static int region_node_accept(GtGenomeNode *gn, GtNodeVisitor *nv, GtError *err)
{
GtRegionNode *rn;
gt_error_check(err);
rn = gt_region_node_cast(gn);
return gt_node_visitor_visit_region_node(nv, rn, err);
}
static void region_node_change_seqid(GtGenomeNode *gn, GtStr *seqid)
{
GtRegionNode *rn = gt_region_node_cast(gn);
gt_assert(rn && seqid);
gt_str_delete(rn->seqid);
rn->seqid = gt_str_ref(seqid);
}
static GtFeatureIndex *in_stream_test_data(GtError *error)
{
GtFeatureIndex *fi;
GtFeatureNode *fn;
GtGenomeNode *gn;
GtRegionNode *rn;
GtStr *seqid;
fi = gt_feature_index_memory_new();
seqid = gt_str_new_cstr("chr1");
gn = gt_region_node_new(seqid, 1, 100000);
rn = gt_region_node_cast(gn);
gt_feature_index_add_region_node(fi, rn, error);
gt_genome_node_delete(gn);
gn = gt_feature_node_new(seqid, "region", 500, 5000, GT_STRAND_BOTH);
fn = gt_feature_node_cast(gn);
gt_feature_index_add_feature_node(fi, fn, error);
gt_genome_node_delete(gn);
gn = gt_feature_node_new(seqid, "region", 50000, 75000, GT_STRAND_BOTH);
fn = gt_feature_node_cast(gn);
gt_feature_index_add_feature_node(fi, fn, error);
gt_genome_node_delete(gn);
gt_str_delete(seqid);
seqid = gt_str_new_cstr("scf0001");
gn = gt_region_node_new(seqid, 1, 10000);
rn = gt_region_node_cast(gn);
gt_feature_index_add_region_node(fi, rn, error);
gt_genome_node_delete(gn);
gn = gt_feature_node_new(seqid, "mRNA", 4000, 6000, GT_STRAND_REVERSE);
fn = gt_feature_node_cast(gn);
gt_feature_index_add_feature_node(fi, fn, error);
gt_genome_node_delete(gn);
gn = gt_feature_node_new(seqid, "mRNA", 7000, 9500, GT_STRAND_FORWARD);
fn = gt_feature_node_cast(gn);
gt_feature_index_add_feature_node(fi, fn, error);
gt_genome_node_delete(gn);
gt_str_delete(seqid);
return fi;
}
GtGenomeNode* gt_region_node_new(GtStr *seqid, unsigned long start,
unsigned long end)
{
GtGenomeNode *gn = gt_genome_node_create(gt_region_node_class());
GtRegionNode *rn = gt_region_node_cast(gn);
gt_assert(seqid);
gt_assert(start <= end);
rn->seqid = gt_str_ref(seqid);
rn->range.start = start;
rn->range.end = end;
return gn;
}
static void region_node_set_range(GtGenomeNode *gn, const GtRange *range)
{
GtRegionNode *rn = gt_region_node_cast(gn);
rn->range = *range;
}
static GtRange region_node_get_range(GtGenomeNode *gn)
{
GtRegionNode *rn = gt_region_node_cast(gn);
return rn->range;
}
static GtStr* region_node_get_seqid(GtGenomeNode *gn)
{
GtRegionNode *rn = gt_region_node_cast(gn);
return rn->seqid;
}
static void region_node_free(GtGenomeNode *gn)
{
GtRegionNode *rn = gt_region_node_cast(gn);
gt_assert(rn && rn->seqid);
gt_str_delete(rn->seqid);
}
static int chseqids_stream_next(GtNodeStream *gs, GtGenomeNode **gn,
GtError *err)
{
GtChseqidsStream *cs;
GtGenomeNode *node, **gn_a, **gn_b;
GtFeatureNode *feature_node;
GtStr *changed_seqid;
unsigned long i;
int rval, had_err = 0;
gt_error_check(err);
cs = chseqids_stream_cast(gs);
if (!cs->sequence_regions_processed) {
while (!had_err) {
if (!(had_err = gt_node_stream_next(cs->in_stream, &node, err))) {
if (node)
gt_array_add(cs->gt_genome_node_buffer, node);
else
break;
if (!gt_region_node_try_cast(node))
break; /* no more sequence regions */
}
}
/* now the buffer contains only sequence regions (except the last entry)
-> change sequence ids */
for (i = 0; !had_err && i < gt_array_size(cs->gt_genome_node_buffer); i++) {
node = *(GtGenomeNode**) gt_array_get(cs->gt_genome_node_buffer, i);
if (gt_genome_node_get_seqid(node)) {
if ((changed_seqid = gt_mapping_map_string(cs->chseqids_mapping,
gt_str_get(gt_genome_node_get_seqid(node)),
err))) {
if ((feature_node = gt_feature_node_try_cast(node))) {
rval = gt_genome_node_traverse_children(node, changed_seqid,
change_sequence_id, true,
err);
gt_assert(!rval); /* change_sequence_id() is sane */
}
else
gt_genome_node_change_seqid(node, changed_seqid);
gt_str_delete(changed_seqid);
}
else
had_err = -1;
}
}
/* sort them */
if (!had_err)
gt_genome_nodes_sort(cs->gt_genome_node_buffer);
/* consolidate them */
for (i = 1; !had_err && i + 1 < gt_array_size(cs->gt_genome_node_buffer);
i++) {
gn_a = gt_array_get(cs->gt_genome_node_buffer, i-1);
gn_b = gt_array_get(cs->gt_genome_node_buffer, i);
if (gt_genome_nodes_are_equal_region_nodes(*gn_a, *gn_b)) {
gt_region_node_consolidate(gt_region_node_cast(*gn_b),
gt_region_node_cast(*gn_a));
gt_genome_node_delete(*gn_a);
*gn_a = NULL;
}
}
cs->sequence_regions_processed = true;
}
/* return non-null nodes from buffer */
while (!had_err &&
cs->buffer_index < gt_array_size(cs->gt_genome_node_buffer)) {
node = *(GtGenomeNode**) gt_array_get(cs->gt_genome_node_buffer,
cs->buffer_index);
cs->buffer_index++;
if (node) {
*gn = node;
return had_err;
}
}
if (!had_err)
had_err = gt_node_stream_next(cs->in_stream, gn, err);
if (!had_err && *gn) {
if (gt_genome_node_get_seqid(*gn)) {
changed_seqid = gt_mapping_map_string(cs->chseqids_mapping,
gt_str_get(gt_genome_node_get_seqid(*gn)),
err);
gt_assert(changed_seqid); /* is always defined, because an undefined
mapping would be catched earlier */
if ((feature_node = gt_feature_node_try_cast(*gn))) {
rval = gt_genome_node_traverse_children(*gn, changed_seqid,
change_sequence_id, true, err);
gt_assert(!rval); /* change_sequence_id() is sane */
}
else
gt_genome_node_change_seqid(*gn, changed_seqid);
gt_str_delete(changed_seqid);
}
}
return had_err;
}