static int m2i_change_target_seqids(GtFeatureNode *fn, const char *target,
GtRegionMapping *region_mapping,
GtError *err)
{
GtStrArray *target_ids;
GtArray *target_ranges, *target_strands;
GtStr *desc, *new_seqid;
unsigned long i;
int had_err;
gt_error_check(err);
gt_assert(fn && target && region_mapping);
target_ids = gt_str_array_new();
target_ranges = gt_array_new(sizeof (GtRange));
target_strands = gt_array_new(sizeof (GtStrand));
desc = gt_str_new();
new_seqid = gt_str_new();
had_err = gt_gff3_parser_parse_all_target_attributes(target, false,
target_ids,
target_ranges,
target_strands, "", 0,
err);
for (i = 0; !had_err && i < gt_str_array_size(target_ids); i++) {
GtStr *seqid;
gt_str_reset(desc);
gt_str_reset(new_seqid);
seqid = gt_str_array_get_str(target_ids, i);
had_err = gt_region_mapping_get_description(region_mapping, desc, seqid,
err);
if (!had_err)
gt_regular_seqid_save(new_seqid, desc);
gt_str_array_set(target_ids, i, new_seqid);
}
if (!had_err) {
GtStr *new_target = gt_str_new();
gt_gff3_parser_build_target_str(new_target, target_ids, target_ranges,
target_strands);
gt_feature_node_set_attribute(fn, GT_GFF_TARGET, gt_str_get(new_target));
gt_str_delete(new_target);
}
gt_str_delete(new_seqid);
gt_str_delete(desc);
gt_array_delete(target_strands);
gt_array_delete(target_ranges);
gt_str_array_delete(target_ids);
return had_err;
}
static int gt_extract_feature_sequence_generic(GtStr *sequence,
GtGenomeNode *gn,
const char *type, bool join, GtStr *seqid,
GtStrArray *target_ids,
unsigned int *out_phase_offset,
GtRegionMapping *region_mapping, GtError *err)
{
GtFeatureNode *fn;
GtRange range;
unsigned int phase_offset = 0;
char *outsequence;
const char *target;
int had_err = 0;
gt_error_check(err);
fn = gt_genome_node_cast(gt_feature_node_class(), gn);
gt_assert(fn);
if (seqid)
gt_str_append_str(seqid, gt_genome_node_get_seqid(gn));
if (target_ids &&
(target = gt_feature_node_get_attribute(fn, GT_GFF_TARGET))) {
had_err = gt_gff3_parser_parse_all_target_attributes(target, false,
target_ids, NULL,
NULL, "", 0, err);
}
if (!had_err) {
if (join) {
GtFeatureNodeIterator *fni;
GtFeatureNode *child;
bool reverse_strand = false,
first_child = true,
first_child_of_type_seen = false;
GtPhase phase = GT_PHASE_UNDEFINED;
/* in this case we have to traverse the children */
fni = gt_feature_node_iterator_new_direct(gt_feature_node_cast(gn));
while (!had_err && (child = gt_feature_node_iterator_next(fni))) {
if (first_child) {
if (target_ids &&
(target = gt_feature_node_get_attribute(child, GT_GFF_TARGET))) {
gt_str_array_reset(target_ids);
had_err = gt_gff3_parser_parse_all_target_attributes(target, false,
target_ids,
NULL,
NULL, "", 0,
err);
}
first_child = false;
}
if (!had_err) {
if (extract_join_feature((GtGenomeNode*) child, type, region_mapping,
sequence, &reverse_strand,
&first_child_of_type_seen,
&phase, err)) {
had_err = -1;
}
if (phase != GT_PHASE_UNDEFINED) {
phase_offset = (int) phase;
}
}
}
gt_feature_node_iterator_delete(fni);
gt_assert(phase_offset <= (unsigned int) GT_PHASE_UNDEFINED);
if (!had_err && gt_str_length(sequence)) {
if (reverse_strand) {
had_err = gt_reverse_complement(gt_str_get(sequence),
gt_str_length(sequence), err);
}
}
}
else if (gt_feature_node_get_type(fn) == type) {
GtPhase phase = gt_feature_node_get_phase(fn);
gt_assert(!had_err);
if (phase != GT_PHASE_UNDEFINED)
phase_offset = (unsigned int) phase;
/* otherwise we only have to look at this feature */
range = gt_genome_node_get_range(gn);
gt_assert(range.start); /* 1-based coordinates */
had_err = gt_region_mapping_get_sequence(region_mapping, &outsequence,
gt_genome_node_get_seqid(gn),
range.start, range.end, err);
if (!had_err) {
gt_str_append_cstr_nt(sequence, outsequence, gt_range_length(&range));
gt_free(outsequence);
if (gt_feature_node_get_strand(fn) == GT_STRAND_REVERSE) {
had_err = gt_reverse_complement(gt_str_get(sequence),
gt_str_length(sequence), err);
}
}
}
}
if (out_phase_offset && phase_offset != GT_PHASE_UNDEFINED) {
*out_phase_offset = phase_offset;
}
return had_err;
}
