static int gt_genomediff_arguments_check(int rest_argc,
void *tool_arguments,
GtError *err)
{
GtGenomediffArguments *arguments = tool_arguments;
bool prepared_index;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if (rest_argc == 0) {
gt_error_set(err, "give at least one file (base)name!");
had_err = -1;
}
if (!had_err) {
if (strcmp("esa", gt_str_get(arguments->indextype)) == 0)
arguments->with_esa = true;
else if (strcmp("pck", gt_str_get(arguments->indextype)) == 0)
arguments->with_pck = true;
}
prepared_index = (arguments->with_esa || arguments->with_pck);
if (!had_err && arguments->user_max_depth != -1 && !arguments->with_pck)
gt_warning("option -maxdepth does only apply to -indextype pck");
if (!had_err &&
prepared_index && gt_encseq_options_mirrored_value(arguments->loadopts))
gt_warning("option -mirrored is ignored with esa and pck index");
if (!had_err && prepared_index && rest_argc > 1) {
gt_error_set(err, "there should be only one basename argument with "
"-indextype esa|pck");
had_err = -1;
}
if (rest_argc == 1 && gt_str_length(arguments->indexname) != 0) {
gt_error_set(err, "Option -indexname is only needed with sequence files, "
"if one file is given as argument, this should be an index.");
had_err = -1;
}
if (!had_err && rest_argc > 1 && gt_str_length(arguments->indexname) == 0) {
gt_error_set(err, "more than one input file given, please use -indexname "
"for basename of indices created during run.");
had_err = -1;
}
if (!had_err)
arguments->with_units = gt_option_is_set(arguments->ref_unitfile);
return had_err;
}
static int check_boundaries_visitor_check_rec(GtFeatureNode *parent,
GtFeatureNode *child,
GtError *err)
{
GtFeatureNodeIterator *fni;
GtFeatureNode *node;
GtRange range,
p_range;
int had_err = 0;
range = gt_genome_node_get_range((GtGenomeNode*) child);
p_range = gt_genome_node_get_range((GtGenomeNode*) parent);
if (range.start < p_range.start || range.end > p_range.end) {
gt_warning("%s child range " GT_WU "-" GT_WU " (file %s, line %u) not "
"contained in %s parent range " GT_WU "-" GT_WU " (file %s, "
"line %u)",
gt_feature_node_get_type(child),
range.start, range.end,
gt_genome_node_get_filename((GtGenomeNode*) child),
gt_genome_node_get_line_number((GtGenomeNode*) child),
gt_feature_node_get_type(parent),
p_range.start, p_range.end,
gt_genome_node_get_filename((GtGenomeNode*) parent),
gt_genome_node_get_line_number((GtGenomeNode*) parent));
}
fni = gt_feature_node_iterator_new_direct(child);
while ((node = gt_feature_node_iterator_next(fni))) {
had_err = check_boundaries_visitor_check_rec(child, node, err);
}
gt_feature_node_iterator_delete(fni);
return had_err;
}
static int gt_splicesiteinfo_runner(int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
SpliceSiteInfoArguments *arguments = tool_arguments;
GtNodeStream *gff3_in_stream = NULL,
*add_introns_stream = NULL,
*splice_site_info_stream = NULL;
GtRegionMapping *region_mapping;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if (!had_err) {
/* create gff3 input stream */
gff3_in_stream = gt_gff3_in_stream_new_unsorted(argc - parsed_args,
argv + parsed_args);
/* create region mapping */
region_mapping = gt_seqid2file_region_mapping_new(arguments->s2fi, err);
if (!region_mapping)
had_err = -1;
}
if (!had_err) {
/* create addintrons stream (if necessary) */
if (arguments->addintrons)
add_introns_stream = gt_add_introns_stream_new(gff3_in_stream);
/* create extract feature stream */
splice_site_info_stream = gt_splice_site_info_stream_new(
arguments->addintrons
? add_introns_stream
: gff3_in_stream,
region_mapping);
/* pull the features through the stream and free them afterwards */
had_err = gt_node_stream_pull(splice_site_info_stream, err);
}
if (!had_err) {
if (!gt_splice_site_info_stream_show(splice_site_info_stream,
arguments->outfp)) {
gt_warning("input file(s) contained no intron, use option -addintrons to "
"add introns automatically");
}
}
/* free */
gt_node_stream_delete(splice_site_info_stream);
gt_node_stream_delete(add_introns_stream);
gt_node_stream_delete(gff3_in_stream);
return had_err;
}
static int toolbox_iterate(void *key, void *value, void *data,
GT_UNUSED GtError *err)
{
const char *name = key;
GtToolinfo *toolinfo = value;
IterateInfo *info = data;
gt_error_check(err);
gt_assert(key && value && data);
if (!toolinfo->hidden) {
if (toolinfo->tool)
info->func(name, toolinfo->tool, info->data);
else
gt_warning("skipping tool '%s' in iterator (not a GtTool object)", name);
}
return 0;
}
static int gt_speck_arguments_check(GT_UNUSED int rest_argc,
void *tool_arguments,
GT_UNUSED GtError *err)
{
SpeccheckArguments *arguments = tool_arguments;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if (arguments->outfp && arguments->colored) {
gt_warning("file output requested ('-o'), disabling colored output");
arguments->colored = false;
}
return had_err = 0;
}
static int gt_speck_arguments_check(GT_UNUSED int rest_argc,
void *tool_arguments,
GT_UNUSED GtError *err)
{
SpeccheckArguments *arguments = tool_arguments;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if ((arguments->outfp || (!arguments->outfp && !isatty(STDOUT_FILENO)))
&& arguments->colored) {
gt_warning("not printing to terminal, disabling colored output");
arguments->colored = false;
}
return had_err = 0;
}
static int determine_outfp(void *data, GtError *err)
{
GtOutputFileInfo *ofi = (GtOutputFileInfo*) data;
GtFileMode file_mode;
int had_err = 0;
gt_error_check(err);
gt_assert(ofi);
if (!gt_str_length(ofi->output_filename))
*ofi->outfp = NULL; /* no output file given -> use stdout */
else { /* outputfile given -> create generic file pointer */
gt_assert(!(ofi->gzip && ofi->bzip2));
if (ofi->gzip)
file_mode = GT_FILE_MODE_GZIP;
else if (ofi->bzip2)
file_mode = GT_FILE_MODE_BZIP2;
else
file_mode = GT_FILE_MODE_UNCOMPRESSED;
if (file_mode != GT_FILE_MODE_UNCOMPRESSED &&
strcmp(gt_str_get(ofi->output_filename) +
gt_str_length(ofi->output_filename) -
strlen(gt_file_mode_suffix(file_mode)),
gt_file_mode_suffix(file_mode))) {
gt_warning("output file '%s' doesn't have correct suffix '%s', appending "
"it", gt_str_get(ofi->output_filename),
gt_file_mode_suffix(file_mode));
gt_str_append_cstr(ofi->output_filename, gt_file_mode_suffix(file_mode));
}
if (!ofi->force && gt_file_exists(gt_str_get(ofi->output_filename))) {
gt_error_set(err, "file \"%s\" exists already, use option -%s to "
"overwrite", gt_str_get(ofi->output_filename),
GT_FORCE_OPT_CSTR);
had_err = -1;
}
if (!had_err) {
*ofi->outfp = gt_file_xopen_file_mode(file_mode,
gt_str_get(ofi->output_filename),
"w");
gt_assert(*ofi->outfp);
}
}
return had_err;
}
static GtStr* make_id_unique(GtGFF3Visitor *gff3_visitor, GtFeatureNode *fn)
{
GtUword i = 1;
GtStr *id = gt_str_new_cstr(gt_feature_node_get_attribute(fn, "ID"));
if (gt_cstr_table_get(gff3_visitor->used_ids, gt_str_get(id))) {
GtStr *buf = gt_str_new();
while (!id_string_is_unique(id, buf, gff3_visitor->used_ids, i++));
gt_warning("feature ID \"%s\" not unique: changing to %s", gt_str_get(id),
gt_str_get(buf));
gt_str_set(id, gt_str_get(buf));
gt_str_delete(buf);
}
/* update table with the new id */
gt_cstr_table_add(gff3_visitor->used_ids, gt_str_get(id));
/* store (unique) id */
gt_hashmap_add(gff3_visitor->feature_node_to_unique_id_str, fn, id);
return id;
}
void gt_lib_init(void)
{
const char *bookkeeping;
bookkeeping = getenv("GT_MEM_BOOKKEEPING");
gt_ma_init(bookkeeping && !strcmp(bookkeeping, "on"));
proc_env_options();
if (spacepeak && !(bookkeeping && !strcmp(bookkeeping, "on")))
gt_warning("GT_ENV_OPTIONS=-spacepeak used without GT_MEM_BOOKKEEPING=on");
gt_fa_init();
if (spacepeak) {
gt_spacepeak_init();
gt_ma_enable_global_spacepeak();
gt_fa_enable_global_spacepeak();
}
gt_log_init();
if (showtime) gt_showtime_enable();
gt_symbol_init();
gt_class_prealloc_run();
gt_ya_rand_init(0);
}
int gt_ltrdigest_arguments_check(GT_UNUSED int rest_argc, void *tool_arguments,
GtError* err)
{
GtLTRdigestOptions *arguments = tool_arguments;
int had_err = 0;
if (arguments->nthreads > 0) {
gt_warning("The '-threads' option is deprecated. Please use the '-j'"
"option of the 'gt' call instead, e.g.:\n"
" gt -j %lu ltrdigest ...", arguments->nthreads);
}
/* -trnas */
if (!had_err && arguments->trna_lib && gt_str_length(arguments->trna_lib) > 0)
{
if (!gt_file_exists(gt_str_get(arguments->trna_lib)))
{
gt_error_set(err, "File '%s' does not exist!",
gt_str_get(arguments->trna_lib));
had_err = -1;
}
}
if (!had_err)
{
GtHMM *hmm;
GtAlphabet *alpha;
alpha = gt_alphabet_new_dna();
hmm = gt_ppt_hmm_new(alpha, &arguments->ppt_opts);
if (!hmm)
{
gt_error_set(err, "PPT HMM parameters are not valid!");
had_err = -1;
}
else
gt_hmm_delete(hmm);
gt_alphabet_delete(alpha);
}
return had_err;
}
static int gtf_show_feature_node(GtFeatureNode *fn, void *data, GtError *err)
{
GtGTFVisitor *gtf_visitor = (GtGTFVisitor*) data;
int had_err = 0;
if (gt_feature_node_has_type(fn, gt_ft_gene)) {
gtf_visitor->gene_id++;
gtf_visitor->transcript_id = 0;
had_err = gtf_show_transcript(fn, gtf_visitor, err);
}
else if (gt_feature_node_has_type(fn, gt_ft_mRNA)) {
had_err = gtf_show_transcript(fn, gtf_visitor, err);
}
else if (!(gt_feature_node_has_type(fn, gt_ft_CDS) ||
gt_feature_node_has_type(fn, gt_ft_exon))) {
gt_warning("skipping GFF3 feature of type \"%s\" (from line %u in file "
"\"%s\")",
gt_feature_node_get_type(fn),
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn));
}
return had_err;
}
int gt_ltrdigest_arguments_check(GT_UNUSED int rest_argc, void *tool_arguments,
GtError* err)
{
GtLTRdigestOptions *arguments = tool_arguments;
int had_err = 0;
if (arguments->nthreads > 0) {
gt_warning("The '-threads' option is deprecated. Please use the '-j'"
"option of the 'gt' call instead, e.g.:\n"
" gt -j "GT_WU" ltrdigest ...", arguments->nthreads);
}
/* -trnas */
if (!had_err && arguments->trna_lib && gt_str_length(arguments->trna_lib) > 0)
{
if (!gt_file_exists(gt_str_get(arguments->trna_lib)))
{
gt_error_set(err, "File '%s' does not exist!",
gt_str_get(arguments->trna_lib));
had_err = -1;
}
}
return had_err;
}
static int gt_xrf_abbr_parse_tree_tag_line(GtIO *xrf_abbr_file, GtStr *tag,
GtStr *value, GtError *err)
{
int had_err = 0;
gt_error_check(err);
gt_log_log("tag");
gt_assert(xrf_abbr_file && tag && value);
do {
had_err = gt_xrf_abbr_parse_tree_proc_any_char(xrf_abbr_file, tag,
false, err);
} while (!had_err && gt_xrf_abbr_parse_tree_any_char(xrf_abbr_file, false));
if (!had_err)
had_err = gt_io_expect(xrf_abbr_file, XRF_SEPARATOR_CHAR, err);
while (!had_err && gt_io_peek(xrf_abbr_file) == XRF_BLANK_CHAR)
gt_io_next(xrf_abbr_file);
if (!had_err) {
do {
had_err = gt_xrf_abbr_parse_tree_proc_any_char(xrf_abbr_file, value,
true, err);
} while (!had_err && gt_xrf_abbr_parse_tree_any_char(xrf_abbr_file, true));
}
if (!had_err) {
if (gt_io_peek(xrf_abbr_file) == XRF_COMMENT_CHAR)
had_err = gt_xrf_abbr_parse_tree_comment_line(xrf_abbr_file, err);
else
had_err = gt_io_expect(xrf_abbr_file, GT_END_OF_LINE, err);
}
if (!had_err && !gt_xrf_abbr_parse_tree_valid_label(gt_str_get(tag))) {
gt_warning("file \"%s\": line "GT_WU": unknown label \"%s\"",
gt_io_get_filename(xrf_abbr_file),
gt_io_get_line_number(xrf_abbr_file),
gt_str_get(tag));
}
gt_log_log("parsed line %s/%s", gt_str_get(tag), gt_str_get(value));
return had_err;
}
static int construct_mRNAs(GT_UNUSED void *key, void *value, void *data,
GtError *err)
{
ConstructionInfo *cinfo = (ConstructionInfo*) data;
GtArray *gt_genome_node_array = (GtArray*) value,
*mRNAs = (GtArray*) cinfo->mRNAs;
GtGenomeNode *mRNA_node, *first_node, *gn;
const char *tname;
GtStrand mRNA_strand;
GtRange mRNA_range;
GtStr *mRNA_seqid;
GtUword i;
int had_err = 0;
gt_error_check(err);
gt_assert(key && value && data);
/* at least one node in array */
gt_assert(gt_array_size(gt_genome_node_array));
/* determine the range and the strand of the mRNA */
first_node = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, 0);
mRNA_range = gt_genome_node_get_range(first_node);
mRNA_strand = gt_feature_node_get_strand((GtFeatureNode*) first_node);
mRNA_seqid = gt_genome_node_get_seqid(first_node);
/* TODO: support discontinuous start/stop codons */
for (i = 0; !had_err && i < gt_array_size(gt_genome_node_array); i++) {
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
if (gt_feature_node_get_attribute((GtFeatureNode*) gn,
GTF_PARSER_STOP_CODON_FLAG)) {
GtUword j;
GtRange stop_codon_rng = gt_genome_node_get_range(gn);
bool found_cds = false;
for (j = 0; !had_err && j < gt_array_size(gt_genome_node_array); j++) {
GtGenomeNode* gn2;
GtRange this_rng;
const char *this_type;
gn2 = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, j);
if (gn == gn2) continue;
this_rng = gt_genome_node_get_range(gn2);
this_type = gt_feature_node_get_type((GtFeatureNode*) gn2);
if (this_type == gt_symbol(gt_ft_CDS)) {
if (gt_range_contains(&this_rng, &stop_codon_rng)) {
if (cinfo->tidy) {
gt_warning("stop codon on line %u in file %s is contained in "
"CDS in line %u",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn),
gt_genome_node_get_line_number(gn2));
found_cds = true;
} else {
gt_error_set(err, "stop codon on line %u in file %s is "
"contained in CDS in line %u",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn),
gt_genome_node_get_line_number(gn2));
had_err = -1;
}
break;
}
if (this_rng.end + 1 == stop_codon_rng.start) {
this_rng.end = stop_codon_rng.end;
gt_genome_node_set_range(gn2, &this_rng);
found_cds = true;
break;
}
if (this_rng.start == stop_codon_rng.end + 1) {
this_rng.start = stop_codon_rng.start;
gt_genome_node_set_range(gn2, &this_rng);
found_cds = true;
break;
}
}
}
if (!found_cds) {
if (!had_err) {
if (cinfo->tidy) {
gt_warning("found stop codon on line %u in file %s with no "
"flanking CDS, ignoring it",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn));
} else {
gt_error_set(err, "found stop codon on line %u in file %s with no "
"flanking CDS",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn));
had_err = -1;
break;
}
}
} else {
gt_array_rem(gt_genome_node_array, i);
gt_genome_node_delete(gn);
}
}
}
for (i = 1; !had_err && i < gt_array_size(gt_genome_node_array); i++) {
GtRange range;
GtStrand strand;
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
range = gt_genome_node_get_range(gn);
mRNA_range = gt_range_join(&mRNA_range, &range);
strand = gt_feature_node_get_strand((GtFeatureNode*) gn);
if (strand != mRNA_strand) {
gt_error_set(err, "feature %s on line %u has strand %c, but the "
"parent transcript has strand %c",
(const char*) key,
gt_genome_node_get_line_number(gn),
GT_STRAND_CHARS[strand],
GT_STRAND_CHARS[mRNA_strand]);
had_err = -1;
break;
} else {
mRNA_strand = gt_strand_join(mRNA_strand, strand);
}
if (!had_err && gt_str_cmp(mRNA_seqid, gt_genome_node_get_seqid(gn))) {
gt_error_set(err, "The features on lines %u and %u refer to different "
"genomic sequences (``seqname''), although they have the same "
"gene IDs (``gene_id'') which must be globally unique",
gt_genome_node_get_line_number(first_node),
gt_genome_node_get_line_number(gn));
had_err = -1;
break;
}
}
if (!had_err) {
mRNA_node = gt_feature_node_new(mRNA_seqid, gt_ft_mRNA, mRNA_range.start,
mRNA_range.end, mRNA_strand);
gt_feature_node_add_attribute(((GtFeatureNode*) mRNA_node), "ID", key);
gt_feature_node_add_attribute(((GtFeatureNode*) mRNA_node), "transcript_id",
key);
if ((tname = gt_hashmap_get(cinfo->transcript_id_to_name_mapping,
(const char*) key)) && strlen(tname) > 0) {
gt_feature_node_add_attribute((GtFeatureNode*) mRNA_node, GT_GFF_NAME,
tname);
}
/* register children */
for (i = 0; i < gt_array_size(gt_genome_node_array); i++) {
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
gt_feature_node_add_child((GtFeatureNode*) mRNA_node,
(GtFeatureNode*) gt_genome_node_ref(gn));
}
/* store the mRNA */
gt_array_add(mRNAs, mRNA_node);
}
return had_err;
}
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 condenseq_io(GtCondenseq *condenseq,
FILE* fp,
GtIOFunc io_func,
GtError *err)
{
int had_err = 0;
int file_format = GT_CONDENSEQ_VERSION;
GtUword idx;
had_err = gt_condenseq_io_one(condenseq->orig_length);
if (!had_err)
had_err = gt_condenseq_io_one(file_format);
if (!had_err && file_format != GT_CONDENSEQ_VERSION) {
gt_error_set(err, "condenseq index is format version %d, current is "
"%d -- please re-encode",
file_format, GT_CONDENSEQ_VERSION);
had_err = -1;
}
if (!had_err)
had_err = gt_condenseq_io_one(condenseq->orig_num_seq);
if (!had_err)
had_err = gt_condenseq_io_one(condenseq->ldb_nelems);
if (!had_err) {
if (condenseq->ldb_nelems == 0) {
gt_warning("compression of condenseq did not succeed in finding any "
"compressable similarities, maybe the input is to small or "
"the chosen parameters should be reconsidered.");
}
if (condenseq->links == NULL) {
condenseq->links = gt_calloc((size_t) condenseq->ldb_nelems,
sizeof (*condenseq->links));
condenseq->ldb_allocated = condenseq->ldb_nelems;
}
had_err = gt_condenseq_io_one(condenseq->udb_nelems);
}
if (!had_err) {
gt_assert(condenseq->udb_nelems > 0);
if (condenseq->uniques == NULL) {
condenseq->uniques = gt_malloc(sizeof (*condenseq->uniques) *
condenseq->udb_nelems );
condenseq->udb_allocated = condenseq->udb_nelems;
}
}
for (idx = 0; !had_err && idx < condenseq->ldb_nelems; idx++) {
had_err = condenseq_linkentry_io(&condenseq->links[idx], fp, io_func, err);
}
for (idx = 0; !had_err && idx < condenseq->udb_nelems; idx++) {
had_err = condenseq_uniqueentry_io(&condenseq->uniques[idx], fp, io_func,
err);
}
if (!had_err && condenseq->orig_num_seq > (GtUword) 1) {
condenseq->ssptab = gt_intset_io(condenseq->ssptab, fp, err);
if (condenseq->ssptab == NULL)
had_err = 1;
}
if (!had_err)
had_err = gt_condenseq_io_one(condenseq->id_len);
if (!had_err) {
if (condenseq->id_len == GT_UNDEF_UWORD) {
condenseq->sdstab = gt_intset_io(condenseq->sdstab, fp, err);
if (condenseq->sdstab == NULL)
had_err = 1;
}
}
if (!had_err)
had_err = gt_condenseq_io_one(condenseq->ids_total_len);
if (!had_err) {
condenseq->orig_ids = gt_realloc(condenseq->orig_ids,
(size_t) condenseq->ids_total_len);
had_err = io_func(condenseq->orig_ids, sizeof (*condenseq->orig_ids),
(size_t) condenseq->ids_total_len, fp, err);
}
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;
}
static int inter_feature_in_children(GtFeatureNode *current_feature, void *data,
GT_UNUSED GtError *err)
{
GtInterFeatureVisitor *aiv = (GtInterFeatureVisitor*) data;
GtFeatureNode *inter_node;
GtRange previous_range, current_range, inter_range;
GtStrand previous_strand, /*current_strand, */inter_strand;
GtStr *parent_seqid;
gt_error_check(err);
gt_assert(current_feature);
if (gt_feature_node_has_type(current_feature, aiv->outside_type)) {
if (aiv->previous_feature) {
/* determine inter range */
previous_range = gt_genome_node_get_range((GtGenomeNode*)
aiv->previous_feature);
current_range = gt_genome_node_get_range((GtGenomeNode*) current_feature);
if (previous_range.end >= current_range.start) {
gt_warning("overlapping boundary features " GT_WU "-" GT_WU " and "
GT_WU "-" GT_WU ", " "not placing '%s' inter-feature",
previous_range.start,
previous_range.end,
current_range.start,
current_range.end,
aiv->inter_type);
return 0;
}
if (current_range.start - previous_range.end < 2) {
gt_warning("no space for inter-feature '%s' between " GT_WU " and "
GT_WU,
aiv->inter_type,
previous_range.end,
current_range.start);
return 0;
}
inter_range.start = previous_range.end + 1;
inter_range.end = current_range.start - 1;
/* determine inter strand */
previous_strand = gt_feature_node_get_strand(aiv->previous_feature);
/*current_strand = gt_feature_node_get_strand(current_feature);*/
gt_assert(previous_strand == gt_feature_node_get_strand(current_feature));
inter_strand = previous_strand;
/* determine sequence id */
parent_seqid =
gt_genome_node_get_seqid((GtGenomeNode*) aiv->parent_feature);
gt_assert(!gt_str_cmp(parent_seqid,
gt_genome_node_get_seqid((GtGenomeNode*)
aiv->previous_feature)));
gt_assert(!gt_str_cmp(parent_seqid,
gt_genome_node_get_seqid((GtGenomeNode*)
current_feature)));
/* create inter feature */
inter_node = (GtFeatureNode*)
gt_feature_node_new(parent_seqid, aiv->inter_type,
inter_range.start, inter_range.end,
inter_strand);
gt_feature_node_add_child(aiv->parent_feature, inter_node);
}
aiv->previous_feature = current_feature;
}
return 0;
}
static int parse_range(GtRange *range, const char *start, const char *end,
unsigned int line_number, const char *filename,
bool tidy, bool correct_negative, GtError *err)
{
long start_val, end_val;
char *ep;
gt_assert(start && end && filename);
gt_error_check(err);
range->start = GT_UNDEF_ULONG;
range->end = GT_UNDEF_ULONG;
/* parse and check start */
errno = 0;
start_val = strtol(start, &ep, 10);
if (start[0] == '\0' || *ep != '\0') {
gt_error_set(err, "could not parse number '%s' on line %u in file '%s'",
start, line_number, filename);
return -1;
}
if (errno == ERANGE && (start_val == LONG_MAX || start_val == LONG_MIN)) {
gt_error_set(err, "number '%s' out of range on line %u in file '%s'", start,
line_number, filename);
return -1;
}
if (start_val < 0) {
if (tidy || correct_negative) {
gt_warning("start '%s' is negative on line %u in file '%s'; reset to 1",
start, line_number, filename);
start_val = 1;
}
else {
gt_error_set(err, "start '%s' is negative on line %u in file '%s'", start,
line_number, filename);
return -1;
}
}
if (start_val == 0 && tidy) {
gt_warning("start '%s' is zero on line %u in file '%s' (GFF3 files are "
"1-based); reset to 1", start, line_number, filename);
start_val = 1;
}
/* parse and check end */
errno = 0;
end_val = strtol(end, &ep, 10);
if (end[0] == '\0' || *ep != '\0') {
gt_error_set(err, "could not parse number '%s' on line %u in file '%s'",
end, line_number, filename);
return -1;
}
if (errno == ERANGE && (end_val == LONG_MAX || end_val == LONG_MIN)) {
gt_error_set(err, "number '%s' out of range on line %u in file '%s'", end,
line_number, filename);
return -1;
}
if (end_val < 0) {
if (tidy || correct_negative) {
gt_warning("end '%s' is negative on line %u in file '%s'; reset to 1",
end, line_number, filename);
end_val = 1;
}
else {
gt_error_set(err, "end '%s' is negative on line %u in file '%s'", end,
line_number, filename);
return -1;
}
}
if (end_val == 0 && tidy) {
gt_warning("end '%s' is zero on line %u in file '%s' (GFF3 files are "
"1-based); reset to 1", end, line_number, filename);
end_val = 1;
}
/* check range */
if (start_val > end_val) {
if (tidy) {
long tmp_val;
gt_warning("start '%lu' is larger then end '%lu' on line %u in file "
"'%s'; swap them", start_val, end_val, line_number, filename);
tmp_val = start_val;
start_val = end_val;
end_val = tmp_val;
}
else {
gt_error_set(err, "start '%lu' is larger then end '%lu' on line %u in "
"file '%s'", start_val, end_val, line_number, filename);
return -1;
}
}
/* set result */
range->start = start_val;
range->end = end_val;
return 0;
}
static int gt_seqorder_runner(GT_UNUSED int argc, const char **argv,
int parsed_args, void *tool_arguments, GtError *err)
{
GtSeqorderArguments *arguments = tool_arguments;
int had_err = 0;
GtEncseq *encseq;
GtEncseqLoader *loader;
unsigned long i, nofseqs;
gt_error_check(err);
gt_assert(arguments != NULL);
/* load encseq */
loader = gt_encseq_loader_new();
encseq = gt_encseq_loader_load(loader, argv[parsed_args], err);
if (encseq == NULL)
had_err = -1;
if (had_err == 0 && !gt_encseq_has_description_support(encseq))
gt_warning("%s has no description support", argv[parsed_args]);
if (!had_err)
{
nofseqs = gt_encseq_num_of_sequences(encseq);
if (arguments->invert)
{
for (i = nofseqs; i > 0; i--)
gt_seqorder_output(i - 1, encseq);
}
else if (arguments->shuffle)
{
unsigned long *seqnums;
seqnums = gt_malloc(sizeof (unsigned long) * nofseqs);
gt_seqorder_get_shuffled_seqnums(nofseqs, seqnums);
for (i = 0; i < nofseqs; i++)
gt_seqorder_output(seqnums[i], encseq);
gt_free(seqnums);
}
else
{
GtSuffixsortspace *suffixsortspace;
gt_assert(arguments->sort || arguments->revsort);
suffixsortspace
= gt_suffixsortspace_new(nofseqs,
/* Use iterator over sequence separators:
saves a lot of binary searches */
gt_encseq_seqstartpos(encseq, nofseqs-1),
false,NULL);
gt_seqorder_sort(suffixsortspace, encseq);
if (arguments->sort)
for (i = 0; i < nofseqs; i++)
gt_seqorder_output(gt_encseq_seqnum(encseq,
gt_suffixsortspace_getdirect(suffixsortspace, i)), encseq);
else
for (i = nofseqs; i > 0; i--)
gt_seqorder_output(gt_encseq_seqnum(encseq,
gt_suffixsortspace_getdirect(suffixsortspace, i - 1)), encseq);
gt_suffixsortspace_delete(suffixsortspace, false);
}
}
gt_encseq_loader_delete(loader);
gt_encseq_delete(encseq);
return had_err;
}
static int calc_spliced_alignments(GthSACollection *sa_collection,
GthChainCollection *chain_collection,
GthCallInfo *call_info,
GthInput *input,
GthStat *stat,
GtUword gen_file_num,
GtUword ref_file_num,
bool directmatches,
GthMatchInfo *match_info,
GthDNACompletePathMatrixJT
dna_complete_path_matrix_jt,
GthProteinCompletePathMatrixJT
protein_complete_path_matrix_jt)
{
const unsigned char *ref_seq_tran, *ref_seq_orig, *ref_seq_tran_rc = NULL,
*ref_seq_orig_rc = NULL;
GtUword chainctr, gen_offset = GT_UNDEF_UWORD, gen_total_length,
ref_total_length;
GtFile *outfp = call_info->out->outfp;
GtRange gen_seq_bounds, gen_seq_bounds_rc;
bool refseqisdna;
GthChain *chain;
GtRange range;
GthSA *saA;
int rval;
gt_assert(sa_collection && chain_collection);
refseqisdna = gth_input_ref_file_is_dna(input, ref_file_num);
for (chainctr = 0;
chainctr < gth_chain_collection_size(chain_collection);
chainctr++) {
chain = gth_chain_collection_get(chain_collection, chainctr);
if (++match_info->call_number > call_info->firstalshown &&
call_info->firstalshown > 0) {
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "<!--\n");
if (!call_info->out->gff3out) {
gt_file_xprintf(outfp, "Maximal matching %s count (%u) reached.\n",
refseqisdna ? "EST" : "protein",
call_info->firstalshown);
gt_file_xprintf(outfp, "Only the first %u matches will be "
"displayed.\n", call_info->firstalshown);
}
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "-->\n");
match_info->max_call_number_reached = true;
break; /* break out of loop */
}
/* compute considered genomic regions if not set by -frompos */
if (!gth_input_use_substring_spec(input)) {
gen_seq_bounds = gth_input_get_genomic_range(input, chain->gen_file_num,
chain->gen_seq_num);
gen_total_length = gt_range_length(&gen_seq_bounds);
gen_offset = gen_seq_bounds.start;
gen_seq_bounds_rc = gen_seq_bounds;
}
else {
/* genomic multiseq contains exactly one sequence */
gt_assert(gth_input_num_of_gen_seqs(input, chain->gen_file_num) == 1);
gen_total_length = gth_input_genomic_file_total_length(input,
chain
->gen_file_num);
gen_seq_bounds.start = gth_input_genomic_substring_from(input);
gen_seq_bounds.end = gth_input_genomic_substring_to(input);
gen_offset = 0;
gen_seq_bounds_rc.start = gen_total_length - 1 - gen_seq_bounds.end;
gen_seq_bounds_rc.end = gen_total_length - 1 - gen_seq_bounds.start;
}
/* "retrieving" the reference sequence */
range = gth_input_get_reference_range(input, chain->ref_file_num,
chain->ref_seq_num);
ref_seq_tran = gth_input_current_ref_seq_tran(input) + range.start;
ref_seq_orig = gth_input_current_ref_seq_orig(input) + range.start;
if (refseqisdna) {
ref_seq_tran_rc = gth_input_current_ref_seq_tran_rc(input) + range.start;
ref_seq_orig_rc = gth_input_current_ref_seq_orig_rc(input) + range.start;
}
ref_total_length = range.end - range.start + 1;
/* check if protein sequences have a stop amino acid */
if (!refseqisdna && !match_info->stop_amino_acid_warning &&
ref_seq_orig[ref_total_length - 1] != GT_STOP_AMINO) {
GtStr *ref_id = gt_str_new();
gth_input_save_ref_id(input, ref_id, chain->ref_file_num,
chain->ref_seq_num);
gt_warning("protein sequence '%s' (#" GT_WU " in file %s) does not end "
"with a stop amino acid ('%c'). If it is not a protein "
"fragment you should add a stop amino acid to improve the "
"prediction. For example with `gt seqtransform "
"-addstopaminos` (see http://genometools.org for details).",
gt_str_get(ref_id), chain->ref_seq_num,
gth_input_get_reference_filename(input, chain->ref_file_num),
GT_STOP_AMINO);
match_info->stop_amino_acid_warning = true;
gt_str_delete(ref_id);
}
/* allocating space for alignment */
saA = gth_sa_new_and_set(directmatches, true, input, chain->gen_file_num,
chain->gen_seq_num, chain->ref_file_num,
chain->ref_seq_num, match_info->call_number,
gen_total_length, gen_offset, ref_total_length);
/* extend the DP borders to the left and to the right */
gth_chain_extend_borders(chain, &gen_seq_bounds, &gen_seq_bounds_rc,
gen_total_length, gen_offset);
/* From here on the dp positions always refer to the forward strand of the
genomic DNA. */
/* call the Dynamic Programming */
if (refseqisdna) {
rval = call_dna_DP(directmatches, call_info, input, stat,
sa_collection, saA, gen_file_num, ref_file_num,
gen_total_length, gen_offset, &gen_seq_bounds,
&gen_seq_bounds_rc, ref_total_length, range.start,
chainctr, gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig,
ref_seq_tran_rc, ref_seq_orig_rc, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
else {
rval = call_protein_DP(directmatches, call_info, input,
stat, sa_collection, saA, gen_file_num,
ref_file_num, gen_total_length, gen_offset,
&gen_seq_bounds, &gen_seq_bounds_rc,
ref_total_length, range.start, chainctr,
gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
/* check return value */
if (rval == GTH_ERROR_DP_PARAMETER_ALLOCATION_FAILED) {
/* statistics bookkeeping */
gth_stat_increment_numoffailedDPparameterallocations(stat);
gth_stat_increment_numofundeterminedSAs(stat);
/* free space */
gth_sa_delete(saA);
match_info->call_number--;
continue; /* continue with the next DP range */
}
else if (rval)
return -1;
}
if (!call_info->out->xmlout && !call_info->out->gff3out && !directmatches &&
!match_info->significant_match_found &&
match_info->call_number <= call_info->firstalshown) {
show_no_match_line(gth_input_get_alphatype(input, ref_file_num), outfp);
}
return 0;
}
static int check_cds_phases(GtArray *cds_features, GtCDSCheckVisitor *v,
bool is_multi, bool second_pass, GtError *err)
{
GtPhase current_phase, correct_phase = GT_PHASE_ZERO;
GtFeatureNode *fn;
GtStrand strand;
unsigned long i, current_length;
int had_err = 0;
gt_error_check(err);
gt_assert(cds_features);
gt_assert(gt_array_size(cds_features));
fn = *(GtFeatureNode**) gt_array_get_first(cds_features);
strand = gt_feature_node_get_strand(fn);
if (strand == GT_STRAND_REVERSE)
gt_array_reverse(cds_features);
for (i = 0; !had_err && i < gt_array_size(cds_features); i++) {
fn = *(GtFeatureNode**) gt_array_get(cds_features, i);
/* the first phase can be anything (except being undefined), because the
GFF3 spec says:
NOTE 4 - CDS features MUST have have a defined phase field. Otherwise it
is not possible to infer the correct polypeptides corresponding to
partially annotated genes. */
if ((!i && gt_feature_node_get_phase(fn) == GT_PHASE_UNDEFINED) ||
(i && gt_feature_node_get_phase(fn) != correct_phase)) {
if (gt_hashmap_get(v->cds_features, fn)) {
if (v->tidy && !is_multi && !gt_feature_node_has_children(fn)) {
/* we can split the feature */
gt_warning("%s feature on line %u in file \"%s\" has multiple "
"parents which require different phases; split feature",
gt_ft_CDS,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn));
gt_hashmap_add(v->cds_features_to_split, fn, fn);
v->splitting_is_necessary = true; /* split later */
}
else {
gt_error_set(err, "%s feature on line %u in file \"%s\" has multiple "
"parents which require different phases",
gt_ft_CDS,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn));
had_err = -1;
}
}
else {
if (v->tidy) {
if (!second_pass) {
gt_warning("%s feature on line %u in file \"%s\" has the wrong "
"phase %c -> correcting it to %c", gt_ft_CDS,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn),
GT_PHASE_CHARS[gt_feature_node_get_phase(fn)],
GT_PHASE_CHARS[correct_phase]);
}
gt_feature_node_set_phase(fn, correct_phase);
}
else {
gt_error_set(err, "%s feature on line %u in file \"%s\" has the "
"wrong phase %c (should be %c)", gt_ft_CDS,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn),
GT_PHASE_CHARS[gt_feature_node_get_phase(fn)],
GT_PHASE_CHARS[correct_phase]);
had_err = -1;
}
}
}
if (!had_err) {
current_phase = gt_feature_node_get_phase(fn);
current_length = gt_genome_node_get_length((GtGenomeNode*) fn);
correct_phase = (3 - (current_length - current_phase) % 3) % 3;
gt_hashmap_add(v->cds_features, fn, fn); /* record CDS feature */
}
}
return had_err;
}
static int gt_seqtranslate_runner(int argc, const char **argv, int parsed_args,
void *tool_arguments, GT_UNUSED GtError *err)
{
GtTranslateArguments *arguments = tool_arguments;
GtSeqIterator *si = NULL;
GtSequenceBuffer *sb = NULL;
GtStrArray *infiles;
int had_err = 0,
rval,
i;
GtStr *translations[3];
translations[0] = gt_str_new();
translations[1] = gt_str_new();
translations[2] = gt_str_new();
gt_error_check(err);
gt_assert(arguments);
infiles = gt_str_array_new();
for (i = parsed_args; i < argc; i++) {
gt_str_array_add_cstr(infiles, argv[i]);
}
sb = gt_sequence_buffer_new_guess_type(infiles, err);
if (!sb)
had_err = -1;
if (!had_err) {
si = gt_seq_iterator_sequence_buffer_new_with_buffer(sb);
if (!si)
had_err = -1;
}
if (!had_err) {
char *desc;
const GtUchar *sequence;
GtUword len;
while (!had_err && (rval = gt_seq_iterator_next(si,
&sequence,
&len, &desc, err))) {
if (rval < 0) {
had_err = -1;
break;
}
if (len < GT_CODON_LENGTH) {
gt_warning("sequence '%s' is shorter than codon length of %d, skipping",
desc, GT_CODON_LENGTH);
} else {
had_err = gt_seqtranslate_do_translation(arguments, (char*) sequence,
len, desc,
translations, false, err);
if (!had_err && arguments->reverse) {
char *revseq = gt_cstr_dup_nt((char*) sequence, len);
had_err = gt_reverse_complement(revseq, len, err);
if (!had_err) {
had_err = gt_seqtranslate_do_translation(arguments, revseq, len,
desc, translations, true,
err);
}
gt_free(revseq);
}
}
}
}
gt_str_delete(translations[0]);
gt_str_delete(translations[1]);
gt_str_delete(translations[2]);
gt_str_array_delete(infiles);
gt_seq_iterator_delete(si);
gt_sequence_buffer_delete(sb);
return had_err;
}
static int output_sequence(GtEncseq *encseq, GtEncseqDecodeArguments *args,
const char *filename, GtError *err)
{
GtUword i, j, sfrom, sto;
int had_err = 0;
bool has_desc;
GtEncseqReader *esr;
gt_assert(encseq);
if (!(has_desc = gt_encseq_has_description_support(encseq)))
gt_warning("Missing description support for file %s", filename);
if (strcmp(gt_str_get(args->mode), "fasta") == 0) {
/* specify a single sequence to extract */
if (args->seq != GT_UNDEF_UWORD) {
if (args->seq >= gt_encseq_num_of_sequences(encseq)) {
gt_error_set(err,
"requested sequence "GT_WU" exceeds number of sequences "
"("GT_WU")", args->seq,
gt_encseq_num_of_sequences(encseq));
return -1;
}
sfrom = args->seq;
sto = args->seq + 1;
} else if (args->seqrng.start != GT_UNDEF_UWORD
&& args->seqrng.end != GT_UNDEF_UWORD) {
/* specify a sequence range to extract */
if (args->seqrng.start >= gt_encseq_num_of_sequences(encseq)
|| args->seqrng.end >= gt_encseq_num_of_sequences(encseq)) {
gt_error_set(err,
"range "GT_WU"-"GT_WU" includes a sequence number "
"exceeding the total number of sequences ("GT_WU")",
args->seqrng.start,
args->seqrng.end,
gt_encseq_num_of_sequences(encseq));
return -1;
}
sfrom = args->seqrng.start;
sto = args->seqrng.end + 1;
} else {
/* extract all sequences */
sfrom = 0;
sto = gt_encseq_num_of_sequences(encseq);
}
for (i = sfrom; i < sto; i++) {
GtUword desclen, startpos, len;
char buf[BUFSIZ];
const char *desc = NULL;
/* XXX: maybe make this distinction in the functions via readmode? */
if (!GT_ISDIRREVERSE(args->rm)) {
startpos = gt_encseq_seqstartpos(encseq, i);
len = gt_encseq_seqlength(encseq, i);
if (has_desc) {
desc = gt_encseq_description(encseq, &desclen, i);
} else {
(void) snprintf(buf, BUFSIZ, "sequence "GT_WU"", i);
desclen = strlen(buf);
desc = buf;
}
} else {
startpos = gt_encseq_seqstartpos(encseq, i);
len = gt_encseq_seqlength(encseq,
gt_encseq_num_of_sequences(encseq)-1-i);
startpos = gt_encseq_total_length(encseq)
- (gt_encseq_seqstartpos(encseq,
gt_encseq_num_of_sequences(
encseq)-1-i) + len);
if (has_desc) {
desc = gt_encseq_description(encseq,
&desclen,
gt_encseq_num_of_sequences(encseq)-1-i);
} else {
(void) snprintf(buf, BUFSIZ, "sequence "GT_WU"", i);
desclen = strlen(buf);
desc = buf;
}
}
gt_assert(desc);
/* output description */
gt_xfputc(GT_FASTA_SEPARATOR, stdout);
gt_xfwrite(desc, 1, desclen, stdout);
gt_xfputc('\n', stdout);
/* XXX: make this more efficient by writing in a buffer first and then
showing the result */
if (args->singlechars) {
for (j = 0; j < len; j++) {
gt_xfputc(gt_encseq_get_decoded_char(encseq,
startpos + j,
args->rm),
stdout);
}
} else {
esr = gt_encseq_create_reader_with_readmode(encseq, args->rm, startpos);
for (j = 0; j < len; j++) {
gt_xfputc(gt_encseq_reader_next_decoded_char(esr), stdout);
}
gt_encseq_reader_delete(esr);
}
gt_xfputc('\n', stdout);
}
}
if (strcmp(gt_str_get(args->mode), "concat") == 0) {
GtUword from = 0,
to = gt_encseq_total_length(encseq) - 1;
if (args->rng.start != GT_UNDEF_UWORD && args->rng.end != GT_UNDEF_UWORD) {
if (args->rng.end > to) {
had_err = -1;
gt_error_set(err,
"end of range ("GT_WU") exceeds encoded sequence length "
"("GT_WU")", args->rng.end, to);
}
if (!had_err) {
from = args->rng.start;
to = args->rng.end;
}
}
if (!had_err) {
if (args->singlechars) {
for (j = from; j <= to; j++) {
char cc = gt_encseq_get_decoded_char(encseq, j, args->rm);
if (cc == (char) SEPARATOR)
cc = gt_str_get(args->sepchar)[0];
gt_xfputc(cc, stdout);
}
} else {
esr = gt_encseq_create_reader_with_readmode(encseq, args->rm, from);
if (esr) {
for (j = from; j <= to; j++) {
char cc = gt_encseq_reader_next_decoded_char(esr);
if (cc == (char) SEPARATOR)
cc = gt_str_get(args->sepchar)[0];
gt_xfputc(cc, stdout);
}
gt_encseq_reader_delete(esr);
}
}
gt_xfputc('\n', stdout);
}
}
return had_err;
}
static int gt_sketch_page_runner(GT_UNUSED int argc,
const char **argv,
int parsed_args,
void *tool_arguments,
GtError *err)
{
SketchPageArguments *arguments = tool_arguments;
int had_err = 0;
GtFeatureIndex *features = NULL;
GtRange qry_range, sequence_region_range;
GtStyle *sty = NULL;
GtStr *prog, *gt_style_file;
GtDiagram *d = NULL;
GtLayout *l = NULL;
GtBioseq *bioseq = NULL;
GtCanvas *canvas = NULL;
const char *seqid = NULL, *outfile;
unsigned long start, height, num_pages = 0;
double offsetpos, usable_height;
cairo_surface_t *surf = NULL;
cairo_t *cr = NULL;
GtTextWidthCalculator *twc;
gt_error_check(err);
features = gt_feature_index_memory_new();
if (cairo_version() < CAIRO_VERSION_ENCODE(1, 8, 6))
gt_warning("Your cairo library (version %s) is older than version 1.8.6! "
"These versions contain a bug which may result in "
"corrupted PDF output!", cairo_version_string());
/* get style */
sty = gt_style_new(err);
if (gt_str_length(arguments->stylefile) == 0)
{
prog = gt_str_new();
gt_str_append_cstr_nt(prog, argv[0],
gt_cstr_length_up_to_char(argv[0], ' '));
gt_style_file = gt_get_gtdata_path(gt_str_get(prog), err);
gt_str_delete(prog);
gt_str_append_cstr(gt_style_file, "/sketch/default.style");
}
else
{
gt_style_file = gt_str_ref(arguments->stylefile);
}
had_err = gt_style_load_file(sty, gt_str_get(gt_style_file), err);
outfile = argv[parsed_args];
if (!had_err)
{
/* get features */
had_err = gt_feature_index_add_gff3file(features, argv[parsed_args+1], err);
if (!had_err && gt_str_length(arguments->seqid) == 0) {
seqid = gt_feature_index_get_first_seqid(features);
if (seqid == NULL)
{
gt_error_set(err, "GFF input file must contain a sequence region!");
had_err = -1;
}
}
else if (!had_err
&& !gt_feature_index_has_seqid(features,
gt_str_get(arguments->seqid)))
{
gt_error_set(err, "sequence region '%s' does not exist in GFF input file",
gt_str_get(arguments->seqid));
had_err = -1;
}
else if (!had_err)
seqid = gt_str_get(arguments->seqid);
}
/* set text */
if (gt_str_length(arguments->text) == 0)
{
gt_str_delete(arguments->text);
arguments->text = gt_str_new_cstr(argv[parsed_args+1]);
}
if (!had_err)
{
/* set display range */
gt_feature_index_get_range_for_seqid(features, &sequence_region_range,
seqid);
qry_range.start = (arguments->range.start == GT_UNDEF_ULONG ?
sequence_region_range.start :
arguments->range.start);
qry_range.end = (arguments->range.end == GT_UNDEF_ULONG ?
sequence_region_range.end :
arguments->range.end);
/* set output format */
if (strcmp(gt_str_get(arguments->format), "pdf") == 0)
{
surf = cairo_pdf_surface_create(outfile,
mm_to_pt(arguments->pwidth),
mm_to_pt(arguments->pheight));
}
else if (strcmp(gt_str_get(arguments->format), "ps") == 0)
{
surf = cairo_ps_surface_create(outfile,
mm_to_pt(arguments->pwidth),
mm_to_pt(arguments->pheight));
}
gt_log_log("created page with %.2f:%.2f dimensions\n",
mm_to_pt(arguments->pwidth),
mm_to_pt(arguments->pheight));
offsetpos = TEXT_SPACER + arguments->theight + TEXT_SPACER;
usable_height = mm_to_pt(arguments->pheight)
- arguments->theight
- arguments->theight
- 4*TEXT_SPACER;
if (gt_str_length(arguments->seqfile) > 0) {
bioseq = gt_bioseq_new(gt_str_get(arguments->seqfile), err);
}
cr = cairo_create(surf);
cairo_set_font_size(cr, 8);
twc = gt_text_width_calculator_cairo_new(cr, sty);
for (start = qry_range.start; start <= qry_range.end;
start += arguments->width)
{
GtRange single_range;
GtCustomTrack *ct = NULL;
const char *seq;
single_range.start = start;
single_range.end = start + arguments->width;
if (had_err)
break;
d = gt_diagram_new(features, seqid, &single_range, sty, err);
if (!d) {
had_err = -1;
break;
}
if (bioseq) {
seq = gt_bioseq_get_sequence(bioseq, 0);
ct = gt_custom_track_gc_content_new(seq,
gt_bioseq_get_sequence_length(bioseq, 0),
800, 70, 0.4, true);
gt_diagram_add_custom_track(d, ct);
}
l = gt_layout_new_with_twc(d, mm_to_pt(arguments->width), sty, twc, err);
had_err = gt_layout_get_height(l, &height, err);
if (!had_err) {
if (gt_double_smaller_double(usable_height - 10 - 2*TEXT_SPACER
- arguments->theight, offsetpos + height))
{
draw_header(cr, gt_str_get(arguments->text), argv[parsed_args+1],
seqid, num_pages, mm_to_pt(arguments->pwidth),
mm_to_pt(arguments->pheight),
arguments->theight);
cairo_show_page(cr);
offsetpos = TEXT_SPACER + arguments->theight + TEXT_SPACER;
num_pages++;
}
canvas = gt_canvas_cairo_context_new(sty,
cr,
offsetpos,
mm_to_pt(arguments->pwidth),
height,
NULL,
err);
if (!canvas)
had_err = -1;
offsetpos += height;
if (!had_err)
had_err = gt_layout_sketch(l, canvas, err);
}
gt_canvas_delete(canvas);
gt_layout_delete(l);
gt_diagram_delete(d);
if (ct)
gt_custom_track_delete(ct);
}
draw_header(cr, gt_str_get(arguments->text), argv[parsed_args+1], seqid,
num_pages, mm_to_pt(arguments->pwidth),
mm_to_pt(arguments->pheight),
arguments->theight);
cairo_show_page(cr);
num_pages++;
gt_log_log("finished, should be %lu pages\n", num_pages);
gt_text_width_calculator_delete(twc);
cairo_destroy(cr);
cairo_surface_flush(surf);
cairo_surface_finish(surf);
cairo_surface_destroy(surf);
cairo_debug_reset_static_data();
if (bioseq)
gt_bioseq_delete(bioseq);
gt_style_delete(sty);
gt_str_delete(gt_style_file);
gt_feature_index_delete(features);
}
return had_err;
}