static GtStr* region_mapping_map(GtRegionMapping *rm,
const char *sequence_region, GtError *err)
{
gt_error_check(err);
gt_assert(rm && sequence_region);
if (!rm->mapping)
return gt_str_ref(gt_str_array_get_str(rm->sequence_filenames, 0));
else
return gt_mapping_map_string(rm->mapping, sequence_region, err);
}
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 GtOPrval parse_options(int *parsed_args,
Cmppairwiseopt *pw,
int argc, const char **argv, GtError *err)
{
GtOptionParser *op;
GtOption *optionstrings,
*optionfiles,
*optioncharlistlen,
*optiontext,
*optionshowedist;
GtStrArray *charlistlen;
GtOPrval oprval;
gt_error_check(err);
charlistlen = gt_str_array_new();
pw->strings = gt_str_array_new();
pw->files = gt_str_array_new();
pw->text = gt_str_new();
pw->charlistlen = NULL;
pw->showedist = false;
op = gt_option_parser_new("options", "Apply function to pairs of strings.");
gt_option_parser_set_mail_address(op, "<*****@*****.**>");
optionstrings = gt_option_new_string_array("ss", "use two strings",
pw->strings);
gt_option_parser_add_option(op, optionstrings);
optionfiles = gt_option_new_filename_array("ff", "use two files",
pw->files);
gt_option_parser_add_option(op, optionfiles);
optioncharlistlen = gt_option_new_string_array("a",
"use character list and length",
charlistlen);
gt_option_parser_add_option(op, optioncharlistlen);
optiontext = gt_option_new_string("t", "use text", pw->text, NULL);
gt_option_parser_add_option(op, optiontext);
optionshowedist = gt_option_new_bool("e", "output unit edit distance",
&pw->showedist, false);
gt_option_parser_add_option(op, optionshowedist);
gt_option_exclude(optionstrings, optionfiles);
gt_option_exclude(optionstrings, optioncharlistlen);
gt_option_exclude(optionstrings, optiontext);
gt_option_exclude(optionfiles, optioncharlistlen);
gt_option_exclude(optionfiles, optiontext);
gt_option_exclude(optioncharlistlen, optiontext);
gt_option_imply(optionshowedist, optionstrings);
oprval = gt_option_parser_parse(op, parsed_args, argc, argv, gt_versionfunc,
err);
if (oprval == GT_OPTION_PARSER_OK)
{
if (gt_option_is_set(optionstrings))
{
if (gt_str_array_size(pw->strings) != 2UL)
{
gt_error_set(err, "option -ss requires two string arguments");
oprval = GT_OPTION_PARSER_ERROR;
}
} else
{
if (gt_option_is_set(optionfiles))
{
if (gt_str_array_size(pw->files) != 2UL)
{
gt_error_set(err, "option -ff requires two filename arguments");
oprval = GT_OPTION_PARSER_ERROR;
}
} else
{
if (gt_option_is_set(optioncharlistlen))
{
GtWord readint;
if (gt_str_array_size(charlistlen) != 2UL)
{
gt_error_set(err,
"option -a requires charlist and length argument");
oprval = GT_OPTION_PARSER_ERROR;
}
pw->charlistlen = gt_malloc(sizeof *pw->charlistlen);
pw->charlistlen->charlist =
gt_str_ref(gt_str_array_get_str(charlistlen,
0));
if (sscanf(gt_str_array_get(charlistlen,1UL), GT_WD, &readint) != 1 ||
readint < 1L)
{
gt_error_set(err,
"option -a requires charlist and length argument");
oprval = GT_OPTION_PARSER_ERROR;
}
pw->charlistlen->len = (GtUword) readint;
} else
{
if (!gt_option_is_set(optiontext))
{
gt_error_set(err,
"use exactly one of the options -ss, -ff, -a, -t");
oprval = GT_OPTION_PARSER_ERROR;
}
}
}
}
}
gt_option_parser_delete(op);
if (oprval == GT_OPTION_PARSER_OK && *parsed_args != argc)
{
gt_error_set(err, "superfluous program parameters");
oprval = GT_OPTION_PARSER_ERROR;
}
gt_str_array_delete(charlistlen);
return oprval;
}
int gt_paircmp(int argc, const char **argv, GtError *err)
{
int parsed_args;
Cmppairwiseopt cmppairwise;
GtOPrval oprval;
gt_error_check(err);
oprval = parse_options(&parsed_args, &cmppairwise, argc, argv, err);
if (oprval == GT_OPTION_PARSER_OK)
{
gt_assert(parsed_args == argc);
showsimpleoptions(&cmppairwise);
if (cmppairwise.showedist)
{
GtUword edist, len1, len2;
GtStr *s1, *s2;
gt_assert(gt_str_array_size(cmppairwise.strings) >= 2);
s1 = gt_str_array_get_str(cmppairwise.strings,0);
s2 = gt_str_array_get_str(cmppairwise.strings,1UL);
len1 = gt_str_length(s1);
len2 = gt_str_length(s2);
edist = gt_computegreedyunitedist((const GtUchar *) gt_str_get(s1),
len1,
(const GtUchar *) gt_str_get(s2),
len2);
printf(GT_WU " " GT_WU " " GT_WU " " GT_WU "%% errors\n",
edist, len1,len2,(200 * edist)/(len1+len2));
}
else if (cmppairwise.print)
{
gt_print_edist_alignment(
(const GtUchar *) gt_str_array_get(cmppairwise.strings,0),0,
(GtUword) strlen(gt_str_array_get(cmppairwise.strings,0)),
(const GtUchar *) gt_str_array_get(cmppairwise.strings,1UL),0,
(GtUword) strlen(gt_str_array_get(cmppairwise.strings,1UL)));
} else
{
size_t idx;
Checkfunctiontabentry checkfunction_tab[] = {
MAKECheckfunctiontabentry(gt_checkgreedyunitedist),
MAKECheckfunctiontabentry(gt_checklinearspace),
MAKECheckfunctiontabentry(gt_checklinearspace_local),
MAKECheckfunctiontabentry(gt_checkaffinelinearspace),
MAKECheckfunctiontabentry(gt_checkaffinelinearspace_local)
};
for (idx = 0; idx < sizeof checkfunction_tab/sizeof checkfunction_tab[0];
idx++)
{
GtUword testcases
= applycheckfunctiontosimpleoptions(checkfunction_tab[idx].function,
&cmppairwise);
printf("# number of testcases for %s: " GT_WU "\n",
checkfunction_tab[idx].name,testcases);
}
}
}
freesimpleoption(&cmppairwise);
if (oprval == GT_OPTION_PARSER_REQUESTS_EXIT)
{
return 0;
}
if (oprval == GT_OPTION_PARSER_ERROR)
{
return -1;
}
return 0;
}
GtHcrEncoder *gt_hcr_encoder_new(GtStrArray *files, GtAlphabet *alpha,
bool descs, GtQualRange qrange, GtTimer *timer,
GtError *err)
{
GtBaseQualDistr *bqd;
GtHcrEncoder *hcr_enc;
GtSeqIterator *seqit;
GtStrArray *file;
int had_err = 0,
status;
GtUword len1,
len2,
i,
num_of_reads = 0;
const GtUchar *seq,
*qual;
char *desc;
gt_error_check(err);
gt_assert(alpha && files);
if (timer != NULL)
gt_timer_show_progress(timer, "get <base,qual> distr", stdout);
if (qrange.start != GT_UNDEF_UINT)
if (qrange.start == qrange.end) {
gt_error_set(err, "qrange.start must unequal qrange.end");
return NULL;
}
hcr_enc = gt_malloc(sizeof (GtHcrEncoder));
hcr_enc->files = files;
hcr_enc->num_of_files = gt_str_array_size(files);
hcr_enc->num_of_reads = 0;
hcr_enc->page_sampling = false;
hcr_enc->regular_sampling = false;
hcr_enc->sampling_rate = 0;
hcr_enc->pagesize = gt_pagesize();
if (descs) {
hcr_enc->encdesc_encoder = gt_encdesc_encoder_new();
if (timer != NULL)
gt_encdesc_encoder_set_timer(hcr_enc->encdesc_encoder, timer);
}
else
hcr_enc->encdesc_encoder = NULL;
hcr_enc->seq_encoder = gt_malloc(sizeof (GtHcrSeqEncoder));
hcr_enc->seq_encoder->alpha = alpha;
hcr_enc->seq_encoder->sampling = NULL;
hcr_enc->seq_encoder->fileinfos = gt_calloc((size_t) hcr_enc->num_of_files,
sizeof (*(hcr_enc->seq_encoder->fileinfos)));
hcr_enc->seq_encoder->qrange = qrange;
bqd = hcr_base_qual_distr_new(alpha, qrange);
/* check if reads in the same file are of same length and get
<base, quality> pair distribution */
for (i = 0; i < hcr_enc->num_of_files; i++) {
file = gt_str_array_new();
gt_str_array_add(file, gt_str_array_get_str(files, i));
seqit = gt_seq_iterator_fastq_new(file, err);
if (!seqit) {
gt_error_set(err, "cannot initialize GtSeqIteratorFastQ object");
had_err = -1;
}
if (!had_err) {
gt_seq_iterator_set_symbolmap(seqit, gt_alphabet_symbolmap(alpha));
gt_seq_iterator_set_quality_buffer(seqit, &qual);
status = gt_seq_iterator_next(seqit, &seq, &len1, &desc, err);
if (status == 1) {
num_of_reads = 1UL;
while (!had_err) {
status = gt_seq_iterator_next(seqit, &seq, &len2, &desc, err);
if (status == -1)
had_err = -1;
if (status != 1)
break;
if (len2 != len1) {
gt_error_set(err, "reads have to be of equal length");
had_err = -1;
break;
}
if (hcr_base_qual_distr_add(bqd, qual, seq, len1) != 0)
had_err = -1;
len1 = len2;
num_of_reads++;
}
}
else if (status == -1)
had_err = -1;
if (!had_err) {
if (i == 0)
hcr_enc->seq_encoder->fileinfos[i].readnum = num_of_reads;
else
hcr_enc->seq_encoder->fileinfos[i].readnum =
hcr_enc->seq_encoder->fileinfos[i - 1].readnum + num_of_reads;
hcr_enc->seq_encoder->fileinfos[i].readlength = len1;
}
}
hcr_enc->num_of_reads += num_of_reads;
gt_str_array_delete(file);
gt_seq_iterator_delete(seqit);
}
if (!had_err)
hcr_base_qual_distr_trim(bqd);
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "build huffman tree for sequences and"
" qualities", stdout);
hcr_enc->seq_encoder->huffman =
gt_huffman_new(bqd,
hcr_base_qual_distr_func,
(GtUword) bqd->ncols * bqd->nrows);
}
if (!had_err) {
hcr_enc->seq_encoder->qual_offset = bqd->qual_offset;
hcr_base_qual_distr_delete(bqd);
return hcr_enc;
}
return NULL;
}
static int gt_genomediff_runner(int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
bool mirrored = false;
int had_err = 0,
i;
GtEncseq *encseq = NULL;
GtGenomediffArguments *arguments = tool_arguments;
GtLogger *logger;
GtShuUnitFileInfo *unit_info = NULL;
GtTimer *timer = NULL;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose,
GT_LOGGER_DEFLT_PREFIX,
stdout);
gt_assert(logger);
for (i = parsed_args; i < argc; i++) {
gt_str_array_add_cstr(arguments->filenames, argv[i]);
}
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("start");
gt_timer_start(timer);
gt_assert(timer);
}
if (arguments->with_units) {
gt_logger_log(logger, "unitfile option set, filename is %s\n",
gt_str_get(arguments->unitfile));
}
if (timer != NULL)
gt_timer_show_progress(timer, "start shu search", stdout);
if (gt_str_array_size(arguments->filenames) > 1UL) {
GtEncseqEncoder *ee = gt_encseq_encoder_new();
gt_encseq_encoder_set_timer(ee, timer);
gt_encseq_encoder_set_logger(ee, logger);
/* kr only makes sense for dna, so we can check this already with ee */
gt_encseq_encoder_set_input_dna(ee);
had_err = gt_encseq_encoder_encode(ee, arguments->filenames,
gt_str_get(arguments->indexname), err);
gt_encseq_encoder_delete(ee);
}
else {
gt_str_append_str(arguments->indexname,
gt_str_array_get_str(arguments->filenames, 0));
if (arguments->with_esa || arguments->with_pck) {
GtStr *current_line = gt_str_new();
FILE *prj_fp;
const char *buffer;
char **elements = NULL;
prj_fp = gt_fa_fopen_with_suffix(gt_str_get(arguments->indexname),
GT_PROJECTFILESUFFIX,"rb",err);
if (prj_fp == NULL)
had_err = -1;
while (!had_err && gt_str_read_next_line(current_line, prj_fp) != EOF) {
buffer = gt_str_get(current_line);
if (elements != NULL) {
gt_free(elements[0]);
gt_free(elements[1]);
}
gt_free(elements);
elements = gt_cstr_split(buffer, '=');
gt_log_log("%s", elements[0]);
if (strcmp("mirrored", elements[0]) == 0) {
gt_log_log("%s", elements[1]);
if (strcmp("1", elements[1]) == 0) {
mirrored = true;
gt_log_log("sequences are treated as mirrored");
}
}
gt_str_reset(current_line);
}
gt_str_delete(current_line);
if (elements != NULL) {
gt_free(elements[0]);
gt_free(elements[1]);
}
gt_free(elements);
gt_fa_xfclose(prj_fp);
}
}
if (!had_err) {
GtEncseqLoader *el = gt_encseq_loader_new_from_options(arguments->loadopts,
err);
if (mirrored)
gt_encseq_loader_mirror(el);
encseq =
gt_encseq_loader_load(el, gt_str_get(arguments->indexname), err);
gt_encseq_loader_delete(el);
}
if (encseq == NULL)
had_err = -1;
if (!had_err) {
unit_info = gt_shu_unit_info_new(encseq);
if (arguments->with_units)
had_err = gt_shu_unit_file_info_read(arguments->unitfile, unit_info,
logger, err);
}
if (!had_err) {
uint64_t **shusums = NULL;
if (arguments->with_esa || arguments->with_pck) {
shusums = gt_genomediff_shulen_sum(arguments, unit_info,
logger, timer, err);
if (shusums == NULL)
had_err = -1;
}
else {
const bool doesa = true;
GenomediffInfo gd_info;
Suffixeratoroptions sopts;
sopts.beverbose = arguments->verbose;
sopts.indexname = arguments->indexname;
sopts.db = NULL;
sopts.encopts = NULL;
sopts.genomediff = true;
sopts.inputindex = arguments->indexname;
sopts.loadopts = arguments->loadopts;
sopts.showprogress = false;
sopts.idxopts = arguments->idxopts;
gt_assert(unit_info != NULL);
gt_array2dim_calloc(shusums, unit_info->num_of_genomes,
unit_info->num_of_genomes);
gd_info.shulensums = shusums;
gd_info.unit_info = unit_info;
had_err = gt_runsuffixerator(doesa, &sopts, &gd_info, logger, err);
}
if (!had_err && shusums != NULL) {
had_err = gt_genomediff_kr_calc(shusums, arguments, unit_info,
arguments->with_pck, logger, timer, err);
gt_array2dim_delete(shusums);
}
}
if (timer != NULL) {
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
gt_logger_delete(logger);
gt_encseq_delete(encseq);
gt_shu_unit_info_delete(unit_info);
return had_err;
}
static int gt_linspace_align_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtLinspaceArguments *arguments = tool_arguments;
int had_err = 0;
GtAlignment *align;
GtWord left_dist = 0, right_dist = 0;
GtSequenceTable *sequence_table1, *sequence_table2;
GtLinspaceManagement *spacemanager;
GtScoreHandler *scorehandler = NULL;
GtTimer *linspacetimer = NULL;
GtAlphabet *alphabet = NULL;
gt_error_check(err);
gt_assert(arguments);
sequence_table1 = gt_sequence_table_new();
sequence_table2 = gt_sequence_table_new();
align = gt_alignment_new();
spacemanager = gt_linspace_management_new();
gt_linspace_management_set_TSfactor(spacemanager,arguments->timesquarefactor);
/* get sequences */
if (gt_str_array_size(arguments->strings) > 0)
{
get_onesequence(sequence_table1, arguments->strings, 0);
sequence_table1->size++;
get_onesequence(sequence_table2, arguments->strings, 1);
sequence_table2->size++;
}
else if (gt_str_array_size(arguments->files) > 0)
{
had_err = get_fastasequences(sequence_table1,
gt_str_array_get_str(arguments->files,0),err);
if (!had_err)
{
had_err = get_fastasequences(sequence_table2,
gt_str_array_get_str(arguments->files,1),err);
}
}
if (arguments->dna)
{
alphabet = gt_alphabet_new_dna();
} else
{
gt_assert(arguments->protein);
alphabet = gt_alphabet_new_protein();
}
gt_encode_sequence_table(alphabet,sequence_table1);
gt_encode_sequence_table(alphabet,sequence_table2);
if (!had_err)
{
scorehandler = gt_arguments2scorehandler(arguments,err);
if (scorehandler == NULL)
{
had_err = -1;
} else
{
if (arguments->global && arguments->protein && !arguments->has_costmatrix)
{
GtScoreHandler *costhandler = gt_scorehandler2costhandler(scorehandler);
gt_scorehandler_delete(scorehandler);
scorehandler = costhandler;
}
}
}
/* get diagonal band */
if (!had_err && arguments->diagonal)
{
if (gt_str_array_size(arguments->diagonalbonds) > 0)
{
had_err = gt_parse_score_value(__LINE__,&left_dist,
gt_str_array_get(arguments->diagonalbonds,0),
false, err);
if (!had_err)
{
had_err = gt_parse_score_value(__LINE__,&right_dist,
gt_str_array_get(arguments->diagonalbonds,1),
false, err);
}
}
}
if (!had_err && arguments->spacetime)
{
linspacetimer = gt_timer_new();
}
/* alignment functions with linear gap costs */
if (!had_err)
{
bool affine;
if (gt_str_array_size(arguments->linearcosts) > 0)
{
affine = false;
} else
{
gt_assert(gt_str_array_size(arguments->affinecosts) > 0);
affine = true;
}
had_err = gt_all_against_all_alignment_check (
affine, align, arguments,
spacemanager,
scorehandler,
gt_alphabet_characters(alphabet),
gt_alphabet_wildcard_show(alphabet),
sequence_table1,
sequence_table2,
left_dist,
right_dist,
linspacetimer,err);
}
/*spacetime option*/
if (!had_err && arguments->spacetime)
{
printf("# combined space peak in kilobytes: %f\n",
GT_KILOBYTES(gt_linspace_management_get_spacepeak(spacemanager)));
gt_timer_show_formatted(linspacetimer,"# TIME overall " GT_WD ".%02ld\n",
stdout);
}
gt_timer_delete(linspacetimer);
gt_linspace_management_delete(spacemanager);
gt_sequence_table_delete(sequence_table1);
gt_sequence_table_delete(sequence_table2);
gt_alignment_delete(align);
gt_alphabet_delete(alphabet);
gt_scorehandler_delete(scorehandler);
return had_err;
}
int gt_paircmp(int argc, const char **argv, GtError *err)
{
int parsed_args;
Cmppairwiseopt cmppairwise;
GtOPrval oprval;
GtFastaReader *reader0 = NULL,
*reader1 = NULL;
gt_error_check(err);
oprval = parse_options(&parsed_args, &cmppairwise, argc, argv, err);
if (oprval == GT_OPTION_PARSER_OK)
{
gt_assert(parsed_args == argc);
showsimpleoptions(&cmppairwise);
if (cmppairwise.showedist)
{
GtUword edist, len1, len2;
GtStr *s1, *s2;
gt_assert(gt_str_array_size(cmppairwise.strings) >= 2);
s1 = gt_str_array_get_str(cmppairwise.strings,0);
s2 = gt_str_array_get_str(cmppairwise.strings,1UL);
len1 = gt_str_length(s1);
len2 = gt_str_length(s2);
edist = gt_computegreedyunitedist((const GtUchar *) gt_str_get(s1),
len1,
(const GtUchar *) gt_str_get(s2),
len2);
printf(GT_WU " " GT_WU " " GT_WU " " GT_WU "%% errors\n",
edist, len1,len2,(200 * edist)/(len1+len2));
}
else if (cmppairwise.print)
{
const GtStr *str0 = gt_str_array_get_str(cmppairwise.strings,0),
*str1 = gt_str_array_get_str(cmppairwise.strings,1);
gt_print_edist_alignment((const GtUchar *) gt_str_get(str0),0,
gt_str_length(str0),
(const GtUchar *) gt_str_get(str1),0,
gt_str_length(str1));
} else
{
size_t idx;
Checkfunctiontabentry checkfunction_tab[] = {
MAKECheckfunctiontabentry(gt_checkgreedyunitedist),
MAKECheckfunctiontabentry(gt_checklinearspace),
MAKECheckfunctiontabentry(gt_checklinearspace_local),
MAKECheckfunctiontabentry(gt_checkaffinelinearspace),
MAKECheckfunctiontabentry(gt_checkaffinelinearspace_local),
MAKECheckfunctiontabentry(gt_checkdiagonalbandalign),
MAKECheckfunctiontabentry(gt_checkdiagonalbandaffinealign)
};
if (cmppairwise.fasta)
{
gt_assert(gt_str_array_size(cmppairwise.files) == 3);
cmppairwise.fastasequences0 = gt_str_array_new();
cmppairwise.fastasequences1 = gt_str_array_new();
reader0 = gt_fasta_reader_rec_new(gt_str_array_get_str(
cmppairwise.files,1UL));
gt_fasta_reader_run(reader0, NULL, save_fastaentry,
NULL, cmppairwise.fastasequences0, err);
reader1 = gt_fasta_reader_rec_new (gt_str_array_get_str(
cmppairwise.files,2UL));
gt_fasta_reader_run(reader1, NULL, save_fastaentry,
NULL, cmppairwise.fastasequences1, err);
gt_error_check(err);
}
for (idx = 0; idx < sizeof checkfunction_tab/sizeof checkfunction_tab[0];
idx++)
{
GtUword testcases;
printf("run %s\n",checkfunction_tab[idx].name);
testcases
= applycheckfunctiontosimpleoptions(checkfunction_tab[idx].function,
&cmppairwise);
printf("# number of testcases for %s: " GT_WU "\n",
checkfunction_tab[idx].name,testcases);
}
gt_fasta_reader_delete(reader0);
gt_fasta_reader_delete(reader1);
}
}
freesimpleoption(&cmppairwise);
if (oprval == GT_OPTION_PARSER_REQUESTS_EXIT)
{
return 0;
}
if (oprval == GT_OPTION_PARSER_ERROR)
{
return -1;
}
return 0;
}
static int gff3_in_stream_plain_next(GtNodeStream *ns, GtGenomeNode **gn,
GtError *err)
{
GtGFF3InStreamPlain *is = gff3_in_stream_plain_cast(ns);
GtStr *filenamestr;
int had_err = 0, status_code;
gt_error_check(err);
if (gt_queue_size(is->genome_node_buffer) > 1) {
/* we still have at least two nodes in the buffer -> serve from there */
*gn = gt_queue_get(is->genome_node_buffer);
return 0;
}
/* the buffer is empty or has one element */
gt_assert(gt_queue_size(is->genome_node_buffer) <= 1);
for (;;) {
/* open file if necessary */
if (!is->file_is_open) {
if (gt_str_array_size(is->files) &&
is->next_file == gt_str_array_size(is->files)) {
break;
}
if (gt_str_array_size(is->files)) {
if (strcmp(gt_str_array_get(is->files, is->next_file), "-") == 0) {
if (is->stdin_argument) {
gt_error_set(err, "multiple specification of argument file \"-\"");
had_err = -1;
break;
}
is->fpin = gt_file_xopen(NULL, "r");
is->file_is_open = true;
is->stdin_argument = true;
}
else {
is->fpin = gt_file_xopen(gt_str_array_get(is->files,
is->next_file), "r");
is->file_is_open = true;
}
is->next_file++;
}
else {
if (is->stdin_processed)
break;
is->fpin = NULL;
is->file_is_open = true;
}
is->line_number = 0;
if (!had_err && is->progress_bar) {
printf("processing file \"%s\"\n", gt_str_array_size(is->files)
? gt_str_array_get(is->files, is->next_file-1) : "stdin");
}
if (!had_err && is->fpin && is->progress_bar) {
gt_progressbar_start(&is->line_number,
gt_file_number_of_lines(gt_str_array_get(is->files,
is->next_file-1)));
}
}
gt_assert(is->file_is_open);
filenamestr = gt_str_array_size(is->files)
? gt_str_array_get_str(is->files, is->next_file-1)
: is->stdinstr;
/* read two nodes */
had_err = gt_gff3_parser_parse_genome_nodes(is->gff3_parser, &status_code,
is->genome_node_buffer,
is->used_types, filenamestr,
&is->line_number, is->fpin,
err);
if (had_err)
break;
if (status_code != EOF) {
had_err = gt_gff3_parser_parse_genome_nodes(is->gff3_parser, &status_code,
is->genome_node_buffer,
is->used_types, filenamestr,
&is->line_number, is->fpin,
err);
if (had_err)
break;
}
if (status_code == EOF) {
/* end of current file */
if (is->progress_bar) gt_progressbar_stop();
gt_file_delete(is->fpin);
is->fpin = NULL;
is->file_is_open = false;
gt_gff3_parser_reset(is->gff3_parser);
if (!gt_str_array_size(is->files)) {
is->stdin_processed = true;
break;
}
continue;
}
gt_assert(gt_queue_size(is->genome_node_buffer));
/* make sure the parsed nodes are sorted */
if (is->ensure_sorting && gt_queue_size(is->genome_node_buffer) > 1) {
GtGenomeNode *last_node = NULL;
/* a sorted stream can have at most one input file */
gt_assert(gt_str_array_size(is->files) == 0 ||
gt_str_array_size(is->files) == 1);
had_err = gt_queue_iterate(is->genome_node_buffer, buffer_is_sorted,
&last_node, err);
}
if (!had_err) {
*gn = gt_queue_get(is->genome_node_buffer);
}
return had_err;
}
gt_assert(!gt_queue_size(is->genome_node_buffer));
*gn = NULL;
return had_err;
}