static int gt_encseq_col_do_grep_desc(GtEncseqCol *esc, GtUword *filenum,
GtUword *seqnum, GtStr *seqid,
GtError *err)
{
GtUword j;
const GtSeqInfo *seq_info_ptr;
GtSeqInfo seq_info;
bool match = false;
int had_err = 0;
gt_error_check(err);
gt_assert(esc && filenum && seqnum && seqid);
/* create cache */
if (!esc->grep_cache)
esc->grep_cache = gt_seq_info_cache_new();
/* try to read from cache */
seq_info_ptr = gt_seq_info_cache_get(esc->grep_cache, gt_str_get(seqid));
if (seq_info_ptr) {
*filenum = seq_info_ptr->filenum;
*seqnum = seq_info_ptr->seqnum;
return 0;
}
for (j = 0; !had_err && j < gt_encseq_num_of_sequences(esc->encseq); j++) {
const char *desc;
char *buf;
GtUword desc_len;
desc = gt_encseq_description(esc->encseq, &desc_len, j);
buf = gt_calloc(desc_len + 1, sizeof (char));
memcpy(buf, desc, desc_len * sizeof (char));
had_err = gt_grep(&match, gt_str_get(seqid), buf, err);
gt_free(buf);
if (!had_err && match) {
*filenum = seq_info.filenum =
gt_encseq_filenum(esc->encseq,
gt_encseq_seqstartpos(esc->encseq, j));
*seqnum = seq_info.seqnum =
j - gt_encseq_filenum_first_seqnum(esc->encseq, *filenum);
gt_seq_info_cache_add(esc->grep_cache, gt_str_get(seqid), &seq_info);
break;
}
}
if (!had_err && !match) {
gt_error_set(err, "no description matched sequence ID '%s'",
gt_str_get(seqid));
had_err = -1;
}
return had_err;
}
static int grep_desc(GtBioseqCol *bsc, GtUword *filenum,
GtUword *seqnum, GtStr *seqid, GtError *err)
{
GtUword i, j, num_matches = 0;
const GtSeqInfo *seq_info_ptr;
GtSeqInfo seq_info;
GtStr *pattern, *escaped;
bool match = false;
int had_err = 0;
gt_error_check(err);
gt_assert(bsc && filenum && seqnum && seqid);
/* create cache */
if (!bsc->grep_cache)
bsc->grep_cache = gt_seq_info_cache_new();
/* try to read from cache */
seq_info_ptr = gt_seq_info_cache_get(bsc->grep_cache, gt_str_get(seqid));
if (seq_info_ptr) {
*filenum = seq_info_ptr->filenum;
*seqnum = seq_info_ptr->seqnum;
return 0;
}
pattern = gt_str_new();
escaped = gt_str_new();
gt_grep_escape_extended(escaped, gt_str_get(seqid), gt_str_length(seqid));
if (bsc->matchdescstart)
gt_str_append_cstr(pattern, "^");
gt_str_append_str(pattern, escaped);
if (bsc->matchdescstart)
gt_str_append_cstr(pattern, "([[:space:]]|$)");
for (i = 0; !had_err && i < bsc->num_of_seqfiles; i++) {
GtBioseq *bioseq = bsc->bioseqs[i];
for (j = 0; !had_err && j < gt_bioseq_number_of_sequences(bioseq); j++) {
const char *desc = gt_bioseq_get_description(bioseq, j);
had_err = gt_grep(&match, gt_str_get(pattern), desc, err);
if (!had_err && match) {
num_matches++;
if (num_matches > 1) {
gt_error_set(err, "query seqid '%s' could match more than one "
"sequence description", gt_str_get(seqid));
had_err = -1;
break;
}
*filenum = i;
*seqnum = j;
/* cache results */
seq_info.filenum = i;
seq_info.seqnum = j;
gt_seq_info_cache_add(bsc->grep_cache, gt_str_get(seqid), &seq_info);
}
}
if (match)
break;
}
gt_str_delete(pattern);
gt_str_delete(escaped);
if (!had_err && num_matches == 0) {
gt_error_set(err, "no description matched sequence ID '%s'",
gt_str_get(seqid));
had_err = -1;
}
return had_err;
}
