void gt_checksortedsuffixes(const char *filename,
int line,
const GtEncseq *encseq,
GtReadmode readmode,
const GtSuffixsortspace *suffixsortspace,
GtUword subbucketleft,
GtUword numberofsuffixes,
bool specialsareequal,
bool specialsareequalatdepth0,
GtUword depth)
{
GtUword idx, pos1, pos2, maxlcp,
totallength = gt_encseq_total_length(encseq);
GtEncseqReader *esr1, *esr2;
int cmp;
gt_assert(!specialsareequal || specialsareequalatdepth0);
esr1 = gt_encseq_create_reader_with_readmode(encseq, readmode, 0);
esr2 = gt_encseq_create_reader_with_readmode(encseq, readmode, 0);
gt_assert(numberofsuffixes > 0);
pos1 = gt_suffixsortspace_get(suffixsortspace,subbucketleft,0);
gt_assert(pos1 < totallength);
for (idx = 1UL; idx < numberofsuffixes; idx++)
{
pos2 = gt_suffixsortspace_get(suffixsortspace,subbucketleft,idx);
if (pos2 < totallength)
{
cmp = gt_encseq_check_comparetwosuffixes(encseq,
readmode,
&maxlcp,
specialsareequal,
specialsareequalatdepth0,
depth,
pos1,
pos2,
esr1,
esr2);
if (cmp > 0)
{
showcomparisonfailure(filename,
line,
"checksortedsuffixes",
encseq,
readmode,
suffixsortspace,
subbucketleft,
depth,
idx-1,
idx,
cmp,
maxlcp);
exit(GT_EXIT_PROGRAMMING_ERROR);
}
gt_assert(depth == 0 || maxlcp <= depth);
}
pos1 = pos2;
}
gt_encseq_reader_delete(esr1);
gt_encseq_reader_delete(esr2);
}
static void gt_seqorder_output(unsigned long seqnum, GtEncseq *encseq)
{
GtEncseqReader *esr;
unsigned long startpos, len, desclen = 0;
const char *desc = NULL;
unsigned long i;
startpos = gt_encseq_seqstartpos(encseq, seqnum);
len = gt_encseq_seqlength(encseq, seqnum);
gt_xfputc(GT_FASTA_SEPARATOR, stdout);
if (gt_encseq_has_description_support(encseq))
{
desc = gt_encseq_description(encseq, &desclen, seqnum);
gt_xfwrite(desc, (size_t)1, (size_t)desclen, stdout);
}
gt_xfputc('\n', stdout);
esr = gt_encseq_create_reader_with_readmode(encseq, GT_READMODE_FORWARD,
startpos);
for (i = 0; i < len; i++)
{
gt_xfputc(gt_encseq_reader_next_decoded_char(esr), stdout);
}
gt_encseq_reader_delete(esr);
gt_xfputc('\n', stdout);
}
static void gt_mmsearchiterator_reinit(GtMMsearchiterator *mmsi,
const GtEncseq *dbencseq,
const ESASuffixptr *suftab,
GtUword leftbound,
GtUword rightbound,
GtUword itvoffset,
GtReadmode readmode,
const GtQuerysubstring *querysubstring,
GtUword minmatchlength)
{
mmsi->suftab = suftab;
if (mmsi->esr == NULL)
{
mmsi->esr = gt_encseq_create_reader_with_readmode(dbencseq, readmode, 0);
} else
{
gt_encseq_reader_reinit_with_readmode(mmsi->esr,dbencseq,readmode,0);
}
mmsi->lcpitv.left = leftbound;
mmsi->lcpitv.right = rightbound;
mmsi->lcpitv.offset = itvoffset;
if (!gt_mmsearch(dbencseq,mmsi->esr,suftab,readmode,&mmsi->lcpitv,
querysubstring,minmatchlength))
{
mmsi->lcpitv.left = 1UL;
mmsi->lcpitv.right = 0;
}
mmsi->sufindex = mmsi->lcpitv.left;
}
static GtMMsearchiterator *gt_mmsearchiterator_new_generic(
const GtEncseq *dbencseq,
const ESASuffixptr *suftab,
unsigned long leftbound,
unsigned long rightbound,
unsigned long itvoffset,
GtReadmode readmode,
const GtQuerysubstring *querysubstring,
unsigned long minmatchlength)
{
GtMMsearchiterator *mmsi = gt_malloc(sizeof *mmsi);
mmsi->lcpitv.left = leftbound;
mmsi->lcpitv.right = rightbound;
mmsi->lcpitv.offset = itvoffset;
mmsi->suftab = suftab;
mmsi->esr = gt_encseq_create_reader_with_readmode(dbencseq, readmode, 0);
if (!gt_mmsearch(dbencseq,mmsi->esr,suftab,readmode,&mmsi->lcpitv,
querysubstring,minmatchlength))
{
mmsi->lcpitv.left = 1UL;
mmsi->lcpitv.right = 0;
}
mmsi->sufindex = mmsi->lcpitv.left;
return mmsi;
}
void gt_checkifprefixesareidentical(const char *filename,
int line,
const GtEncseq *encseq,
GtReadmode readmode,
const GtSuffixsortspace *suffixsortspace,
GtUword subbucketleft,
GtUword width,
GtUword depth)
{
GtUword idx, maxlcp, pos1, pos2;
int cmp;
GtEncseqReader *esr1, *esr2;
esr1 = gt_encseq_create_reader_with_readmode(encseq, readmode, 0);
esr2 = gt_encseq_create_reader_with_readmode(encseq, readmode, 0);
gt_assert(depth > 0);
for (idx = 0; idx < width-1; idx++)
{
pos1 = gt_suffixsortspace_get(suffixsortspace,subbucketleft,idx);
pos2 = gt_suffixsortspace_get(suffixsortspace,subbucketleft,idx+1);
cmp = gt_encseq_check_comparetwosuffixes(encseq,
readmode,
&maxlcp,
false, /* specialsareequal */
true,/* specialsareequalatdepth0 */
depth,pos1,
pos2,
esr1,
esr2);
gt_assert(maxlcp <= depth);
if (cmp != 0 || maxlcp < depth)
{
showcomparisonfailure(filename,
line,
"checkifprefixesareidentical",
encseq,
readmode,
suffixsortspace,
subbucketleft,
depth,
idx,idx+1,cmp,maxlcp);
exit(GT_EXIT_PROGRAMMING_ERROR);
}
}
gt_encseq_reader_delete(esr1);
gt_encseq_reader_delete(esr2);
}
void getencseqkmers(const GtEncseq *encseq,
GtReadmode readmode,
unsigned int kmersize,
void(*processkmercode)(void *,
unsigned long,
const GtKmercode *),
void *processkmercodeinfo)
{
unsigned long currentposition = 0, totallength;
Kmerstream *spwp;
GtUchar charcode;
GtEncseqReader *esr;
unsigned int numofchars, overshoot;
totallength = gt_encseq_total_length(encseq);
if (totallength < (unsigned long) kmersize)
{
return;
}
numofchars = gt_alphabet_num_of_chars(gt_encseq_alphabet(encseq));
spwp = kmerstream_new(numofchars,kmersize);
esr = gt_encseq_create_reader_with_readmode(encseq,readmode,0);
for (currentposition = 0; currentposition < (unsigned long) kmersize;
currentposition++)
{
charcode = gt_encseq_reader_next_encoded_char(esr);
GT_CHECKENCCHAR(charcode,encseq,currentposition,readmode);
spwp->windowwidth++;
updatespecialpositions(spwp,charcode,false,0);
spwp->cyclicwindow[spwp->windowwidth-1] = charcode;
}
kmerstream_newcode(&spwp->currentkmercode,spwp);
processkmercode(processkmercodeinfo,0,&spwp->currentkmercode);
for (currentposition = (unsigned long) kmersize; currentposition<totallength;
currentposition++)
{
charcode = gt_encseq_reader_next_encoded_char(esr);
GT_CHECKENCCHAR(charcode,encseq,currentposition,readmode);
shiftrightwithchar(spwp,charcode);
kmerstream_newcode(&spwp->currentkmercode,spwp);
processkmercode(processkmercodeinfo,currentposition + 1 - spwp->kmersize,
&spwp->currentkmercode);
}
gt_encseq_reader_delete(esr);
for (overshoot=0; overshoot<kmersize; overshoot++)
{
shiftrightwithchar(spwp,(GtUchar) WILDCARD);
kmerstream_newcode(&spwp->currentkmercode,spwp);
processkmercode(processkmercodeinfo,
overshoot + currentposition + 1 - spwp->kmersize,
&spwp->currentkmercode);
}
kmerstream_delete(spwp);
}
static int encseq_lua_create_reader_with_readmode(lua_State *L)
{
GtEncseq **encseq;
GtEncseqReader *reader;
GtUword startpos;
GtReadmode readmode;
encseq = check_encseq(L, 1);
readmode = luaL_checknumber(L, 2);
startpos = luaL_checknumber(L, 3);
luaL_argcheck(L, startpos < gt_encseq_total_length(*encseq), 3,
"cannot exceed total length of encoded sequence");
reader = gt_encseq_create_reader_with_readmode(*encseq, readmode, startpos);
gt_assert(reader);
gt_lua_encseq_reader_push(L, reader);
return 1;
}
/*@notnull@*/ GtKmercodeiterator *gt_kmercodeiterator_encseq_new(
const GtEncseq *encseq,
GtReadmode readmode,
unsigned int kmersize,
unsigned long startpos)
{
GtKmercodeiterator *kmercodeiterator;
unsigned int numofchars;
GtUchar charcode;
gt_assert(!GT_ISDIRREVERSE(readmode) || startpos == 0);
kmercodeiterator = gt_malloc(sizeof (*kmercodeiterator));
kmercodeiterator->totallength = gt_encseq_total_length(encseq);
kmercodeiterator->startpos = startpos;
gt_assert(startpos < kmercodeiterator->totallength);
if (kmercodeiterator->totallength - startpos < (unsigned long) kmersize)
{
kmercodeiterator->inputexhausted = true;
kmercodeiterator->fb = NULL;
kmercodeiterator->encseq = encseq;
kmercodeiterator->esr = NULL;
kmercodeiterator->spwp = NULL;
} else
{
kmercodeiterator->inputexhausted = false;
kmercodeiterator->fb = NULL;
kmercodeiterator->encseq = encseq;
kmercodeiterator->readmode = readmode;
kmercodeiterator->esr = gt_encseq_create_reader_with_readmode(encseq,
readmode,
startpos);
numofchars = gt_alphabet_num_of_chars(gt_encseq_alphabet(encseq));
kmercodeiterator->spwp = kmerstream_new(numofchars,kmersize);
kmercodeiterator->hasprocessedfirst = false;
for (kmercodeiterator->currentposition = startpos;
kmercodeiterator->currentposition < startpos+(unsigned long) kmersize;
kmercodeiterator->currentposition++)
{
charcode = gt_encseq_reader_next_encoded_char(kmercodeiterator->esr);
kmercodeiterator->spwp->windowwidth++;
updatespecialpositions(kmercodeiterator->spwp,charcode,false,0);
kmercodeiterator->spwp->cyclicwindow[kmercodeiterator->
spwp->windowwidth-1] = charcode;
}
}
return kmercodeiterator;
}
void gt_encseq2symbolstring(FILE *fpout,
const GtEncseq *encseq,
GtReadmode readmode,
unsigned long start,
unsigned long wlen,
unsigned long width)
{
unsigned long j, idx, lastpos;
GtUchar currentchar;
GtEncseqReader *esr;
const GtAlphabet *alpha;
esr = gt_encseq_create_reader_with_readmode(encseq, readmode, start);
gt_assert(width > 0);
lastpos = start + wlen - 1;
alpha = gt_encseq_alphabet(encseq);
for (idx = start, j = 0; /* Nothing */ ; idx++)
{
currentchar = gt_encseq_reader_next_encoded_char(esr);
if (currentchar == (GtUchar) SEPARATOR)
{
fprintf(fpout,"\n>\n");
j = 0;
} else
{
gt_alphabet_echo_pretty_symbol(alpha,fpout,currentchar);
}
if (idx == lastpos)
{
fprintf(fpout,"\n");
break;
}
if (currentchar != (GtUchar) SEPARATOR)
{
j++;
if (j >= width)
{
fprintf(fpout,"\n");
j = 0;
}
}
}
gt_encseq_reader_delete(esr);
}
Myersonlineresources *gt_newMyersonlineresources(unsigned int numofchars,
bool nowildcards,
const GtEncseq *encseq,
ProcessIdxMatch processmatch,
void *processmatchinfo)
{
Myersonlineresources *mor;
mor = gt_malloc(sizeof *mor);
mor->eqsvectorrev = gt_malloc(sizeof *mor->eqsvectorrev * numofchars);
mor->eqsvector = gt_malloc(sizeof *mor->eqsvector * numofchars);
mor->encseq = encseq;
mor->esr = gt_encseq_create_reader_with_readmode(encseq, GT_READMODE_REVERSE,
0);
gt_assert(numofchars <= GT_MAXALPHABETCHARACTER);
mor->alphasize = numofchars;
mor->totallength = gt_encseq_total_length(encseq);
mor->nowildcards = nowildcards;
mor->processmatch = processmatch;
mor->processmatchinfo = processmatchinfo;
return mor;
}
static void gt_wtree_encseq_fill_bits(GtWtreeEncseq *we)
{
unsigned int level_idx;
GtUword sym_idx;
GtEncseqReader *er =
gt_encseq_create_reader_with_readmode(we->encseq, GT_READMODE_FORWARD, 0);
gt_assert(we != NULL);
for (level_idx = 0; level_idx < we->levels; level_idx++) {
for (sym_idx = 0;
sym_idx < we->parent_instance.members->length;
sym_idx++) {
GtWtreeSymbol c_sym =
gt_wtree_encseq_map(we, gt_encseq_reader_next_encoded_char(er));
if (gt_wtree_encseq_set_nodestart_and_current_fo(we, level_idx, c_sym)) {
/*0*/
if (we->current_fo != NULL) {
gt_assert(we->node_start + we->current_fo->offset < we->num_of_bits);
we->current_fo->offset++;
we->current_fo->left_size++;
}
}
else {
if (we->current_fo != NULL) {
gt_assert(we->node_start + we->current_fo->offset < we->num_of_bits);
GT_SETIBIT(we->bits, we->node_start + we->current_fo->offset);
we->current_fo->offset++;
}
}
}
gt_encseq_reader_reinit_with_readmode(er, we->encseq,
GT_READMODE_FORWARD, 0);
}
gt_encseq_reader_delete(er);
gt_wtree_encseq_fill_offset_delete(we->root_fo);
we->root_fo = we->current_fo = NULL;
gt_encseq_delete(we->encseq);
we->encseq = NULL;
}
void gt_lookaheadsearchPSSM(const GtEncseq *encseq,
const Profilematrix *prof)
{
unsigned long firstpos, bufsize;
GtUchar currentchar;
unsigned long pos;
GtEncseqReader *esr;
unsigned long totallength = gt_encseq_total_length(encseq);
GtUchar *buffer;
esr = gt_encseq_create_reader_with_readmode(encseq,GT_READMODE_FORWARD,0);
buffer = gt_malloc(sizeof *buffer * prof->dimension);
firstpos = bufsize = 0;
for (pos=0; pos < totallength; pos++)
{
currentchar = gt_encseq_reader_next_encoded_char(esr);
if (ISSPECIAL(currentchar))
{
bufsize = firstpos = 0;
} else
{
if (bufsize < prof->dimension)
{
buffer[bufsize++] = currentchar;
} else
{
buffer[firstpos++] = currentchar;
if (firstpos == prof->dimension)
{
firstpos = 0;
}
}
}
}
gt_encseq_reader_delete(esr);
gt_free(buffer);
}
double *gt_encseq_get_gc(const GtEncseq *encseq,
bool with_special,
bool calculate,
GT_UNUSED GtError *err)
{
GtEncseqReader *reader;
GtAlphabet *alphabet;
double *gc_content;
/* unit = file or sequence depending on per_file */
unsigned long char_idx, totallength, max_unit,
seq_idx = 0,
nextsep = 0,
at_count = 0,
gc_count = 0,
default_count = 0;
bool is_mirrored_encseq;
GtUchar acgt[8], current_c;
alphabet = gt_encseq_alphabet(encseq);
gt_assert(gt_alphabet_is_dna(alphabet));
gt_alphabet_encode_seq(alphabet, acgt,
"aAtTcCgG", 8UL);
totallength = gt_encseq_total_length(encseq);
reader = gt_encseq_create_reader_with_readmode(encseq,
GT_READMODE_FORWARD,
0);
is_mirrored_encseq = gt_encseq_is_mirrored(encseq);
if (is_mirrored_encseq)
{
max_unit = GT_DIV2(gt_encseq_num_of_sequences(encseq));
gc_content = gt_calloc((size_t) GT_MULT2(max_unit), sizeof (double));
}
else
{
max_unit = gt_encseq_num_of_sequences(encseq);
gc_content = gt_calloc((size_t) max_unit, sizeof (double));
}
nextsep = gt_encseq_seqstartpos(encseq, seq_idx) +
gt_encseq_seqlength(encseq, seq_idx);
for (char_idx = 0; char_idx < totallength; char_idx++)
{
if (nextsep == char_idx)
{
if (calculate)
{
calculate_gc(encseq,
gc_content,
with_special,
seq_idx,
gc_count,
at_count);
}
else
{
gc_content[seq_idx] = (double) gc_count;
}
seq_idx++;
nextsep = gt_encseq_seqstartpos(encseq, seq_idx) +
gt_encseq_seqlength(encseq, seq_idx);
gt_encseq_reader_reinit_with_readmode(reader,
encseq,
GT_READMODE_FORWARD,
char_idx + 1UL);
gc_count = at_count = default_count = 0UL;
continue;
}
current_c = gt_encseq_reader_next_encoded_char(reader);
if (current_c == acgt[0] ||
current_c == acgt[1] ||
current_c == acgt[2] ||
current_c == acgt[3])
{
at_count++;
}
else
{
if (current_c == acgt[4] ||
current_c == acgt[5] ||
current_c == acgt[6] ||
current_c == acgt[7])
{
gc_count++;
}
else
{
default_count++;
}
}
}
if (calculate)
{
calculate_gc(encseq,
gc_content,
with_special,
seq_idx,
gc_count,
at_count);
}
else
{
gc_content[seq_idx] = (double) gc_count;
}
gt_encseq_reader_delete(reader);
if (is_mirrored_encseq)
{
unsigned long double_max_unit = GT_MULT2(max_unit);
for (seq_idx = 0; seq_idx < max_unit; seq_idx++)
{
gc_content[double_max_unit - seq_idx - 1] =
gc_content[seq_idx];
}
}
return gc_content;
}
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;
}
enum verifyBWTSeqErrCode
gt_BWTSeqVerifyIntegrity(BWTSeq *bwtSeq, const char *projectName,
int checkFlags,
GtUword tickPrint, FILE *fp,
GtLogger *verbosity, GtError *err)
{
Suffixarray suffixArray;
struct extBitsRetrieval extBits;
bool suffixArrayIsInitialized = false, extBitsAreInitialized = false;
enum verifyBWTSeqErrCode retval = VERIFY_BWTSEQ_NO_ERROR;
do
{
GtUword seqLen;
gt_assert(bwtSeq && projectName && err);
gt_error_check(err);
initExtBitsRetrieval(&extBits);
extBitsAreInitialized = true;
if (gt_mapsuffixarray(&suffixArray,
SARR_SUFTAB | SARR_ESQTAB, projectName, verbosity, err))
{
gt_error_set(err, "Cannot load reference suffix array project with"
" demand for suffix table file and encoded sequence"
" for project: %s", projectName);
retval = VERIFY_BWTSEQ_REFLOAD_ERROR;
break;
}
suffixArrayIsInitialized = true;
seqLen = gt_encseq_total_length(suffixArray.encseq) + 1;
if (BWTSeqLength(bwtSeq) != seqLen)
{
gt_error_set(err, "length mismatch for suffix array project %s and "
"bwt sequence index", projectName);
retval = VERIFY_BWTSEQ_LENCOMPARE_ERROR;
break;
}
if (checkFlags & VERIFY_BWTSEQ_SUFVAL
&& BWTSeqHasLocateInformation(bwtSeq))
{
GtUword i;
for (i = 0; i < seqLen && retval == VERIFY_BWTSEQ_NO_ERROR; ++i)
{
if (gt_BWTSeqPosHasLocateInfo(bwtSeq, i, &extBits))
{
GtUword sfxArrayValue = gt_BWTSeqLocateMatch(bwtSeq, i,
&extBits);
if (sfxArrayValue != ESASUFFIXPTRGET(suffixArray.suftab,i))
{
gt_error_set(err, "Failed suffix array value comparison"
" at position "GT_WU": "GT_WU" != "GT_WU"",
i, sfxArrayValue,
ESASUFFIXPTRGET(suffixArray.suftab,i));
retval = VERIFY_BWTSEQ_SUFVAL_ERROR;
break;
}
}
if (tickPrint && !((i + 1) % tickPrint))
putc('.', fp);
}
if (tickPrint)
putc('\n', fp);
if (retval != VERIFY_BWTSEQ_NO_ERROR)
break;
}
else if (checkFlags & VERIFY_BWTSEQ_SUFVAL)
{
gt_error_set(err, "check of suffix array values was requested,"
" but index contains no locate information!");
retval = VERIFY_BWTSEQ_SUFVAL_ERROR;
break;
}
else if (!(checkFlags & VERIFY_BWTSEQ_SUFVAL)
&& BWTSeqHasLocateInformation(bwtSeq))
{
fputs("Not checking suftab values.\n", stderr);
}
if (BWTSeqHasLocateInformation(bwtSeq))
{
GtUword nextLocate = BWTSeqTerminatorPos(bwtSeq);
if (suffixArray.longest.defined &&
suffixArray.longest.valueunsignedlong != nextLocate)
{
gt_error_set(err, "terminator/0-rotation position mismatch "GT_WU""
" vs. "GT_WU"", suffixArray.longest.valueunsignedlong,
nextLocate);
retval = VERIFY_BWTSEQ_TERMPOS_ERROR;
break;
}
if ((checkFlags & VERIFY_BWTSEQ_LFMAPWALK)
&& (bwtSeq->featureToggles & BWTReversiblySorted))
{
GtUword i = seqLen;
/* handle first symbol specially because the encseq
* will not return the terminator symbol */
{
Symbol sym = BWTSeqGetSym(bwtSeq, nextLocate);
if (sym != UNDEFBWTCHAR)
{
gt_error_set(err, "symbol mismatch at position "GT_WU": "
"%d vs. reference symbol %d", i - 1, (int)sym,
(int)UNDEFBWTCHAR);
retval = VERIFY_BWTSEQ_LFMAPWALK_ERROR;
break;
}
--i;
nextLocate = BWTSeqLFMap(bwtSeq, nextLocate, &extBits);
}
while (i > 0)
{
Symbol symRef =
gt_encseq_get_encoded_char(suffixArray.encseq,
--i,
suffixArray.readmode);
Symbol symCmp = BWTSeqGetSym(bwtSeq, nextLocate);
if (symCmp != symRef)
{
gt_error_set(err, "symbol mismatch at position "GT_WU": "
"%d vs. reference symbol %d", i, symCmp, symRef);
retval = VERIFY_BWTSEQ_LFMAPWALK_ERROR;
break;
}
nextLocate = BWTSeqLFMap(bwtSeq, nextLocate, &extBits);
}
if (retval != VERIFY_BWTSEQ_NO_ERROR)
break;
}
else if ((checkFlags & VERIFY_BWTSEQ_LFMAPWALK)
&& !(bwtSeq->featureToggles & BWTReversiblySorted))
{
gt_error_set(err, "requested complete backwards regeneration in index"
" without regeneration capability");
retval = VERIFY_BWTSEQ_LFMAPWALK_IMP_ERROR;
break;
}
}
if (checkFlags & VERIFY_BWTSEQ_CONTEXT)
{
BWTSeqContextRetriever *bwtSeqCR =
gt_BWTSeqCRLoad(bwtSeq, projectName, CTX_MAP_ILOG_AUTOSIZE);
if (!bwtSeqCR)
{
gt_error_set(err, "cannot load BWT sequence context access table"
" for project %s", projectName);
retval = VERIFY_BWTSEQ_CONTEXT_LOADFAIL;
break;
}
fputs("Checking context regeneration.\n", stderr);
{
GtUword i, start, subSeqLen,
maxSubSeqLen = MIN(MAX(MIN_CONTEXT_LEN, seqLen/CONTEXT_FRACTION),
MAX_CONTEXT_LEN),
numTries = MIN(MAX_NUM_CONTEXT_CHECKS,
MAX(2, seqLen/CONTEXT_INTERVAL));
Symbol *contextBuf = gt_malloc(sizeof (Symbol) * MAX_CONTEXT_LEN);
GtEncseqReader *esr =
gt_encseq_create_reader_with_readmode(suffixArray.encseq,
suffixArray.readmode,
0);
for (i = 0; i < numTries && retval == VERIFY_BWTSEQ_NO_ERROR; ++i)
{
GtUword j, end, inSubSeqLen;
subSeqLen = random()%maxSubSeqLen + 1;
start = random()%(seqLen - subSeqLen + 1);
end = start + subSeqLen;
inSubSeqLen = subSeqLen - ((end==seqLen)?1:0);
gt_BWTSeqCRAccessSubseq(bwtSeqCR, start, subSeqLen, contextBuf);
gt_encseq_reader_reinit_with_readmode(esr, suffixArray.encseq,
suffixArray.readmode, start);
for (j = 0; j < inSubSeqLen; ++j)
{
Symbol symRef = gt_encseq_reader_next_encoded_char(esr);
Symbol symCmp = contextBuf[j];
if (symCmp != symRef)
{
gt_error_set(err, "symbol mismatch at position "GT_WU": "
"%d vs. reference symbol %d", start + j, (int)symCmp,
(int)symRef);
retval = VERIFY_BWTSEQ_CONTEXT_SYMFAIL;
break;
}
}
while (j < subSeqLen)
{
Symbol symRef = UNDEFBWTCHAR;
Symbol symCmp = contextBuf[j];
if (symCmp != symRef)
{
gt_error_set(err, "symbol mismatch at position "GT_WU": "
"%d vs. reference symbol %d", start + j, (int)symCmp,
(int)symRef);
retval = VERIFY_BWTSEQ_CONTEXT_SYMFAIL;
break;
}
++j;
}
}
if (retval == VERIFY_BWTSEQ_NO_ERROR)
fputs("Context regeneration completed successfully.\n", stderr);
gt_encseq_reader_delete(esr);
gt_free(contextBuf);
}
gt_deleteBWTSeqCR(bwtSeqCR);
}
} while (0);
if (suffixArrayIsInitialized) gt_freesuffixarray(&suffixArray);
if (extBitsAreInitialized) destructExtBitsRetrieval(&extBits);
return retval;
}
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_encseq_bitextract_runner(GT_UNUSED int argc, const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GT_UNUSED GtError *err)
{
GtEncseqBitextractArguments *arguments = tool_arguments;
GtEncseqLoader *el;
GtEncseq *encseq;
int had_err = 0;
bool fwd, it1, GT_UNUSED it2;
char buffer[BUFSIZ];
GtEndofTwobitencoding etbe;
GtEncseqReader *esr;
GtSpecialrangeiterator *sri;
GtRange srng;
GtReadmode rm;
gt_error_check(err);
gt_assert(arguments);
el = gt_encseq_loader_new();
encseq = gt_encseq_loader_load(el, argv[parsed_args], err);
if (!encseq)
had_err = -1;
if (!had_err && arguments->mirror) {
had_err = gt_encseq_mirror(encseq, err);
}
if (!had_err) {
rm = gt_readmode_parse(gt_str_get(arguments->readmode), NULL);
fwd = GT_ISDIRREVERSE(rm) ? false : true;
}
if (!had_err && arguments->bitpos != GT_UNDEF_ULONG) {
if (arguments->bitpos >= gt_encseq_total_length(encseq)) {
gt_error_set(err, "position %lu exceeds encoded sequence length of %lu",
arguments->bitpos, gt_encseq_total_length(encseq));
had_err = -1;
}
if (!had_err) {
unsigned long ret;
esr = gt_encseq_create_reader_with_readmode(encseq, rm,
arguments->bitpos);
ret = gt_encseq_extract2bitencwithtwobitencodingstoppos(&etbe, esr,
encseq,
rm, arguments->bitpos);
gt_bitsequence_tostring(buffer, etbe.tbe);
printf("Twobitencoding %s\n"
"unitsnotspecial %u\n"
"position %lu\n"
"returnvalue %lu\n",
buffer,
etbe.unitsnotspecial,
arguments->bitpos,
ret);
gt_encseq_reader_delete(esr);
}
}
if (!had_err && arguments->stoppos != GT_UNDEF_ULONG) {
if (arguments->stoppos >= gt_encseq_total_length(encseq)) {
gt_error_set(err, "position %lu exceeds encoded sequence length of %lu",
arguments->stoppos, gt_encseq_total_length(encseq));
had_err = -1;
}
if (!had_err) {
esr = gt_encseq_create_reader_with_readmode(encseq, rm, 0);
/* check stoppos stuff */
gt_encseq_reader_reinit_with_readmode(esr, encseq, rm,
arguments->stoppos);
printf("%lu: %lu\n", arguments->stoppos,
gt_getnexttwobitencodingstoppos(fwd, esr));
gt_encseq_reader_delete(esr);
}
}
if (!had_err && arguments->specialranges) {
/* check specialrangeiterator stuff */
if (gt_encseq_has_specialranges(encseq)) {
sri = gt_specialrangeiterator_new(encseq, fwd);
while (true) {
it1 = gt_specialrangeiterator_next(sri, &srng);
if (it1)
printf("%lu:%lu\n", srng.start, srng.end);
else break;
}
gt_specialrangeiterator_delete(sri);
}
}
gt_encseq_delete(encseq);
gt_encseq_loader_delete(el);
return had_err;
}
