static int gt_sort_stream_next(GtNodeStream *ns, GtGenomeNode **gn,
GtError *err)
{
GtSortStream *sort_stream;
GtGenomeNode *node, *eofn;
int had_err = 0;
gt_error_check(err);
sort_stream = gt_sort_stream_cast(ns);
if (!sort_stream->sorted) {
while (!(had_err = gt_node_stream_next(sort_stream->in_stream, &node,
err)) && node) {
if ((eofn = gt_eof_node_try_cast(node)))
gt_genome_node_delete(eofn); /* get rid of EOF nodes */
else
gt_array_add(sort_stream->nodes, node);
}
if (!had_err) {
gt_genome_nodes_sort_stable(sort_stream->nodes);
sort_stream->sorted = true;
}
}
if (!had_err) {
gt_assert(sort_stream->sorted);
if (sort_stream->idx < gt_array_size(sort_stream->nodes)) {
*gn = *(GtGenomeNode**) gt_array_get(sort_stream->nodes,
sort_stream->idx);
sort_stream->idx++;
/* join region nodes with the same sequence ID */
if (gt_region_node_try_cast(*gn)) {
GtRange range_a, range_b;
while (sort_stream->idx < gt_array_size(sort_stream->nodes)) {
node = *(GtGenomeNode**) gt_array_get(sort_stream->nodes,
sort_stream->idx);
if (!gt_region_node_try_cast(node) ||
gt_str_cmp(gt_genome_node_get_seqid(*gn),
gt_genome_node_get_seqid(node))) {
/* the next node is not a region node with the same ID */
break;
}
range_a = gt_genome_node_get_range(*gn);
range_b = gt_genome_node_get_range(node);
range_a = gt_range_join(&range_a, &range_b);
gt_genome_node_set_range(*gn, &range_a);
gt_genome_node_delete(node);
sort_stream->idx++;
}
}
return 0;
}
}
if (!had_err) {
gt_array_reset(sort_stream->nodes);
*gn = NULL;
}
return had_err;
}
static int feature_index_lua_add_region_node(lua_State *L)
{
GtFeatureIndex **fi;
GtGenomeNode **gn;
GtRegionNode *rn;
gt_assert(L);
fi = check_feature_index(L, 1);
gn = check_genome_node(L, 2);
rn = gt_region_node_try_cast(*gn);
luaL_argcheck(L, rn, 2, "not a region node");
gt_feature_index_add_region_node(*fi, rn);
return 0;
}
static int feature_index_lua_add_region_node(lua_State *L)
{
GtFeatureIndex **fi;
GtGenomeNode **gn;
GtRegionNode *rn;
GtError *err;
gt_assert(L);
fi = check_feature_index(L, 1);
gn = check_genome_node(L, 2);
rn = gt_region_node_try_cast(*gn);
luaL_argcheck(L, rn, 2, "not a region node");
err = gt_error_new();
if (gt_feature_index_add_region_node(*fi, rn, err))
return gt_lua_error(L, err);
gt_error_delete(err);
return 0;
}
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;
}