;
}
HANDLE_ERROR;
if (!transcript_id) {
gt_error_set(err, "missing attribute \"%s\" on line " GT_WU
" in file \"%s\"", TRANSCRIPT_ID_ATTRIBUTE, line_number,
filename);
had_err = -1;
}
HANDLE_ERROR;
/* process the mandatory attributes */
if (!(transcript_id_hash = gt_hashmap_get(parser->gene_id_hash,
gene_id))) {
transcript_id_hash = gt_hashmap_new(GT_HASH_STRING, gt_free_func,
(GtFree) gt_array_delete);
gt_hashmap_add(parser->gene_id_hash, gt_cstr_dup(gene_id),
transcript_id_hash);
}
gt_assert(transcript_id_hash);
if (!(gt_genome_node_array = gt_hashmap_get(transcript_id_hash,
transcript_id))) {
gt_genome_node_array = gt_array_new(sizeof (GtGenomeNode*));
gt_hashmap_add(transcript_id_hash, gt_cstr_dup(transcript_id),
gt_genome_node_array);
}
gt_assert(gt_genome_node_array);
/* save optional gene_name and transcript_name attributes */
if (transcript_name
&& !gt_hashmap_get(parser->transcript_id_to_name_mapping,
transcript_id)) {
gt_hashmap_add(parser->transcript_id_to_name_mapping,
gt_cstr_dup(transcript_id),
gt_cstr_dup(transcript_name));
}
if (gene_name && !gt_hashmap_get(parser->gene_id_to_name_mapping,
gene_id)) {
gt_hashmap_add(parser->gene_id_to_name_mapping,
gt_cstr_dup(gene_id),
gt_cstr_dup(gene_name));
}
/* get seqid */
seqid_str = gt_hashmap_get(parser->seqid_to_str_mapping, seqname);
if (!seqid_str) {
seqid_str = gt_str_new_cstr(seqname);
gt_hashmap_add(parser->seqid_to_str_mapping, gt_str_get(seqid_str),
seqid_str);
}
gt_assert(seqid_str);
/* construct the new feature */
gn = gt_feature_node_new(seqid_str, type, range.start, range.end,
gt_strand_value);
gt_genome_node_set_origin(gn, filenamestr, line_number);
/* set source */
source_str = gt_hashmap_get(parser->source_to_str_mapping, source);
if (!source_str) {
source_str = gt_str_new_cstr(source);
gt_hashmap_add(parser->source_to_str_mapping, gt_str_get(source_str),
source_str);
}
gt_assert(source_str);
gt_feature_node_set_source((GtFeatureNode*) gn, source_str);
if (score_is_defined)
gt_feature_node_set_score((GtFeatureNode*) gn, score_value);
if (phase_value != GT_PHASE_UNDEFINED)
gt_feature_node_set_phase((GtFeatureNode*) gn, phase_value);
gt_array_add(gt_genome_node_array, gn);
}
gt_str_reset(line_buffer);
}
/* process all region nodes */
if (!had_err)
gt_region_node_builder_build(parser->region_node_builder, genome_nodes);
/* process all feature nodes */
cinfo.genome_nodes = genome_nodes;
cinfo.gene_id_to_name_mapping = parser->gene_id_to_name_mapping;
cinfo.transcript_id_to_name_mapping = parser->transcript_id_to_name_mapping;
if (!had_err) {
had_err = gt_hashmap_foreach(parser->gene_id_hash, construct_genes,
&cinfo, err);
}
/* free */
gt_splitter_delete(splitter);
gt_splitter_delete(attribute_splitter);
gt_str_delete(line_buffer);
return had_err;
}
int gt_queue_unit_test(GtError *err)
{
long check_counter = 0, check_counter_reverse = 1023;
unsigned long i;
int had_err = 0;
GtQueue *q;
gt_error_check(err);
/* without wraparound */
q = gt_queue_new();
gt_ensure(had_err, !gt_queue_size(q));
for (i = 0; !had_err && i < 1024; i++) {
gt_queue_add(q, (void*) i);
gt_ensure(had_err, gt_queue_size(q) == i + 1);
}
if (!had_err)
had_err = gt_queue_iterate(q, check_queue, &check_counter, err);
if (!had_err) {
had_err = gt_queue_iterate_reverse(q, check_queue_reverse,
&check_counter_reverse, err);
}
gt_ensure(had_err, gt_queue_iterate(q, fail_func, NULL, NULL));
gt_ensure(had_err, gt_queue_iterate_reverse(q, fail_func, NULL, NULL));
if (!had_err) {
gt_queue_remove(q, (void*) 0);
gt_ensure(had_err, gt_queue_size(q) == 1023);
}
for (i = 1; !had_err && i < 1024; i++) {
gt_ensure(had_err, gt_queue_head(q) == (void*) i);
gt_ensure(had_err, gt_queue_get(q) == (void*) i);
gt_ensure(had_err, gt_queue_size(q) == 1024 - i - 1);
}
gt_ensure(had_err, !gt_queue_size(q));
gt_queue_delete(q);
/* with wraparound (without full queue) */
if (!had_err) {
q = gt_queue_new();
gt_ensure(had_err, !gt_queue_size(q));
for (i = 0; !had_err && i < 1024; i++) {
gt_queue_add(q, (void*) i);
gt_ensure(had_err, gt_queue_size(q) == i + 1);
}
check_counter = 0;
check_counter_reverse = 1023;
if (!had_err)
had_err = gt_queue_iterate(q, check_queue, &check_counter, err);
gt_ensure(had_err, gt_queue_iterate(q, fail_func, NULL, NULL));
gt_ensure(had_err, gt_queue_iterate_reverse(q, fail_func, NULL, NULL));
if (!had_err) {
had_err = gt_queue_iterate_reverse(q, check_queue_reverse,
&check_counter_reverse, err);
}
for (i = 0; !had_err && i < 512; i++) {
gt_ensure(had_err, gt_queue_head(q) == (void*) i);
gt_ensure(had_err, gt_queue_get(q) == (void*) i);
gt_ensure(had_err, gt_queue_size(q) == 1024 - i - 1);
}
for (i = 0; !had_err && i < 512; i++) {
gt_queue_add(q, (void*) (i + 1024));
gt_ensure(had_err, gt_queue_size(q) == 512 + i + 1);
}
check_counter = 512;
check_counter_reverse = 1535;
if (!had_err)
had_err = gt_queue_iterate(q, check_queue, &check_counter, err);
if (!had_err) {
had_err = gt_queue_iterate_reverse(q, check_queue_reverse,
&check_counter_reverse, err);
}
gt_ensure(had_err, gt_queue_iterate(q, fail_func, NULL, NULL));
gt_ensure(had_err, gt_queue_iterate_reverse(q, fail_func, NULL, NULL));
if (!had_err) {
gt_queue_remove(q, (void*) 512);
gt_ensure(had_err, gt_queue_size(q) == 1023);
}
for (i = 1; !had_err && i < 1024; i++) {
gt_ensure(had_err, gt_queue_head(q) == (void*) (512 + i));
gt_ensure(had_err, gt_queue_get(q) == (void*) (512 + i));
gt_ensure(had_err, gt_queue_size(q) == 1024 - i - 1);
}
gt_ensure(had_err, !gt_queue_size(q));
gt_queue_delete(q);
}
/* with wraparound (with full queue) */
if (!had_err) {
q = gt_queue_new();
gt_ensure(had_err, !gt_queue_size(q));
for (i = 0; !had_err && i < 1024; i++) {
gt_queue_add(q, (void*) i);
gt_ensure(had_err, gt_queue_size(q) == i + 1);
}
check_counter = 0;
check_counter_reverse = 1023;
if (!had_err)
had_err = gt_queue_iterate(q, check_queue, &check_counter, err);
if (!had_err) {
had_err = gt_queue_iterate_reverse(q, check_queue_reverse,
&check_counter_reverse, err);
}
gt_ensure(had_err, gt_queue_iterate(q, fail_func, NULL, NULL));
gt_ensure(had_err, gt_queue_iterate_reverse(q, fail_func, NULL, NULL));
for (i = 0; !had_err && i < 512; i++) {
gt_ensure(had_err, gt_queue_head(q) == (void*) i);
gt_ensure(had_err, gt_queue_get(q) == (void*) i);
gt_ensure(had_err, gt_queue_size(q) == 1024 - i - 1);
}
for (i = 0; !had_err && i < 1024; i++) {
gt_queue_add(q, (void*) (i + 1024));
gt_ensure(had_err, gt_queue_size(q) == 512 + i + 1);
}
check_counter = 512;
check_counter_reverse = 2047;
if (!had_err)
had_err = gt_queue_iterate(q, check_queue, &check_counter, err);
if (!had_err) {
had_err = gt_queue_iterate_reverse(q, check_queue_reverse,
&check_counter_reverse, err);
}
gt_ensure(had_err, gt_queue_iterate(q, fail_func, NULL, NULL));
gt_ensure(had_err, gt_queue_iterate_reverse(q, fail_func, NULL, NULL));
if (!had_err) {
gt_queue_remove(q, (void*) 512);
gt_ensure(had_err, gt_queue_size(q) == 1535);
}
for (i = 1; !had_err && i < 1536; i++) {
gt_ensure(had_err, gt_queue_head(q) == (void*) (512 + i));
gt_ensure(had_err, gt_queue_get(q) == (void*) (512 + i));
gt_ensure(had_err, gt_queue_size(q) == 1536 - i - 1);
}
gt_ensure(had_err, !gt_queue_size(q));
gt_queue_delete(q);
}
/* test a corner case */
if (!had_err) {
q = gt_queue_new();
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 1);
if (!had_err)
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 2);
gt_ensure(had_err, gt_queue_get(q));
gt_ensure(had_err, gt_queue_size(q) == 1);
if (!had_err)
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 2);
gt_ensure(had_err, gt_queue_get(q));
gt_ensure(had_err, gt_queue_size(q) == 1);
if (!had_err)
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 2);
gt_ensure(had_err, gt_queue_get(q));
gt_ensure(had_err, gt_queue_size(q) == 1);
gt_ensure(had_err, gt_queue_get(q));
gt_ensure(had_err, gt_queue_size(q) == 0);
if (!had_err)
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 1);
gt_ensure(had_err, gt_queue_get(q));
gt_ensure(had_err, gt_queue_size(q) == 0);
gt_queue_delete(q);
}
/* gt_queue_remove() corner case */
if (!had_err) {
q = gt_queue_new();
gt_queue_add(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 1);
gt_queue_remove(q, (void*) 1);
gt_ensure(had_err, gt_queue_size(q) == 0);
gt_queue_delete(q);
}
/* gt_queue_remove() corner case */
if (!had_err) {
q = gt_queue_new();
gt_queue_add(q, (void*) 0);
gt_queue_add(q, (void*) 1);
gt_queue_add(q, (void*) 2);
gt_queue_add(q, (void*) 3);
gt_ensure(had_err, gt_queue_get(q) == (void*) 0);
gt_ensure(had_err, gt_queue_get(q) == (void*) 1);
gt_queue_add(q, (void*) 4);
gt_queue_add(q, (void*) 5);
gt_queue_remove(q, (void*) 4);
gt_queue_remove(q, (void*) 2);
gt_queue_remove(q, (void*) 5);
gt_queue_remove(q, (void*) 3);
gt_ensure(had_err, gt_queue_size(q) == 0);
gt_queue_delete(q);
}
/* delete with contents */
if (!had_err) {
q = gt_queue_new();
gt_ensure(had_err, !gt_queue_size(q));
if (!had_err)
gt_queue_add(q, gt_calloc(1, 16));
gt_ensure(had_err, gt_queue_size(q) == 1);
if (!had_err)
gt_queue_add(q, gt_calloc(1, 32));
gt_ensure(had_err, gt_queue_size(q) == 2);
gt_queue_delete_with_contents(q);
}
return had_err;
}
// Main method
int main(int argc, char * const *argv)
{
GtError *error;
GtLogger *logger;
GtQueue *streams;
GtNodeStream *stream, *last_stream;
CanonGFF3Options options = { NULL, NULL, false };
gt_lib_init();
error = gt_error_new();
canon_gff3_parse_options(argc, argv + 0, &options, error);
streams = gt_queue_new();
logger = gt_logger_new(true, "", stderr);
stream = gt_gff3_in_stream_new_unsorted(argc - optind, (const char **)
argv+optind);
gt_gff3_in_stream_check_id_attributes((GtGFF3InStream *)stream);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream *)stream);
gt_queue_add(streams, stream);
last_stream = stream;
if(options.infer)
{
GtHashmap *type_parents = gt_hashmap_new(GT_HASH_STRING, gt_free_func,
gt_free_func);
gt_hashmap_add(type_parents, gt_cstr_dup("mRNA"), gt_cstr_dup("gene"));
gt_hashmap_add(type_parents, gt_cstr_dup("tRNA"), gt_cstr_dup("gene"));
stream = agn_infer_parent_stream_new(last_stream,
type_parents);
gt_hashmap_delete(type_parents);
gt_queue_add(streams, stream);
last_stream = stream;
}
stream = agn_gene_stream_new(last_stream, logger);
gt_queue_add(streams, stream);
last_stream = stream;
if(options.source != NULL)
{
GtNodeVisitor *ssv = gt_set_source_visitor_new(options.source);
stream = gt_visitor_stream_new(last_stream, ssv);
gt_queue_add(streams, stream);
last_stream = stream;
}
stream = gt_gff3_out_stream_new(last_stream, options.outstream);
if(!options.infer)
gt_gff3_out_stream_retain_id_attributes((GtGFF3OutStream *)stream);
gt_queue_add(streams, stream);
last_stream = stream;
if(gt_node_stream_pull(last_stream, error) == -1)
{
fprintf(stderr, "[CanonGFF3] error processing node stream: %s",
gt_error_get(error));
}
while(gt_queue_size(streams) > 0)
{
stream = gt_queue_get(streams);
gt_node_stream_delete(stream);
}
gt_queue_delete(streams);
if(options.source != NULL)
gt_str_delete(options.source);
if(options.outstream != NULL)
gt_file_delete(options.outstream);
gt_error_delete(error);
gt_logger_delete(logger);
gt_lib_clean();
return 0;
}
static int advancefastabufferstate(GtFastaBuffer *fb, GtError *err)
{
int currentchar;
unsigned long currentoutpos = 0, currentfileadd = 0, currentfileread = 0;
GtUchar charcode;
gt_error_check(err);
while (true)
{
if (currentoutpos >= (unsigned long) OUTPUTFILEBUFFERSIZE)
{
if (fb->filelengthtab != NULL)
{
fb->filelengthtab[fb->filenum].length
+= (uint64_t) currentfileread;
fb->filelengthtab[fb->filenum].effectivelength
+= (uint64_t) currentfileadd;
}
break;
}
if (fb->nextfile)
{
if (fb->filelengthtab != NULL)
{
fb->filelengthtab[fb->filenum].length = 0;
fb->filelengthtab[fb->filenum].effectivelength = 0;
}
fb->nextfile = false;
fb->indesc = false;
fb->firstseqinfile = true;
currentfileadd = 0;
currentfileread = 0;
fb->linenum = (uint64_t) 1;
fb->inputstream = gt_file_xopen(gt_str_array_get(fb->filenametab,
(unsigned long) fb->filenum),
"rb");
fb->currentinpos = 0;
fb->currentfillpos = 0;
} else
{
currentchar = ownbuffer_genfile_getc(fb,fb->inputstream);
if (currentchar == EOF)
{
gt_file_delete(fb->inputstream);
fb->inputstream = NULL;
if (fb->filelengthtab != NULL)
{
fb->filelengthtab[fb->filenum].length += currentfileread;
fb->filelengthtab[fb->filenum].effectivelength += currentfileadd;
}
if ((unsigned long) fb->filenum == gt_str_array_size(fb->filenametab)-1)
{
fb->complete = true;
break;
}
fb->filenum++;
fb->nextfile = true;
} else
{
currentfileread++;
if (fb->indesc)
{
if (currentchar == NEWLINESYMBOL)
{
fb->linenum++;
fb->indesc = false;
}
if (fb->descptr != NULL)
{
if (currentchar == NEWLINESYMBOL)
{
GT_STOREINARRAY(&fb->headerbuffer, char, 128, '\0');
gt_queue_add(fb->descptr,
gt_cstr_dup(fb->headerbuffer.spacechar));
fb->headerbuffer.nextfreechar = 0;
} else
{
GT_STOREINARRAY(&fb->headerbuffer, char, 128, currentchar);
}
}
} else
{
if (!isspace((int) currentchar))
{
if (currentchar == FASTASEPARATOR)
{
if (fb->firstoverallseq)
{
fb->firstoverallseq = false;
fb->firstseqinfile = false;
} else
{
if (fb->firstseqinfile)
{
fb->firstseqinfile = false;
} else
{
currentfileadd++;
}
fb->outputbuffer[currentoutpos++] = (GtUchar) SEPARATOR;
fb->lastspeciallength++;
}
fb->indesc = true;
} else
{
if (fb->symbolmap == NULL)
{
fb->outputbuffer[currentoutpos++] = (GtUchar) currentchar;
} else
{
charcode = fb->symbolmap[(unsigned int) currentchar];
if (charcode == (GtUchar) UNDEFCHAR)
{
gt_error_set(err,
"illegal character '%c': file \"%s\", line %llu",
currentchar,
gt_str_array_get(fb->filenametab, fb->filenum),
(unsigned long long) fb->linenum);
return -1;
}
if (ISSPECIAL(charcode))
{
fb->lastspeciallength++;
} else
{
if (fb->lastspeciallength > 0)
{
fb->lastspeciallength = 0;
}
if (fb->characterdistribution != NULL)
{
fb->characterdistribution[charcode]++;
}
}
fb->outputbuffer[currentoutpos++] = charcode;
}
currentfileadd++;
}
}
}
}
GtNodeVisitor*
agn_gaeval_visitor_new(GtNodeStream *astream, AgnGaevalParams gparams)
{
agn_assert(astream);
// Create the node visitor
GtNodeVisitor *nv = gt_node_visitor_create(gaeval_visitor_class());
AgnGaevalVisitor *v = gaeval_visitor_cast(nv);
v->alignments = gt_feature_index_memory_new();
v->tsvout = NULL;
v->params = gparams;
// Check that sum of weights is 1.0
double weights_total = gparams.alpha + gparams.beta +
gparams.gamma + gparams.epsilon;
if(fabs(weights_total - 1.0) > 0.0001)
{
fprintf(stderr, "[AgnGaevalVisitor::agn_gaeval_visitor_new] warning: "
"sum of weights is not 1.0 %.3lf; integrity calculations will be "
"incorrect\n", weights_total);
}
// Set up node stream to load alignment features into memory
GtQueue *streams = gt_queue_new();
GtNodeStream *stream, *last_stream;
GtHashmap *typestokeep = gt_hashmap_new(GT_HASH_STRING, NULL, NULL);
gt_hashmap_add(typestokeep, "cDNA_match", "cDNA_match");
gt_hashmap_add(typestokeep, "EST_match", "EST_match");
gt_hashmap_add(typestokeep, "nucleotide_match", "nucleotide_match");
stream = agn_filter_stream_new(astream, typestokeep);
gt_queue_add(streams, stream);
last_stream = stream;
stream = gt_feature_out_stream_new(last_stream, v->alignments);
gt_queue_add(streams, stream);
last_stream = stream;
stream = gt_inter_feature_stream_new(last_stream, "cDNA_match", "match_gap");
gt_queue_add(streams, stream);
last_stream = stream;
stream = gt_inter_feature_stream_new(last_stream, "EST_match", "match_gap");
gt_queue_add(streams, stream);
last_stream = stream;
stream = gt_inter_feature_stream_new(last_stream, "nucleotide_match",
"match_gap");
gt_queue_add(streams, stream);
last_stream = stream;
// Process the node stream
GtError *error = gt_error_new();
int result = gt_node_stream_pull(last_stream, error);
if(result == -1)
{
fprintf(stderr, "[AEGeAn::AgnGaevalStream] error parsing alignments: %s\n",
gt_error_get(error));
gt_node_visitor_delete(nv);
return NULL;
}
gt_error_delete(error);
gt_hashmap_delete(typestokeep);
while(gt_queue_size(streams) > 0)
{
stream = gt_queue_get(streams);
gt_node_stream_delete(stream);
}
gt_queue_delete(streams);
return nv;
}
static int select_visitor_feature_node(GtNodeVisitor *nv,
GtFeatureNode *fn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *fv;
bool filter_node = false;
gt_error_check(err);
fv = select_visitor_cast(nv);
fv->current_feature++;
if ((!gt_str_length(fv->seqid) || /* no seqid was specified or seqids are
equal */
!gt_str_cmp(fv->seqid, gt_genome_node_get_seqid((GtGenomeNode*) fn))) &&
(!gt_str_length(fv->source) || /* no source was specified or sources are
equal */
!strcmp(gt_str_get(fv->source), gt_feature_node_get_source(fn)))) {
GtRange range = gt_genome_node_get_range((GtGenomeNode*) fn);
/* enforce maximum gene length */
/* XXX: we (spuriously) assume that genes are always root nodes */
if (fn && gt_feature_node_has_type(fn, gt_ft_gene)) {
if (fv->max_gene_length != GT_UNDEF_ULONG &&
gt_range_length(&range) > fv->max_gene_length) {
filter_node = true;
}
else if (fv->max_gene_num != GT_UNDEF_ULONG &&
fv->gene_num >= fv->max_gene_num) {
filter_node = true;
}
else if (fv->min_gene_score != GT_UNDEF_DOUBLE &&
gt_feature_node_get_score(fn) < fv->min_gene_score) {
filter_node = true;
}
else if (fv->max_gene_score != GT_UNDEF_DOUBLE &&
gt_feature_node_get_score(fn) > fv->max_gene_score) {
filter_node = true;
}
else if (fv->feature_num != GT_UNDEF_ULONG &&
fv->feature_num != fv->current_feature) {
filter_node = true;
}
if (!filter_node)
fv->gene_num++; /* gene passed filter */
}
}
else
filter_node = true;
if (!filter_node)
filter_node = filter_contain_range(fn, fv->contain_range);
if (!filter_node)
filter_node = filter_overlap_range(fn, fv->overlap_range);
if (!filter_node)
filter_node = filter_strand(fn, fv->strand);
if (!filter_node)
filter_node = filter_targetstrand(fn, fv->targetstrand);
if (!filter_node)
filter_node = filter_has_CDS(fn, fv->has_CDS);
if (!filter_node)
filter_node = filter_min_average_ssp(fn, fv->min_average_splice_site_prob);
if (filter_node)
gt_genome_node_delete((GtGenomeNode*) fn);
else
gt_queue_add(fv->node_buffer, fn);
return 0;
}
static int gt_sequence_buffer_fasta_advance(GtSequenceBuffer *sb, GtError *err)
{
int currentchar, ret = 0;
unsigned long currentoutpos = 0, currentfileadd = 0, currentfileread = 0;
GtSequenceBufferMembers *pvt;
GtSequenceBufferFasta *sbf;
gt_error_check(err);
sbf = (GtSequenceBufferFasta*) sb;
pvt = sb->pvt;
while (true)
{
if (currentoutpos >= (unsigned long) OUTBUFSIZE)
{
if (pvt->filelengthtab != NULL)
{
pvt->filelengthtab[pvt->filenum].length
+= (uint64_t) currentfileread;
pvt->filelengthtab[pvt->filenum].effectivelength
+= (uint64_t) currentfileadd;
}
break;
}
if (sbf->nextfile)
{
if (pvt->filelengthtab != NULL)
{
pvt->filelengthtab[pvt->filenum].length = 0;
pvt->filelengthtab[pvt->filenum].effectivelength = 0;
}
sbf->nextfile = false;
sbf->indesc = false;
sbf->firstseqinfile = true;
currentfileadd = 0;
currentfileread = 0;
pvt->linenum = (uint64_t) 1;
pvt->inputstream = gt_file_xopen(gt_str_array_get(pvt->filenametab,
(unsigned long) pvt->filenum),
"rb");
pvt->currentinpos = 0;
pvt->currentfillpos = 0;
} else
{
currentchar = inlinebuf_getchar(sb, pvt->inputstream);
if (currentchar == EOF)
{
gt_file_delete(pvt->inputstream);
pvt->inputstream = NULL;
if (pvt->filelengthtab != NULL)
{
pvt->filelengthtab[pvt->filenum].length += currentfileread;
pvt->filelengthtab[pvt->filenum].effectivelength += currentfileadd;
}
if ((unsigned long) pvt->filenum
== gt_str_array_size(pvt->filenametab)-1)
{
pvt->complete = true;
break;
}
pvt->filenum++;
sbf->nextfile = true;
} else
{
currentfileread++;
if (sbf->indesc)
{
if (currentchar == NEWLINESYMBOL)
{
pvt->linenum++;
sbf->indesc = false;
}
if (pvt->descptr != NULL)
{
if (currentchar == NEWLINESYMBOL)
{
gt_queue_add(pvt->descptr,
gt_cstr_dup(gt_str_get(sbf->headerbuffer)));
gt_str_reset(sbf->headerbuffer);
} else
{
gt_str_append_char(sbf->headerbuffer, currentchar);
}
}
} else
{
if (!isspace((int) currentchar))
{
if (currentchar == FASTASEPARATOR)
{
if (sbf->firstoverallseq)
{
sbf->firstoverallseq = false;
sbf->firstseqinfile = false;
} else
{
if (sbf->firstseqinfile)
{
sbf->firstseqinfile = false;
} else
{
currentfileadd++;
}
pvt->outbuf[currentoutpos++] = (GtUchar) SEPARATOR;
pvt->lastspeciallength++;
}
sbf->indesc = true;
} else
{
if ((ret = process_char(sb, currentoutpos, currentchar, err)))
return ret;
currentoutpos++;
currentfileadd++;
}
}
}
}
}
}
if (sbf->firstoverallseq)
{
gt_error_set(err,"no sequences in multiple fasta file(s) %s ...",
gt_str_array_get(pvt->filenametab,0));
return -2;
}
pvt->nextfree = currentoutpos;
return 0;
}
static int add_ids_visitor_feature_node(GtNodeVisitor *nv, GtFeatureNode *fn,
GtError *err)
{
AutomaticSequenceRegion *auto_sr;
GtAddIDsVisitor *aiv;
const char *seqid;
bool is_circular;
aiv = add_ids_visitor_cast(nv);
seqid = gt_str_get(gt_genome_node_get_seqid((GtGenomeNode*) fn));
if (aiv->ensure_sorting && !gt_cstr_table_get(aiv->defined_seqids, seqid)) {
gt_error_set(err, "the file %s is not sorted (seqid \"%s\" on line %u has "
"not been previously introduced with a \"%s\" line)",
gt_genome_node_get_filename((GtGenomeNode*) fn), seqid,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
GT_GFF_SEQUENCE_REGION);
return -1;
}
if (!gt_cstr_table_get(aiv->defined_seqids, seqid)) {
GtFeatureNodeIterator *fni;
GtFeatureNode *node;
GtRange range = gt_genome_node_get_range((GtGenomeNode*) fn);
is_circular = gt_feature_node_get_attribute(fn, GT_GFF_IS_CIRCULAR)
? true : false;
if (!is_circular) {
fni = gt_feature_node_iterator_new(fn);
while ((node = gt_feature_node_iterator_next(fni))) {
GtRange node_range = gt_genome_node_get_range((GtGenomeNode*) node);
range = gt_range_join(&range, &node_range);
}
gt_feature_node_iterator_delete(fni);
}
/* sequence region has not been previously introduced -> check if one has
already been created automatically */
auto_sr = gt_hashmap_get(aiv->undefined_sequence_regions, seqid);
if (!auto_sr) {
GtStr *seqid_str;
/* sequence region has not been createad automatically -> do it now */
gt_warning("seqid \"%s\" on line %u in file \"%s\" has not been "
"previously introduced with a \"%s\" line, create such a line "
"automatically", seqid,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn),
GT_GFF_SEQUENCE_REGION);
auto_sr = automatic_sequence_region_new(is_circular);
seqid_str = gt_genome_node_get_seqid((GtGenomeNode*) fn);
auto_sr->sequence_region = gt_region_node_new(seqid_str, range.start,
range.end);
gt_hashmap_add(aiv->undefined_sequence_regions, gt_str_get(seqid_str),
auto_sr);
}
else {
if (auto_sr->is_circular) {
gt_assert(!is_circular); /* XXX */
}
else if (is_circular) {
gt_assert(!auto_sr->is_circular); /* XXX */
auto_sr->is_circular = true;
gt_genome_node_set_range(auto_sr->sequence_region, &range);
}
else {
GtRange joined_range,
sr_range = gt_genome_node_get_range(auto_sr->sequence_region);
/* update the range of the sequence region */
joined_range = gt_range_join(&range, &sr_range);
gt_genome_node_set_range(auto_sr->sequence_region, &joined_range);
}
}
gt_array_add(auto_sr->feature_nodes, fn);
}
else
gt_queue_add(aiv->node_buffer, fn);
return 0;
}
static int bed_rest(GtBEDParser *bed_parser, GtIO *bed_file, GtError *err)
{
GtUword block_count = 0;
GtGenomeNode *gn = NULL;
GtRange range;
GtStr *seqid;
int had_err;
gt_error_check(err);
/* column 1.: chrom */
seqid = get_seqid(bed_parser);
had_err = skip_blanks(bed_file, err);
/* column 2.: chromStart */
if (!had_err) {
word(bed_parser->word, bed_file);
had_err = skip_blanks(bed_file, err);
}
/* column 3.: chromEnd */
if (!had_err) {
word(bed_parser->another_word, bed_file);
had_err = parse_bed_range(&range, bed_parser->word,
bed_parser->another_word, bed_parser->offset,
bed_file, false, err);
}
if (!had_err) {
/* add region */
gt_region_node_builder_add_region(bed_parser->region_node_builder,
gt_str_get(seqid), range);
/* create feature */
gn = gt_feature_node_new(seqid,
bed_parser->feature_type
? bed_parser->feature_type
: BED_FEATURE_TYPE,
range.start, range.end, GT_STRAND_BOTH);
gt_queue_add(bed_parser->feature_nodes, gn);
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* optional column 4.: name */
if (!had_err) {
word(bed_parser->word, bed_file);
if (gt_str_length(bed_parser->word)) {
gt_feature_node_add_attribute((GtFeatureNode*) gn, GT_GFF_NAME,
gt_str_get(bed_parser->word));
}
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* optional column 5.: score */
if (!had_err) {
word(bed_parser->word, bed_file);
if (gt_str_length(bed_parser->word)) {
bool score_is_defined;
float score_value;
had_err = gt_parse_score(&score_is_defined, &score_value,
gt_str_get(bed_parser->word),
gt_io_get_line_number(bed_file),
gt_io_get_filename(bed_file), err);
if (!had_err && score_is_defined)
gt_feature_node_set_score((GtFeatureNode*) gn, score_value);
}
}
if (!had_err && bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
/* optional column 6.: strand */
if (!had_err) {
word(bed_parser->word, bed_file);
if (gt_str_length(bed_parser->word)) {
GtStrand strand;
had_err = gt_parse_strand(&strand, gt_str_get(bed_parser->word),
gt_io_get_line_number(bed_file),
gt_io_get_filename(bed_file), err);
if (!had_err)
gt_feature_node_set_strand((GtFeatureNode*) gn, strand);
}
}
if (!had_err && bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
/* optional column 7.: thickStart */
if (!had_err) {
word(bed_parser->word, bed_file);
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* optional column 8.: thickEnd */
if (!had_err) {
word(bed_parser->another_word, bed_file);
if (gt_str_length(bed_parser->another_word)) {
gt_assert(gt_str_length(bed_parser->word));
/* got a thickStart and a thickEnd -> construct corresponding feature */
had_err = parse_bed_range(&range, bed_parser->word,
bed_parser->another_word, bed_parser->offset,
bed_file, true, err);
if (!had_err && range.start <= range.end)
construct_thick_feature(bed_parser, (GtFeatureNode*) gn, range);
}
}
if (!had_err && bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
/* optional column 9.: itemRgb */
if (!had_err) {
word(bed_parser->word, bed_file);
/* we do not use the RGB values */
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* optional column 10.: blockCount */
if (!had_err) {
word(bed_parser->word, bed_file);
if (gt_str_length(bed_parser->word)) {
if (gt_parse_uword(&block_count, gt_str_get(bed_parser->word))) {
gt_error_set(err,
"file \"%s\": line "GT_WU": could not parse blockCount",
gt_io_get_filename(bed_file),
gt_io_get_line_number(bed_file));
had_err = -1;
}
else {
/* reset to parse/process blockSizes and blockStarts properly */
gt_str_reset(bed_parser->word);
gt_str_reset(bed_parser->another_word);
}
}
}
if (!had_err && bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
/* optional column 11.: blockSizes */
if (!had_err) {
word(bed_parser->word, bed_file);
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* optional column 12.: blockStarts */
if (!had_err) {
word(bed_parser->another_word, bed_file);
if (bed_separator(bed_file))
had_err = skip_blanks(bed_file, err);
}
/* process blocks if necessary */
if (!had_err && block_count) {
had_err = process_blocks(bed_parser, (GtFeatureNode*) gn, block_count,
bed_parser->word, bed_parser->another_word,
bed_file, err);
}
/* the end of the line should now be reached */
if (!had_err)
had_err = gt_io_expect(bed_file, GT_END_OF_LINE, err);
return had_err;
}