static int readkeysize(const char *indexname,GtError *err)
{
FILE *fp;
bool haserr = false;
char cc;
gt_error_check(err);
fp = gt_fa_fopen_with_suffix(indexname,GT_KEYSTABFILESUFFIX,"rb",err);
if (fp == NULL)
{
haserr = true;
}
if (!haserr)
{
GT_UNUSED size_t ret;
ret = fread(&cc,sizeof cc, (size_t) 1, fp);
if (ferror(fp))
{
gt_error_set(err,"error when trying to read first byte of file %s%s: %s",
indexname,GT_KEYSTABFILESUFFIX,strerror(errno));
haserr = true;
}
}
gt_assert(cc >= 0);
gt_fa_xfclose(fp);
return haserr ? -1 : (int) cc;
}
int gt_outprjfile(const char *indexname,
GtReadmode readmode,
const GtEncseq *encseq,
GtUword numberofallsortedsuffixes,
unsigned int prefixlength,
GtUword numoflargelcpvalues,
double averagelcp,
GtUword maxbranchdepth,
const Definedunsignedlong *longest,
GtError *err)
{
FILE *prjfp;
bool haserr = false;
gt_error_check(err);
prjfp = gt_fa_fopen_with_suffix(indexname,GT_PROJECTFILESUFFIX,"wb",err);
if (prjfp == NULL)
{
haserr = true;
}
if (!haserr)
{
showprjinfo(prjfp,
readmode,
encseq,
numberofallsortedsuffixes,
prefixlength,
numoflargelcpvalues,
averagelcp,
maxbranchdepth,
longest);
gt_fa_xfclose(prjfp);
}
return haserr ? -1 : 0;
}
int gt_hcr_decoder_decode_range(GtHcrDecoder *hcr_dec, const char *name,
GtUword start, GtUword end,
GtTimer *timer, GtError *err)
{
char qual[BUFSIZ] = {0},
seq[BUFSIZ] = {0};
GtStr *desc = gt_str_new();
int had_err = 0;
GtUword cur_width,
cur_read;
size_t i;
FILE *output;
GT_UNUSED GtHcrSeqDecoder *seq_dec;
gt_error_check(err);
gt_assert(hcr_dec && name);
seq_dec = hcr_dec->seq_dec;
gt_assert(start <= end);
gt_assert(start < seq_dec->num_of_reads && end < seq_dec->num_of_reads);
if (timer != NULL)
gt_timer_show_progress(timer, "decode hcr", stdout);
output = gt_fa_fopen_with_suffix(name, HCRFILEDECODEDSUFFIX, "w", err);
if (output == NULL)
had_err = -1;
for (cur_read = start; had_err == 0 && cur_read <= end; cur_read++) {
if (gt_hcr_decoder_decode(hcr_dec, cur_read, seq, qual, desc, err) != 0)
had_err = -1;
else {
gt_xfputc(HCR_DESCSEPSEQ, output);
if (hcr_dec->encdesc != NULL)
gt_xfputs(gt_str_get(desc), output);
else
fprintf(output, ""GT_WU"", cur_read);
gt_xfputc('\n', output);
for (i = 0, cur_width = 0; i < strlen(seq); i++, cur_width++) {
if (cur_width == HCR_LINEWIDTH) {
cur_width = 0;
gt_xfputc('\n', output);
}
gt_xfputc(seq[i], output);
}
gt_xfputc('\n', output);
gt_xfputc(HCR_DESCSEPQUAL, output);
gt_xfputc('\n', output);
for (i = 0, cur_width = 0; i < strlen(qual); i++, cur_width++) {
if (cur_width == HCR_LINEWIDTH) {
cur_width = 0;
gt_xfputc('\n', output);
}
gt_xfputc(qual[i], output);
}
gt_xfputc('\n', output);
}
}
gt_fa_xfclose(output);
gt_str_delete(desc);
return had_err;
}
static GtHcrSeqDecoder *hcr_seq_decoder_new(GtAlphabet *alpha, const char *name,
GtError *err)
{
GtHcrSeqDecoder *seq_dec = gt_malloc(sizeof (GtHcrSeqDecoder));
GtBaseQualDistr *bqd = NULL;
GtWord end_enc_start_sampling = 0;
FILE *fp = NULL;
GT_UNUSED size_t read,
one = (size_t) 1;
seq_dec->alpha = alpha;
seq_dec->alphabet_size = gt_alphabet_size(alpha);
seq_dec->cur_read = 0;
seq_dec->data_iter = NULL;
seq_dec->file_info_rbt = NULL;
seq_dec->fileinfos = NULL;
seq_dec->filename = gt_str_new_cstr(name);
seq_dec->huff_dec = NULL;
seq_dec->huffman = NULL;
seq_dec->sampling = NULL;
seq_dec->symbols = NULL;
gt_str_append_cstr(seq_dec->filename, HCRFILESUFFIX);
fp = gt_fa_fopen_with_suffix(name, HCRFILESUFFIX, "rb", err);
if (gt_error_is_set(err)) {
hcr_seq_decoder_delete(seq_dec);
seq_dec = NULL;
}
else {
hcr_read_file_info(seq_dec, fp);
bqd = hcr_base_qual_distr_new_from_file(fp, seq_dec->alpha);
seq_dec->qual_offset = bqd->qual_offset;
read = gt_xfread_one(&end_enc_start_sampling, fp);
gt_assert(read == one);
seq_dec->start_of_encoding = decoder_calc_start_of_encoded_data(fp);
seq_decoder_init_huffman(seq_dec, end_enc_start_sampling, bqd, err);
if (gt_error_is_set(err)) {
hcr_seq_decoder_delete(seq_dec);
seq_dec = NULL;
}
}
if (seq_dec != NULL) {
gt_xfseek(fp, end_enc_start_sampling, SEEK_SET);
seq_dec->sampling = gt_sampling_read(fp);
seq_dec->file_info_rbt = seq_decoder_init_file_info(seq_dec->fileinfos,
seq_dec->num_of_files);
}
hcr_base_qual_distr_delete(bqd);
gt_fa_fclose(fp);
return seq_dec;
}
static int hcr_write_seq_qual_data(const char *name, GtHcrEncoder *hcr_enc,
GtTimer *timer, GtError *err)
{
int had_err = 0;
FILE *fp;
GtUword dummy = 0;
GtWord pos;
gt_error_check(err);
fp = gt_fa_fopen_with_suffix(name, HCRFILESUFFIX, "wb", err);
if (fp == NULL)
had_err = -1;
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "write sequences and qualities encoding",
stdout);
hcr_write_file_info(fp, hcr_enc);
had_err = hcr_write_seqdistrtab(fp, hcr_enc);
if (!had_err) {
bool is_not_at_pageborder;
pos = ftell(fp);
gt_xfwrite_one(&dummy, fp);
is_not_at_pageborder = (ftell(fp) % hcr_enc->pagesize) != 0;
if (is_not_at_pageborder)
hcr_enc->seq_encoder->start_of_encoding =
(ftell(fp) / hcr_enc->pagesize + 1) * hcr_enc->pagesize;
else
hcr_enc->seq_encoder->start_of_encoding = ftell(fp);
if (hcr_enc->page_sampling)
hcr_enc->seq_encoder->sampling =
gt_sampling_new_page(hcr_enc->sampling_rate,
(off_t) hcr_enc->seq_encoder->start_of_encoding);
else if (hcr_enc->regular_sampling)
hcr_enc->seq_encoder->sampling =
gt_sampling_new_regular(hcr_enc->sampling_rate,
(off_t) hcr_enc->seq_encoder->start_of_encoding);
had_err = hcr_write_seqs(fp, hcr_enc, err);
}
if (!had_err) {
gt_assert(fp);
gt_xfseek(fp, pos, SEEK_SET);
gt_xfwrite_one(&hcr_enc->seq_encoder->startofsamplingtab, fp);
}
gt_fa_xfclose(fp);
}
return 0;
}
int gt_pckbuckettable_2file(const char *indexname,
const Pckbuckettable *pckbuckettable,
GtError *err)
{
FILE *fp;
unsigned long seqposmaxdepth;
gt_error_check(err);
fp = gt_fa_fopen_with_suffix(indexname,PCKBUCKETTABLE,"wb",err);
if (fp == NULL)
{
return -1;
}
seqposmaxdepth = (unsigned long) pckbuckettable->maxdepth;
gt_xfwrite(&seqposmaxdepth,sizeof (unsigned long),(size_t) 1,fp);
gt_xfwrite(pckbuckettable->mbtab[0],sizeof (Mbtab),
(size_t) pckbuckettable->maxnumofvalues,fp);
gt_fa_fclose(fp);
return 0;
}
int gt_alphabet_to_file(const GtAlphabet *alphabet, const char *indexname,
GtError *err)
{
FILE *al1fp;
bool haserr = false;
gt_error_check(err);
al1fp = gt_fa_fopen_with_suffix(indexname,GT_ALPHABETFILESUFFIX,"wb",err);
if (al1fp == NULL)
{
haserr = true;
}
if (!haserr)
{
GtStr *buf = gt_str_new();
gt_alphabet_to_str(alphabet, buf);
gt_xfwrite(gt_str_get(buf), sizeof (char), (size_t) gt_str_length(buf),
al1fp);
gt_fa_xfclose(al1fp);
gt_str_delete(buf);
}
return haserr ? -1 : 0;
}
static bool scanprjfileuintkeys(Suffixarray *suffixarray,
const char *indexname,
GtLogger *logger,
GtError *err)
{
bool haserr = false;
FILE *fp;
gt_error_check(err);
fp = gt_fa_fopen_with_suffix(indexname,GT_PROJECTFILESUFFIX,"rb",err);
if (fp == NULL)
{
haserr = true;
}
if (!haserr && scanprjfileuintkeysviafileptr(suffixarray,
indexname,logger,
fp,err) != 0)
{
haserr = true;
}
gt_fa_xfclose(fp);
return haserr;
}
/*read condenseq data structure from file*/
GtCondenseq *gt_condenseq_new_from_file(const char *indexname,
GtLogger *logger, GtError *err)
{
int had_err = 0;
FILE* fp;
GtEncseqLoader *esl;
GtEncseq *unique_es;
GtCondenseq *condenseq = NULL;
/*load unique_es*/
esl = gt_encseq_loader_new();
unique_es = gt_encseq_loader_load(esl, indexname, err);
if (!unique_es)
had_err = -1;
if (!had_err) {
gt_encseq_loader_delete(esl);
condenseq = condenseq_new_empty(gt_encseq_alphabet(unique_es));
condenseq->filename = gt_cstr_dup(indexname);
condenseq->unique_es = unique_es;
fp = gt_fa_fopen_with_suffix(indexname, GT_CONDENSEQ_FILE_SUFFIX,
"rb", err);
if (fp == NULL) {
had_err = -1;
}
else {
had_err = condenseq_io(condenseq, fp, gt_io_error_fread, err);
if (!had_err) {
GtUword i;
gt_assert(condenseq->uniques);
gt_assert(condenseq->links);
gt_fa_fclose(fp);
/*create link array for each unique entry*/
for (i = 0; i < condenseq->udb_nelems; i++) {
GT_INITARRAY(&(condenseq->uniques[i].links),uint32_t);
}
/* check for overflows */
if (condenseq->ldb_nelems > (GtUword) ((uint32_t) 0 - (uint32_t) 1)) {
gt_error_set(err, "Overflow, to many link-elements. Can't be stored");
had_err = -1;
}
/* iterate through link entrys and store ids in corresponding unique
entry array */
for (i = 0; !had_err && (GtUword) i < condenseq->ldb_nelems; i++) {
GtUword uid = condenseq->links[i].unique_id;
gt_assert(uid < condenseq->udb_nelems);
GT_STOREINARRAY(&(condenseq->uniques[uid].links),
uint32_t,
10,
(uint32_t) i);
}
}
}
}
if (!had_err) {
gt_assert(condenseq != NULL);
if (condenseq->id_len != GT_UNDEF_UWORD)
gt_logger_log(logger, "IDs const len: " GT_WU, condenseq->id_len);
else
gt_logger_log(logger, "using sdstab to access IDs");
}
if (had_err) {
gt_condenseq_delete(condenseq);
condenseq = NULL;
}
return (condenseq);
}
GtOutlcpinfo *gt_Outlcpinfo_new(const char *indexname,
unsigned int numofchars,
unsigned int prefixlength,
bool withdistribution,
bool swallow_tail_lcpvalues,
GtFinalProcessBucket final_process_bucket,
void *final_process_bucket_info,
GtError *err)
{
bool haserr = false;
GtOutlcpinfo *outlcpinfo;
outlcpinfo = gt_malloc(sizeof (*outlcpinfo));
outlcpinfo->sizeofinfo = sizeof (*outlcpinfo);
outlcpinfo->lcpsubtab.lcptabsum = 0.0;
outlcpinfo->swallow_tail_lcpvalues = swallow_tail_lcpvalues;
if (withdistribution)
{
outlcpinfo->lcpsubtab.distlcpvalues = gt_disc_distri_new();
} else
{
outlcpinfo->lcpsubtab.distlcpvalues = NULL;
}
if (indexname == NULL)
{
outlcpinfo->lcpsubtab.lcp2file = NULL;
if (final_process_bucket != NULL)
{
outlcpinfo->lcpsubtab.lcpprocess
= gt_malloc(sizeof (*outlcpinfo->lcpsubtab.lcpprocess));
outlcpinfo->lcpsubtab.lcpprocess->final_process_bucket
= final_process_bucket;
outlcpinfo->lcpsubtab.lcpprocess->final_process_bucket_info
= final_process_bucket_info;
} else
{
outlcpinfo->lcpsubtab.lcpprocess = NULL;
}
} else
{
outlcpinfo->lcpsubtab.lcpprocess = NULL;
outlcpinfo->lcpsubtab.lcp2file
= gt_malloc(sizeof (*outlcpinfo->lcpsubtab.lcp2file));
outlcpinfo->sizeofinfo += sizeof (*outlcpinfo->lcpsubtab.lcp2file);
outlcpinfo->lcpsubtab.lcp2file->countoutputlcpvalues = 0;
outlcpinfo->lcpsubtab.lcp2file->maxbranchdepth = 0;
outlcpinfo->lcpsubtab.lcp2file->totalnumoflargelcpvalues = 0;
outlcpinfo->lcpsubtab.lcp2file->reservoir = NULL;
outlcpinfo->lcpsubtab.lcp2file->sizereservoir = 0;
outlcpinfo->lcpsubtab.lcp2file->smalllcpvalues = NULL;
GT_INITARRAY(&outlcpinfo->lcpsubtab.lcp2file->largelcpvalues,
Largelcpvalue);
outlcpinfo->lcpsubtab.lcp2file->outfplcptab
= gt_fa_fopen_with_suffix(indexname,GT_LCPTABSUFFIX,"wb",err);
if (outlcpinfo->lcpsubtab.lcp2file->outfplcptab == NULL)
{
haserr = true;
}
if (!haserr)
{
outlcpinfo->lcpsubtab.lcp2file->outfpllvtab
= gt_fa_fopen_with_suffix(indexname,GT_LARGELCPTABSUFFIX,"wb",err);
if (outlcpinfo->lcpsubtab.lcp2file->outfpllvtab == NULL)
{
haserr = true;
}
}
}
outlcpinfo->numsuffixes2output = 0;
outlcpinfo->minchanged = 0;
if (!haserr && prefixlength > 0)
{
outlcpinfo->turnwheel = gt_turningwheel_new(prefixlength,numofchars);
outlcpinfo->sizeofinfo += gt_turningwheel_size();
} else
{
outlcpinfo->turnwheel = NULL;
}
#ifdef SKDEBUG
outlcpinfo->previoussuffix.startpos = 0;
#endif
outlcpinfo->previoussuffix.code = 0;
outlcpinfo->previoussuffix.prefixindex = 0;
outlcpinfo->previoussuffix.defined = false;
outlcpinfo->previousbucketwasempty = false;
outlcpinfo->lcpsubtab.tableoflcpvalues.bucketoflcpvalues = NULL;
outlcpinfo->lcpsubtab.tableoflcpvalues.numofentries = 0;
#ifndef NDEBUG
outlcpinfo->lcpsubtab.tableoflcpvalues.isset = NULL;
#endif
if (haserr)
{
gt_free(outlcpinfo);
return NULL;
}
return outlcpinfo;
}
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 readfirstvaluesfromfile(GtEncseqMetadata *emd,
const char *indexname, GtError *err)
{
FILE *fp;
bool had_err = false;
unsigned long cc, byteoffset = 0, alphatype;
char *alphadef;
gt_error_check(err);
fp = gt_fa_fopen_with_suffix(indexname, GT_ENCSEQFILESUFFIX, "rb", err);
if (fp == NULL)
{
had_err = true;
}
NEXTFREAD(emd->is64bit);
if (!had_err)
{
if ((int) emd->is64bit > 1)
{
gt_error_set(err, "illegal platform code %u in \"%s%s\"", emd->is64bit,
indexname, GT_ENCSEQFILESUFFIX);
had_err = true;
}
if (!had_err && ((emd->is64bit && sizeof (unsigned long) != (size_t) 8)
|| (!emd->is64bit && sizeof (unsigned long) == (size_t) 8)))
{
gt_error_set(err, "trying to load 64-bit index \"%s%s\" on a 32-bit "
"system or vice versa -- please use correct index "
"for this platform", indexname, GT_ENCSEQFILESUFFIX);
had_err = true;
}
}
NEXTFREAD(emd->version);
if (!had_err)
{
if (emd->version < GT_ENCSEQ_VERSION) {
gt_error_set(err, "index \"%s%s\" is format version %lu, current is "
"%lu -- please re-encode",
indexname, GT_ENCSEQFILESUFFIX,
emd->version, GT_ENCSEQ_VERSION);
had_err = true;
}
}
NEXTFREAD(cc);
if (!had_err)
{
if (cc >= (unsigned long) GT_ACCESS_TYPE_UNDEFINED)
{
gt_error_set(err, "illegal type %lu in \"%s%s\"", cc,
indexname, GT_ENCSEQFILESUFFIX);
had_err = true;
}
}
if (!had_err) {
emd->sat = (GtEncseqAccessType) cc;
NEXTFREAD(emd->totallength);
NEXTFREAD(emd->numofdbsequences);
NEXTFREAD(emd->numofdbfiles);
NEXTFREAD(emd->lengthofdbfilenames);
NEXTFREAD(emd->specialcharinfo);
NEXTFREAD(emd->minseqlen);
NEXTFREAD(emd->maxseqlen);
}
NEXTFREAD(alphatype);
if (!had_err) {
if (alphatype > 2UL) {
gt_error_set(err, "illegal alphabet type %lu in \"%s%s\"", alphatype,
indexname, GT_ENCSEQFILESUFFIX);
had_err = true;
}
}
if (!had_err) {
NEXTFREAD(emd->lengthofalphadef);
switch (alphatype) {
case 0:
emd->alpha = gt_alphabet_new_dna();
break;
case 1:
emd->alpha = gt_alphabet_new_protein();
break;
case 2:
gt_assert(emd->lengthofalphadef > 0);
emd->customalphabet = true;
alphadef = gt_malloc(sizeof (char) * emd->lengthofalphadef);
NEXTFREADWSIZE(*(alphadef), emd->lengthofalphadef);
emd->alpha = gt_alphabet_new_from_string(alphadef,
emd->lengthofalphadef,
err);
if (!emd->alpha) {
had_err = true;
}
gt_free(alphadef);
break;
}
gt_assert(emd->alpha != NULL);
}
gt_fa_xfclose(fp);
return had_err ? -1 : 0;
}
int gt_mapfmindex (Fmindex *fmindex,const char *indexname,
GtLogger *logger,GtError *err)
{
FILE *fpin = NULL;
bool haserr = false, storeindexpos = true;
GtSpecialcharinfo specialcharinfo;
gt_error_check(err);
fmindex->mappedptr = NULL;
fmindex->bwtformatching = NULL;
fmindex->alphabet = NULL;
fpin = gt_fa_fopen_with_suffix(indexname,FMASCIIFILESUFFIX,"rb",err);
if (fpin == NULL)
{
haserr = true;
}
if (!haserr)
{
if (scanfmafileviafileptr(fmindex,
&specialcharinfo,
&storeindexpos,
indexname,
fpin,
logger,
err) != 0)
{
haserr = true;
}
}
gt_fa_xfclose(fpin);
if (!haserr)
{
fmindex->bwtformatching = mapbwtencoding(indexname,logger,err);
if (fmindex->bwtformatching == NULL)
{
haserr = true;
}
}
if (!haserr)
{
fmindex->specpos.nextfreeGtPairBwtidx
= (unsigned long) gt_determinenumberofspecialstostore(&specialcharinfo);
fmindex->specpos.spaceGtPairBwtidx = NULL;
fmindex->specpos.allocatedGtPairBwtidx = 0;
fmindex->alphabet = gt_alphabet_ref(
gt_encseq_alphabet(fmindex->bwtformatching));
if (fmindex->alphabet == NULL)
{
haserr = true;
}
}
if (!haserr)
{
GtStr *tmpfilename;
gt_computefmkeyvalues (fmindex,
&specialcharinfo,
fmindex->bwtlength,
fmindex->log2bsize,
fmindex->log2markdist,
gt_alphabet_num_of_chars(fmindex->alphabet),
fmindex->suffixlength,
storeindexpos);
tmpfilename = gt_str_new_cstr(indexname);
gt_str_append_cstr(tmpfilename,FMDATAFILESUFFIX);
if (gt_fillfmmapspecstartptr(fmindex,storeindexpos,tmpfilename,err) != 0)
{
haserr = true;
}
gt_str_delete(tmpfilename);
}
if (haserr)
{
gt_freefmindex(fmindex);
}
return haserr ? -1 : 0;
}
int gt_extractkeysfromdesfile(const char *indexname,
bool sortkeys,
GtLogger *logger,
GtError *err)
{
FILE *fpin, *fpout = NULL;
GtStr *line = NULL;
const char *keyptr;
unsigned long keylen, constantkeylen = 0, linenum;/* incorrectorder = 0;*/
bool haserr = false, firstdesc = true;
char *previouskey = NULL;
Fixedsizekey *keytab = NULL, *keytabptr = NULL;
GtEncseq *encseq = NULL;
unsigned long numofentries = 0;
const unsigned long linewidth = 60UL;
fpin = gt_fa_fopen_with_suffix(indexname,GT_DESTABFILESUFFIX,"rb",err);
if (fpin == NULL)
{
return -1;
}
if (!sortkeys)
{
fpout = gt_fa_fopen_with_suffix(indexname,GT_KEYSTABFILESUFFIX,"wb",err);
if (fpout == NULL)
{
haserr = true;
}
}
if (!haserr)
{
line = gt_str_new();
}
for (linenum = 0; !haserr && gt_str_read_next_line(line, fpin) != EOF;
linenum++)
{
keyptr = desc2key(&keylen,gt_str_get(line),err);
if (keyptr == NULL)
{
haserr = true;
break;
}
if (keylen == 0)
{
gt_error_set(err,"key of length 0 in \"%s\" not expected",
gt_str_get(line));
haserr = true;
break;
}
if (firstdesc)
{
if (keylen > (unsigned long) CHAR_MAX)
{
gt_error_set(err,"key \"%*.*s\" of length %lu not allowed; "
"no key must be larger than %d",
(int) keylen,(int) keylen,keyptr,keylen,CHAR_MAX);
haserr = true;
break;
}
constantkeylen = keylen;
previouskey = gt_malloc(sizeof (char) * (constantkeylen+1));
firstdesc = false;
if (!sortkeys)
{
gt_xfputc((char) constantkeylen,fpout);
} else
{
GtEncseqLoader *el;
if (constantkeylen > (unsigned long) MAXFIXEDKEYSIZE)
{
gt_error_set(err,"key \"%*.*s\" of length %lu not allowed; "
"no key must be larger than %d",
(int) keylen,(int) keylen,keyptr,keylen,
MAXFIXEDKEYSIZE);
haserr = true;
break;
}
el = gt_encseq_loader_new();
gt_encseq_loader_set_logger(el, logger);
encseq = gt_encseq_loader_load(el, indexname, err);
gt_encseq_loader_delete(el);
if (encseq == NULL)
{
haserr = true;
break;
}
numofentries = gt_encseq_num_of_sequences(encseq);
gt_assert(numofentries > 0);
keytab = gt_malloc(sizeof (*keytab) * numofentries);
keytabptr = keytab;
}
} else
{
if (constantkeylen != keylen)
{
gt_error_set(err,"key \"%*.*s\" of length %lu: all keys must be of "
"the same length which for all previously seen "
"headers is %lu",
(int) keylen,(int) keylen,keyptr,keylen,
constantkeylen);
haserr = true;
break;
}
gt_assert(previouskey != NULL);
if (!sortkeys && strncmp(previouskey,keyptr,(size_t) constantkeylen) >= 0)
{
gt_error_set(err,"previous key \"%s\" is not lexicographically smaller "
"than current key \"%*.*s\"",
previouskey,(int) keylen,(int) keylen,keyptr);
haserr = true;
break;
/*
printf("previous key \"%s\" (no %lu) is lexicographically larger "
"than current key \"%*.*s\"\n",
previouskey,linenum,(int) keylen,(int) keylen,keyptr);
incorrectorder++;
*/
}
}
if (!sortkeys)
{
gt_xfwrite(keyptr,sizeof *keyptr,(size_t) keylen,fpout);
gt_xfputc('\0',fpout);
} else
{
gt_assert(keytabptr != NULL);
strncpy(keytabptr->key,keyptr,(size_t) constantkeylen);
keytabptr->key[constantkeylen] = '\0';
keytabptr->seqnum = linenum;
keytabptr++;
}
strncpy(previouskey,keyptr,(size_t) constantkeylen);
previouskey[constantkeylen] = '\0';
gt_str_reset(line);
}
if (!haserr)
{
gt_logger_log(logger,"number of keys of length %lu = %lu",
constantkeylen,linenum);
/*
gt_logger_log(logger,"number of incorrectly ordered keys = %lu",
incorrectorder);
*/
}
gt_str_delete(line);
gt_fa_fclose(fpin);
gt_fa_fclose(fpout);
gt_free(previouskey);
if (!haserr && sortkeys)
{
gt_assert(keytabptr != NULL);
gt_assert(numofentries > 0);
gt_assert(keytabptr == keytab + numofentries);
qsort(keytab,(size_t) numofentries,sizeof (*keytab),compareFixedkeys);
gt_assert(keytabptr != NULL);
for (keytabptr = keytab; !haserr && keytabptr < keytab + numofentries;
keytabptr++)
{
if (giextract_encodedseq2fasta(stdout,
encseq,
keytabptr->seqnum,
NULL,
linewidth,
err) != 0)
{
haserr = true;
break;
}
}
}
if (encseq != NULL)
{
gt_encseq_delete(encseq);
encseq = NULL;
}
gt_free(keytab);
return haserr ? -1 : 0;
}
static int enumeratelcpintervals(const char *inputindex,
Sequentialsuffixarrayreader *ssar,
const char *storeindex,
bool storecounts,
GtUword mersize,
GtUword minocc,
GtUword maxocc,
bool performtest,
GtLogger *logger,
GtError *err)
{
TyrDfsstate *state;
bool haserr = false;
unsigned int alphasize;
gt_error_check(err);
state = gt_malloc(sizeof (*state));
GT_INITARRAY(&state->occdistribution,Countwithpositions);
state->esrspace = gt_encseq_create_reader_with_readmode(
gt_encseqSequentialsuffixarrayreader(ssar),
gt_readmodeSequentialsuffixarrayreader(ssar),
0);
state->mersize = (GtUword) mersize;
state->encseq = gt_encseqSequentialsuffixarrayreader(ssar);
alphasize = gt_alphabet_num_of_chars(gt_encseq_alphabet(state->encseq));
state->readmode = gt_readmodeSequentialsuffixarrayreader(ssar);
state->storecounts = storecounts;
state->minocc = minocc;
state->maxocc = maxocc;
state->totallength = gt_encseq_total_length(state->encseq);
state->performtest = performtest;
state->countoutputmers = 0;
state->merindexfpout = NULL;
state->countsfilefpout = NULL;
GT_INITARRAY(&state->largecounts,Largecount);
if (strlen(storeindex) == 0)
{
state->sizeofbuffer = 0;
state->bytebuffer = NULL;
} else
{
state->sizeofbuffer = MERBYTES(mersize);
state->bytebuffer = gt_malloc(sizeof *state->bytebuffer
* state->sizeofbuffer);
}
if (performtest)
{
state->currentmer = gt_malloc(sizeof *state->currentmer
* state->mersize);
state->suftab = gt_suftabSequentialsuffixarrayreader(ssar);
} else
{
state->currentmer = NULL;
state->suftab = NULL;
}
if (state->mersize > state->totallength)
{
gt_error_set(err,"mersize "GT_WU" > "GT_WU" = totallength not allowed",
state->mersize,
state->totallength);
haserr = true;
} else
{
if (strlen(storeindex) == 0)
{
state->processoccurrencecount = adddistpos2distribution;
} else
{
state->merindexfpout = gt_fa_fopen_with_suffix(storeindex,MERSUFFIX,
"wb",err);
if (state->merindexfpout == NULL)
{
haserr = true;
} else
{
if (state->storecounts)
{
state->countsfilefpout
= gt_fa_fopen_with_suffix(storeindex,COUNTSSUFFIX,"wb",err);
if (state->countsfilefpout == NULL)
{
haserr = true;
}
}
}
state->processoccurrencecount = outputsortedstring2index;
}
if (!haserr)
{
if (gt_depthfirstesa(ssar,
tyr_allocateDfsinfo,
tyr_freeDfsinfo,
tyr_processleafedge,
NULL,
tyr_processcompletenode,
tyr_assignleftmostleaf,
tyr_assignrightmostleaf,
(Dfsstate*) state,
logger,
err) != 0)
{
haserr = true;
}
if (strlen(storeindex) == 0)
{
showfinalstatistics(state,inputindex,logger);
}
}
if (!haserr)
{
if (state->countsfilefpout != NULL)
{
gt_logger_log(logger,"write "GT_WU" mercounts > "GT_WU
" to file \"%s%s\"",
state->largecounts.nextfreeLargecount,
(GtUword) MAXSMALLMERCOUNT,
storeindex,
COUNTSSUFFIX);
gt_xfwrite(state->largecounts.spaceLargecount, sizeof (Largecount),
(size_t) state->largecounts.nextfreeLargecount,
state->countsfilefpout);
}
}
if (!haserr)
{
gt_logger_log(logger,"number of "GT_WU"-mers in index: "GT_WU"",
mersize,
state->countoutputmers);
gt_logger_log(logger,"index size: %.2f megabytes\n",
GT_MEGABYTES(state->countoutputmers * state->sizeofbuffer +
sizeof (GtUword) * EXTRAINTEGERS));
}
}
/* now out EXTRAINTEGERS integer values */
if (!haserr && state->merindexfpout != NULL)
{
outputbytewiseUlongvalue(state->merindexfpout,
(GtUword) state->mersize);
outputbytewiseUlongvalue(state->merindexfpout,(GtUword) alphasize);
}
gt_fa_xfclose(state->merindexfpout);
gt_fa_xfclose(state->countsfilefpout);
GT_FREEARRAY(&state->occdistribution,Countwithpositions);
gt_free(state->currentmer);
gt_free(state->bytebuffer);
GT_FREEARRAY(&state->largecounts,Largecount);
gt_encseq_reader_delete(state->esrspace);
gt_free(state);
return haserr ? -1 : 0;
}
