static void remap_NoCutList(AjPFile outfile, const AjPTable hittable,
AjBool html, const AjPStr enzymes, AjBool blunt,
AjBool sticky, ajuint sitelen, AjBool commercial,
AjBool ambiguity, AjBool limit,
const AjPTable retable)
{
/* for iterating over hittable */
PValue value;
void **keyarray = NULL; /* array for table */
void **valarray = NULL; /* array for table */
ajint i;
/* list of enzymes that cut */
AjPList cutlist;
AjIList citer; /* iterator for cutlist */
AjPStr cutname = NULL;
AjBool found;
/* for parsing value->iso string */
AjPStrTok tok;
char tokens[] = " ,";
AjPStr code = NULL;
const char *p;
/* for reading in enzymes names */
AjPFile enzfile = NULL;
AjPStr *ea;
ajint ne; /* number of enzymes */
AjBool isall = ajTrue;
/* list of enzymes that don't cut */
AjPList nocutlist;
AjIList niter; /* iterator for nocutlist */
AjPStr nocutname = NULL;
/* count of rejected enzymes not matching criteria */
ajint rejected_count = 0;
EmbPPatRestrict enz;
/* for renaming preferred isoschizomers */
AjPList newlist;
/*
**
** Make a list of enzymes('cutlist') that hit
** including the isoschizomer names
**
*/
ajDebug("Make a list of all enzymes that cut\n");
cutlist = ajListstrNew();
nocutlist = ajListstrNew();
ajTableToarrayKeysValues(hittable, &keyarray, &valarray);
for(i = 0; keyarray[i]; i++)
{
value = (PValue) valarray[i];
cutname = ajStrNew();
ajStrAssignRef(&cutname, keyarray[i]);
ajListstrPushAppend(cutlist, cutname);
/* Add to cutlist all isoschizomers of enzymes that cut */
ajDebug("Add to cutlist all isoschizomers of enzymes that cut\n");
/* start token to parse isoschizomers names */
tok = ajStrTokenNewC(value->iso, tokens);
while(ajStrTokenNextParseC(&tok, tokens, &code))
{
cutname = ajStrNew();
ajStrAssignS(&cutname, code);
ajListstrPushAppend(cutlist, cutname);
}
ajStrTokenDel(&tok);
}
ajStrDel(&code);
AJFREE(keyarray);
AJFREE(valarray);
/*
** Read in list of enzymes ('nocutlist') - either all or
** the input enzyme list.
** Exclude those that don't match the selection criteria - count these.
*/
ajDebug("Read in a list of all input enzyme names\n");
ne = 0;
if(!enzymes)
isall = ajTrue;
else
{
/* get input list of enzymes into ea[] */
ne = ajArrCommaList(enzymes, &ea);
if(ajStrMatchCaseC(ea[0], "all"))
isall = ajTrue;
else
{
isall = ajFalse;
for(i=0; i<ne; ++i)
ajStrRemoveWhite(&ea[i]);
}
}
enzfile = ajDatafileNewInNameC(ENZDATA);
/* push all enzyme names without the required criteria onto nocutlist */
enz = embPatRestrictNew();
while(!ajFileIsEof(enzfile))
{
if(!embPatRestrictReadEntry(enz, enzfile))
continue;
/*
** If user entered explicit enzyme list, then check to see if
** this is one of that explicit list
*/
if(!isall)
{
found = AJFALSE;
for(i=0; i<ne; ++i)
if(ajStrMatchCaseS(ea[i], enz->cod))
{
found = AJTRUE;
break;
}
if(!found) /* not in the explicit list */
continue;
ajDebug("RE %S is in the input explicit list of REs\n", enz->cod);
}
/* ignore ncuts==0 as they are unknown */
if(!enz->ncuts)
{
/* number of cut positions */
ajDebug("RE %S has an unknown number of cut positions\n",
enz->cod);
continue;
}
ajDebug("RE %S has a known number of cut sites\n", enz->cod);
if(enz->len < sitelen)
{
/* recognition site length */
ajDebug("RE %S does not have a long enough recognition site\n",
enz->cod);
rejected_count++;
continue;
}
if(!blunt && enz->blunt)
{
/* blunt/sticky */
ajDebug("RE %S is blunt\n", enz->cod);
rejected_count++;
continue;
}
if(!sticky && !enz->blunt)
{
/* blunt/sticky */
ajDebug("RE %S is sticky\n", enz->cod);
rejected_count++;
continue;
}
/* commercially available enzymes have uppercase patterns */
p = ajStrGetPtr(enz->pat);
/*
** The -commercial qualifier is only used if we are searching
** through 'all' of the REBASE database - if we have specified an
** explicit list of enzymes then they are searched for whether or
** not they are commercially available
*/
if((*p >= 'a' && *p <= 'z') && commercial && isall)
{
ajDebug("RE %S is not commercial\n", enz->cod);
rejected_count++;
continue;
}
if(!ambiguity && remap_Ambiguous(enz->pat)) {
ajDebug("RE %S is ambiguous\n", enz->cod);
rejected_count++;
continue;
}
ajDebug("RE %S matches all required criteria\n", enz->cod);
code = ajStrNew();
ajStrAssignS(&code, enz->cod);
ajListstrPushAppend(nocutlist, code);
}
embPatRestrictDel(&enz);
ajFileClose(&enzfile);
for(i=0; i<ne; ++i)
if(ea[i])
ajStrDel(&ea[i]);
if(ne)
AJFREE(ea);
/*
** Change names of enzymes in the non-cutter list
** to that of preferred (prototype)
** enzyme name so that the isoschizomers of cutters
** will be removed from the
** non-cutter list in the next bit.
** Remove duplicate prototype names.
*/
if(limit)
{
newlist = ajListstrNew();
remap_RenamePreferred(nocutlist, retable, newlist);
ajListstrFreeData(&nocutlist);
nocutlist = newlist;
ajListSortUnique(nocutlist, remap_cmpcase, remap_strdel);
}
/*
** Iterate through the list of input enzymes removing those that are in
** the cutlist.
*/
ajDebug("Remove from the nocutlist all enzymes and isoschizomers "
"that cut\n");
/*
** This steps down both lists at the same time, comparing names and
** iterating to the next name in whichever list whose name compares
** alphabetically before the other. Where a match is found, the
** nocutlist item is deleted.
*/
ajListSort(nocutlist, remap_cmpcase);
ajListSort(cutlist, remap_cmpcase);
citer = ajListIterNewread(cutlist);
niter = ajListIterNew(nocutlist);
/*
while((cutname = (AjPStr)ajListIterGet(citer)) != NULL)
ajDebug("dbg cutname = %S\n", cutname);
*/
nocutname = (AjPStr)ajListIterGet(niter);
cutname = (AjPStr)ajListIterGet(citer);
ajDebug("initial cutname, nocutname: '%S' '%S'\n", cutname, nocutname);
while(nocutname != NULL && cutname != NULL)
{
i = ajStrCmpCaseS(cutname, nocutname);
ajDebug("compare cutname, nocutname: %S %S ", cutname, nocutname);
ajDebug("ajStrCmpCase=%d\n", i);
if(i == 0)
{
/* match - so remove from nocutlist */
ajDebug("ajListstrRemove %S\n", nocutname);
ajListstrIterRemove(niter);
nocutname = (AjPStr)ajListIterGet(niter);
/*
** Don't increment the cutname list pointer here
** - there may be more than one entry in the nocutname
** list with the same name because we have converted
** isoschizomers to their preferred name
*/
/* cutname = (AjPStr)ajListIterGet(citer); */
}
else if(i == -1)
/* cutlist name sorts before nocutlist name */
cutname = (AjPStr)ajListIterGet(citer);
else if(i == 1)
/* nocutlist name sorts before cutlist name */
nocutname = (AjPStr)ajListIterGet(niter);
}
ajListIterDel(&citer);
ajListIterDel(&niter);
ajListstrFreeData(&cutlist);
/* Print the resulting list of those that do not cut*/
ajDebug("Print out the list\n");
/* print the title */
if(html)
ajFmtPrintF(outfile, "<H2>");
ajFmtPrintF(outfile, "\n\n# Enzymes that do not cut\n\n");
if(html)
ajFmtPrintF(outfile, "</H2>\n");
if(html)
ajFmtPrintF(outfile, "<PRE>");
/* ajListSort(nocutlist, ajStrVcmp);*/
niter = ajListIterNewread(nocutlist);
i = 0;
while((nocutname = (AjPStr)ajListIterGet(niter)) != NULL)
{
ajFmtPrintF(outfile, "%-10S", nocutname);
/* new line after every 7 names printed */
if(i++ == 7)
{
ajFmtPrintF(outfile, "\n");
i = 0;
}
}
ajListIterDel(&niter);
/* end the output */
ajFmtPrintF(outfile, "\n");
if(html) {ajFmtPrintF(outfile, "</PRE>\n");}
/*
** Print the count of rejected enzymes
** N.B. This is the count of ALL rejected enzymes including all
** isoschizomers
*/
if(html)
ajFmtPrintF(outfile, "<H2>");
ajFmtPrintF(outfile,
"\n\n# No. of cutting enzymes which do not match the\n"
"# SITELEN, BLUNT, STICKY, COMMERCIAL, AMBIGUOUS criteria\n\n");
if(html)
ajFmtPrintF(outfile, "</H2>\n");
ajFmtPrintF(outfile, "%d\n", rejected_count);
ajDebug("Tidy up\n");
ajListstrFreeData(&nocutlist);
ajListstrFreeData(&cutlist);
return;
}
void ajResourceDel(AjPResource *Presource)
{
AjPResource thys;
AjPReslink reslink;
AjPResquery resqry;
AjPResterm resterm;
AjIList iter;
if(!Presource) return;
if(!(*Presource)) return;
thys = *Presource;
ajStrDel(&thys->Id);
ajListstrFreeData(&thys->Idalt);
ajStrDel(&thys->Acc);
ajStrDel(&thys->Name);
ajStrDel(&thys->Desc);
ajStrDel(&thys->Url);
ajStrDel(&thys->Urllink);
ajStrDel(&thys->Urlrest);
ajStrDel(&thys->Urlsoap);
ajListstrFreeData(&thys->Cat);
iter = ajListIterNewread(thys->Taxon);
while(!ajListIterDone(iter))
{
resterm = ajListIterGet(iter);
ajRestermDel(&resterm);
}
ajListIterDel(&iter);
ajListFree(&thys->Taxon);
iter = ajListIterNewread(thys->Edamdat);
while(!ajListIterDone(iter))
{
resterm = ajListIterGet(iter);
ajRestermDel(&resterm);
}
ajListIterDel(&iter);
ajListFree(&thys->Edamdat);
iter = ajListIterNewread(thys->Edamfmt);
while(!ajListIterDone(iter))
{
resterm = ajListIterGet(iter);
ajRestermDel(&resterm);
}
ajListIterDel(&iter);
ajListFree(&thys->Edamfmt);
iter = ajListIterNewread(thys->Edamid);
while(!ajListIterDone(iter))
{
resterm = ajListIterGet(iter);
ajRestermDel(&resterm);
}
ajListIterDel(&iter);
ajListFree(&thys->Edamid);
iter = ajListIterNewread(thys->Edamtpc);
while(!ajListIterDone(iter))
{
resterm = ajListIterGet(iter);
ajRestermDel(&resterm);
}
ajListIterDel(&iter);
ajListFree(&thys->Edamtpc);
iter = ajListIterNewread(thys->Xref);
while(!ajListIterDone(iter))
{
reslink = ajListIterGet(iter);
ajReslinkDel(&reslink);
}
ajListIterDel(&iter);
ajListFree(&thys->Xref);
iter = ajListIterNewread(thys->Query);
while(!ajListIterDone(iter))
{
resqry = ajListIterGet(iter);
ajResqueryDel(&resqry);
}
ajListIterDel(&iter);
ajListFree(&thys->Query);
ajListstrFreeData(&thys->Example);
ajStrDel(&thys->Db);
ajStrDel(&thys->Setdb);
ajStrDel(&thys->Full);
ajStrDel(&thys->Qry);
ajStrDel(&thys->Formatstr);
ajStrDel(&thys->Filename);
ajStrDel(&thys->TextPtr);
AJFREE(*Presource);
*Presource = NULL;
return;
}
int main(int argc, char **argv)
{
AjPList idlist;
AjPList* fieldList = NULL;
AjBool systemsort;
AjBool cleanup;
ajint blastv = 0;
char dbtype = '\0';
ajuint maxindex;
ajuint maxidlen = 0;
ajuint maxlen;
AjPStr version = NULL;
AjPStr seqtype = NULL;
AjPFile elistfile = NULL;
AjPFile* alistfile = NULL;
AjPStr dbname = NULL;
AjPStr release = NULL;
AjPStr datestr = NULL;
AjPStr sortopt = NULL;
void **entryIds = NULL;
AjBool usesrc = AJTRUE;
AjPStr directory;
AjPStr indexdir;
AjPStr filename;
AjPStr exclude;
AjPStr curfilename = NULL;
AjPStr idformat = NULL;
EmbPEntry entry;
PBlastDb db = NULL;
ajuint idCount = 0;
ajuint idDone;
AjPList listTestFiles = NULL;
void ** testFiles = NULL;
ajuint nfiles;
ajuint ifile;
ajuint jfile;
ajuint filesize;
short recsize;
ajuint maxfilelen = 20;
char date[4] =
{
0,0,0,0
};
AjPStr tmpfname = NULL;
AjPStr* fields = NULL;
AjPFile entFile = NULL;
AjPStr* divfiles = NULL;
ajint* maxFieldLen = NULL;
ajuint ifield = 0;
ajuint nfields = 0;
AjPFile logfile = NULL;
ajuint* countField = NULL;
ajuint* fieldTot = NULL;
ajuint idCountFile = 0;
ajuint i = 0;
embInit("dbiblast", argc, argv);
idformat = ajStrNewC("NCBI");
fields = ajAcdGetList("fields");
directory = ajAcdGetDirectoryName("directory");
indexdir = ajAcdGetOutdirName("indexoutdir");
filename = ajAcdGetString("filenames");
exclude = ajAcdGetString("exclude");
dbname = ajAcdGetString("dbname");
release = ajAcdGetString("release");
datestr = ajAcdGetString("date");
systemsort = ajAcdGetBoolean("systemsort");
cleanup = ajAcdGetBoolean("cleanup");
sortopt = ajAcdGetString("sortoptions");
maxindex = ajAcdGetInt("maxindex");
version = ajAcdGetListSingle("blastversion");
seqtype = ajAcdGetListSingle("seqtype");
usesrc = ajAcdGetBoolean("sourcefile");
logfile = ajAcdGetOutfile("outfile");
while(fields[nfields]) /* array ends with a NULL */
nfields++;
if(nfields)
{
AJCNEW(maxFieldLen, nfields);
AJCNEW0(countField, nfields);
AJCNEW0(fieldTot, nfields);
for(ifield=0; ifield < nfields; ifield++)
maxFieldLen[ifield] = (ajint) maxindex * -1;
if(systemsort)
AJCNEW(alistfile, nfields);
else
{
AJCNEW(fieldList, nfields);
for(ifield=0; ifield < nfields; ifield++)
fieldList[ifield] = ajListNew();
}
}
if(ajStrMatchC(datestr, "00/00/00"))
ajFmtPrintS(&datestr, "%D", ajTimeRefTodayFmt("dbindex"));
ajStrRemoveWhite(&dbname); /* used for temp filenames */
embDbiDateSet(datestr, date);
idlist = ajListNew();
if(ajUtilGetBigendian())
readReverse = ajFalse;
else
readReverse = ajTrue;
ajStrToInt(version, &blastv);
dbtype = ajStrGetCharFirst(seqtype);
ajDebug("reading '%S/%S'\n", directory, filename);
ajDebug("writing '%S/'\n", indexdir);
listTestFiles = embDbiFileListExc(directory, filename, exclude);
ajListSort(listTestFiles, ajStrVcmp);
nfiles = ajListToarray(listTestFiles, &testFiles);
if(!nfiles)
ajDie("No input files in '%S' matched filename '%S'",
directory, filename);
embDbiLogHeader(logfile, dbname, release, datestr,
indexdir, maxindex);
embDbiLogFields(logfile, fields, nfields);
embDbiLogSource(logfile, directory, filename, exclude,
(AjPStr*) testFiles, nfiles);
embDbiLogCmdline(logfile);
AJCNEW0(divfiles, nfiles);
/*
** process each input file, one at a time
*/
jfile = 0;
for(ifile=0; ifile < nfiles; ifile++)
{
curfilename = (AjPStr) testFiles[ifile];
if(!dbiblast_blastopenlib(curfilename,
usesrc, blastv, dbtype, &db))
continue; /* could be the wrong file type with "*.*" */
ajDebug("processing filename '%S' ...\n", curfilename);
ajDebug("processing file '%S' ...\n", db->TFile->Name);
ajStrAssignS(&divfiles[jfile], db->TFile->Name);
ajFilenameTrimPath(&divfiles[jfile]);
if(ajStrGetLen(divfiles[jfile]) >= maxfilelen)
maxfilelen = ajStrGetLen(divfiles[jfile]) + 1;
if(systemsort) /* elistfile for entries, alist for fields */
elistfile = embDbiSortOpen(alistfile, jfile,
dbname, fields, nfields);
idCountFile = 0;
for(i=0;i<nfields;i++)
countField[i] = 0;
while((entry=dbiblast_nextblastentry(db, jfile,
idformat, systemsort,
fields,
maxFieldLen,
&maxidlen, countField,
elistfile, alistfile)))
{
idCountFile++;
if(!systemsort) /* save the entry data in lists */
{
embDbiMemEntry(idlist, fieldList, nfields, entry, jfile);
}
}
idCount += idCountFile;
if(systemsort)
{
embDbiSortClose(&elistfile, alistfile, nfields);
/* lost the entry, so can't free it :-) */
}
embDbiLogFile(logfile, curfilename, idCountFile, fields,
countField, nfields);
dbiblast_dbfree(&db);
jfile++;
}
nfiles = jfile;
/*
** write the division.lkp file
*/
embDbiWriteDivision(indexdir, dbname, release, date,
maxfilelen, nfiles, divfiles, NULL);
/*
** Write the entryname.idx index
*/
ajStrAssignC(&tmpfname, "entrynam.idx");
entFile = ajFileNewOutNamePathS(tmpfname, indexdir);
recsize = maxidlen+10;
filesize = 300 + (idCount*(ajint)recsize);
embDbiHeader(entFile, filesize, idCount, recsize, dbname, release, date);
if(systemsort)
idDone = embDbiSortWriteEntry(entFile, maxidlen,
dbname, nfiles, cleanup, sortopt);
else /* save entries in entryIds array */
{
idDone = embDbiMemWriteEntry(entFile, maxidlen,
idlist, &entryIds);
if(idDone != idCount)
ajFatal("Duplicates not allowed for in-memory processing");
}
embDbiHeaderSize(entFile, 300+(idDone*(ajint)recsize), idDone);
ajFileClose(&entFile);
/*
** Write the fields index files
*/
for(ifield=0; ifield < nfields; ifield++)
{
if(maxindex)
maxlen = maxindex;
else
{
if(maxFieldLen[ifield] >= 0)
maxlen = maxFieldLen[ifield];
else
maxlen = - maxFieldLen[ifield];
}
if(systemsort)
fieldTot[ifield] = embDbiSortWriteFields(dbname, release,
date, indexdir,
fields[ifield], maxlen,
nfiles, idCount,
cleanup, sortopt);
else
fieldTot[ifield] = embDbiMemWriteFields(dbname, release,
date, indexdir,
fields[ifield], maxlen,
fieldList[ifield],
entryIds);
}
embDbiLogFinal(logfile,maxindex, maxFieldLen, fields, fieldTot,
nfields, nfiles, idDone, idCount);
if(systemsort)
embDbiRmEntryFile(dbname, cleanup);
ajListMap(idlist, embDbiEntryDelMap, NULL);
ajListFree(&idlist);
AJFREE(entryIds);
ajStrDelarray(&fields);
for(i=0;i<nfields;i++)
{
if(systemsort)
{
ajFileClose(&alistfile[i]);
}
else
{
ajListMap(fieldList[i], embDbiFieldDelMap, NULL);
ajListFree(&fieldList[i]);
}
}
AJFREE(alistfile);
AJFREE(fieldList);
ajStrDel(&version);
ajStrDel(&seqtype);
ajFileClose(&elistfile);
for(i=0;i<nfiles;i++)
{
ajStrDel(&divfiles[i]);
}
AJFREE(countField);
AJFREE(fieldTot);
ajStrDel(&dbname);
ajStrDel(&release);
ajStrDel(&datestr);
ajStrDel(&sortopt);
ajStrDel(&directory);
ajStrDel(&indexdir);
ajStrDel(&filename);
ajStrDel(&exclude);
ajStrDel(&idformat);
ajStrDel(&tmpfname);
AJFREE(maxFieldLen);
ajFileClose(&logfile);
ajListstrFreeData(&listTestFiles);
ajStrDel(&t);
ajStrDel(&id);
ajStrDel(&acc);
ajStrDel(&hline);
ajStrDel(&tmpdes);
ajStrDel(&tmpfd);
ajStrDel(&tmpgi);
ajStrDel(&tmpdb);
ajStrDel(&tmpac);
ajStrDel(&tmpsv);
ajRegFree(&wrdexp);
embDbiEntryDel(&dbiblastEntry);
if(fdl)
{
for(i=0; i < nfields; i++)
ajListFree(&fdl[i]);
AJFREE(fdl);
}
for(i=0;i<nfiles;i++)
{
ajStrDel(&divfiles[i]);
}
AJFREE(divfiles);
AJFREE(testFiles);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPList idlist;
AjPList* fieldList = NULL;
AjBool systemsort;
AjBool cleanup;
ajuint maxindex;
ajuint maxidlen = 0;
ajuint maxlen;
AjPFile elistfile = NULL;
AjPFile* alistfile = NULL;
AjPStr dbname = NULL;
AjPStr release = NULL;
AjPStr datestr = NULL;
AjPStr sortopt = NULL;
void **entryIds = NULL;
AjPStr directory;
AjPStr indexdir;
AjPStr filename;
AjPStr exclude;
AjPStr curfilename = NULL;
AjPFile libr=NULL;
AjPStr idformat = NULL;
EmbPEntry entry;
ajuint idtype = 0;
ajuint idCount = 0;
ajuint idDone;
AjPList listInputFiles = NULL;
void ** inputFiles = NULL;
ajuint nfiles;
ajuint ifile;
ajuint filesize;
short recsize;
ajuint maxfilelen = 20;
char date[4] =
{
0,0,0,0
};
AjPStr tmpfname = NULL;
AjPStr* fields = NULL;
AjPFile entFile = NULL;
AjPStr* divfiles = NULL;
AjPRegexp regIdExp = NULL;
ajint* maxFieldLen = NULL;
ajuint ifield = 0;
ajuint nfields = 0;
AjPFile logfile = NULL;
ajuint* countField = NULL;
ajuint* fieldTot = NULL;
ajuint idCountFile = 0;
ajuint i;
embInit("dbifasta", argc, argv);
idformat = ajAcdGetListSingle("idformat");
fields = ajAcdGetList("fields");
directory = ajAcdGetDirectoryName("directory");
indexdir = ajAcdGetOutdirName("indexoutdir");
filename = ajAcdGetString("filenames");
exclude = ajAcdGetString("exclude");
dbname = ajAcdGetString("dbname");
release = ajAcdGetString("release");
datestr = ajAcdGetString("date");
systemsort = ajAcdGetBoolean("systemsort");
cleanup = ajAcdGetBoolean("cleanup");
sortopt = ajAcdGetString("sortoptions");
maxindex = ajAcdGetInt("maxindex");
logfile = ajAcdGetOutfile("outfile");
while(fields[nfields]) /* array ends with a NULL */
nfields++;
if(nfields)
{
AJCNEW(maxFieldLen, nfields);
AJCNEW0(countField, nfields);
AJCNEW0(fieldTot, nfields);
for(ifield=0; ifield < nfields; ifield++)
maxFieldLen[ifield] = (ajint)maxindex * -1;
if(systemsort)
AJCNEW(alistfile, nfields);
else
{
AJCNEW(fieldList, nfields);
for(ifield=0; ifield < nfields; ifield++)
fieldList[ifield] = ajListNew();
}
}
if(ajStrMatchC(datestr, "00/00/00"))
ajFmtPrintS(&datestr, "%D", ajTimeRefTodayFmt("dbindex"));
ajStrRemoveWhite(&dbname); /* used for temp filenames */
embDbiDateSet(datestr, date);
idlist = ajListNew();
regIdExp = dbifasta_getExpr(idformat, &idtype);
ajDebug("reading '%S/%S'\n", directory, filename);
ajDebug("writing '%S/'\n", indexdir);
listInputFiles = embDbiFileListExc(directory, filename, exclude);
ajListSort(listInputFiles, &ajStrVcmp);
nfiles = (ajuint) ajListToarray(listInputFiles, &inputFiles);
if(!nfiles)
ajDie("No input files in '%S' matched filename '%S'",
directory, filename);
embDbiLogHeader(logfile, dbname, release, datestr,
indexdir, maxindex);
embDbiLogFields(logfile, fields, nfields);
embDbiLogSource(logfile, directory, filename, exclude,
(AjPStr*) inputFiles, nfiles);
embDbiLogCmdline(logfile);
AJCNEW0(divfiles, nfiles);
/*
** process each input file, one at a time
*/
for(ifile=0; ifile < nfiles; ifile++)
{
ajStrAssignS(&curfilename,(AjPStr) inputFiles[ifile]);
embDbiFlatOpenlib(curfilename, &libr);
ajFilenameTrimPath(&curfilename);
if(ajStrGetLen(curfilename) >= maxfilelen)
maxfilelen = ajStrGetLen(curfilename) + 1;
ajDebug("processing filename '%S' ...\n", curfilename);
ajDebug("processing file '%F' ...\n", libr);
ajStrAssignS(&divfiles[ifile], curfilename);
if(systemsort) /* elistfile for entries, alist for fields */
elistfile = embDbiSortOpen(alistfile, ifile,
dbname, fields, nfields);
idCountFile = 0;
for(i=0;i<nfields;i++)
countField[i] = 0;
while((entry=dbifasta_NextFlatEntry(libr, ifile,
regIdExp, idtype,
systemsort, fields,
maxFieldLen, &maxidlen,
countField, elistfile,
alistfile)))
{
idCountFile++;
if(!systemsort) /* save the entry data in lists */
embDbiMemEntry(idlist, fieldList, nfields,
entry, ifile);
entry = NULL;
}
idCount += idCountFile;
if(systemsort)
{
embDbiSortClose(&elistfile, alistfile, nfields);
AJFREE(entry);
}
else
{
embDbiEntryDel(&dbifastaGEntry);
}
embDbiLogFile(logfile, curfilename, idCountFile, fields,
countField, nfields);
}
/* write the division.lkp file */
embDbiWriteDivision(indexdir, dbname, release, date,
maxfilelen, nfiles, divfiles, NULL);
/* Write the entryname.idx index */
ajStrAssignC(&tmpfname, "entrynam.idx");
entFile = ajFileNewOutNamePathS(tmpfname, indexdir);
recsize = maxidlen+10;
filesize = 300 + (idCount*(ajint)recsize);
embDbiHeader(entFile, filesize, idCount, recsize, dbname, release, date);
if(systemsort)
idDone = embDbiSortWriteEntry(entFile, maxidlen,
dbname, nfiles, cleanup, sortopt);
else /* save entries in entryIds array */
{
idDone = embDbiMemWriteEntry(entFile, maxidlen,
idlist, &entryIds);
if(idDone != idCount)
ajFatal("Duplicates not allowed for in-memory processing");
}
embDbiHeaderSize(entFile, 300+(idDone*(ajint)recsize), idDone);
ajFileClose(&entFile);
/* Write the fields index files */
for(ifield=0; ifield < nfields; ifield++)
{
if(maxindex)
maxlen = maxindex;
else
{
if(maxFieldLen[ifield] >= 0)
maxlen = maxFieldLen[ifield];
else
maxlen = - maxFieldLen[ifield];
}
if(systemsort)
fieldTot[ifield] = embDbiSortWriteFields(dbname, release,
date, indexdir,
fields[ifield], maxlen,
nfiles, idCount,
cleanup, sortopt);
else
fieldTot[ifield] = embDbiMemWriteFields(dbname, release,
date, indexdir,
fields[ifield], maxlen,
fieldList[ifield],
entryIds);
}
embDbiLogFinal(logfile,maxindex, maxFieldLen, fields, fieldTot,
nfields, nfiles, idDone, idCount);
if(systemsort)
embDbiRmEntryFile(dbname, cleanup);
ajStrDel(&idformat);
ajStrDelarray(&fields);
ajStrDel(&filename);
ajStrDel(&exclude);
ajStrDel(&dbname);
ajStrDel(&release);
ajStrDel(&datestr);
ajStrDel(&sortopt);
ajStrDel(&directory);
ajStrDel(&indexdir);
ajStrDel(&tmpfname);
ajFileClose(&libr);
ajFileClose(&logfile);
for(i=0;i<nfields;i++)
{
if(systemsort)
{
ajFileClose(&alistfile[i]);
}
else
{
ajListMap(fieldList[i], &embDbiFieldDelMap, NULL);
ajListFree(&fieldList[i]);
}
}
AJFREE(alistfile);
AJFREE(fieldList);
AJFREE(maxFieldLen);
AJFREE(countField);
AJFREE(fieldTot);
for(i=0;i<nfiles;i++)
{
ajStrDel(&divfiles[i]);
}
AJFREE(divfiles);
AJFREE(inputFiles);
embDbiEntryDel(&dbifastaGEntry);
ajStrDel(&dbifastaGRline);
ajStrDel(&dbifastaGTmpId);
if(dbifastaGFdl)
{
for(i=0; i < nfields; i++)
ajListFree(&dbifastaGFdl[i]);
AJFREE(dbifastaGFdl);
}
ajListMap(idlist, &embDbiEntryDelMap, NULL);
ajListFree(&idlist);
ajListstrFreeData(&listInputFiles);
AJFREE(entryIds);
ajRegFree(&dbifastaGIdexp);
ajRegFree(&dbifastaGWrdexp);
ajRegFree(®IdExp);
ajStrDel(&dbifastaGTmpAc);
ajStrDel(&dbifastaGTmpSv);
ajStrDel(&dbifastaGTmpGi);
ajStrDel(&dbifastaGTmpDb);
ajStrDel(&dbifastaGTmpDes);
ajStrDel(&dbifastaGTmpFd);
ajStrDel(&curfilename);
embExit();
return 0;
}
/* @funcstatic acdrelations_procacdfile ***************************************
**
** Process ACD file and write new ACD file with new relations: attributes
** added (replaced if necessary).
**
** @param [r] inf [AjPFile] ACD input file
** @param [r] outf [AjPFile] ACD output file
** @param [r] P [PEdam] edam object
** @param [r] T [PKtype] ktype object
** @return [void]
** @@
******************************************************************************/
static void acdrelations_procacdfile
(AjPFile inf,
AjPFile outf,
PEdam P,
PKtype T)
{
AjPStr line = NULL;
AjPStr tok = NULL;
AjPStr acdtype = NULL;
AjPStr strtmp = NULL;
AjPList strlist = NULL;
AjPStr *strarr = NULL;
ajint nstr = 0;
/* Allocate memory */
line = ajStrNew();
tok = ajStrNew();
acdtype = ajStrNew();
strlist = ajListstrNew();
/* Read next line */
while(ajReadline(inf,&line))
{
ajFmtScanS(line, "%S", &tok);
/* Write application definition or section definition out as-is */
if(ajStrMatchC(tok, "application:") ||
ajStrMatchC(tok, "section:"))
{
ajFmtPrintF(outf, "%S", line);
while(ajReadline(inf,&line))
{
ajFmtPrintF(outf, "%S", line);
ajFmtScanS(line, "%S", &tok);
if(ajStrMatchC(tok, "]"))
break;
}
}
/* Write variables, endsection definitions and comments out as-is */
else if(ajStrMatchC(tok, "variable:") ||
ajStrMatchC(tok, "endsection:") ||
ajStrMatchC(tok, "#"))
ajFmtPrintF(outf, "%S", line);
/* Write out blank lines as-is */
else if (!ajFmtScanS(line, "%S", &tok))
ajFmtPrintF(outf, "%S", line);
/* Process data definition */
else
/* First line of data definition */
{
/* Process and write datatype line */
ajFmtPrintF(outf, "%S", line);
ajFmtScanS(line, "%S", &acdtype);
ajStrRemoveSetC(&acdtype, ":");
/* Process subsequent (attribute) lines */
while(ajReadline(inf,&line))
{
strtmp = ajStrNew();
ajStrAssignS(&strtmp, line);
ajStrRemoveWhite(&strtmp);
ajListstrPushAppend(strlist, strtmp);
ajFmtScanS(line, "%S", &tok);
/* Reached end of data definition */
if(ajStrMatchC(tok, "]"))
{
nstr = ajListstrToarray(strlist, &strarr);
/* Write relations: line */
acdrelations_writerelations(outf, acdtype, strarr, nstr, P, T);
AJFREE(strarr);
ajListstrFreeData(&strlist);
strlist = ajListstrNew();
ajFmtPrintF(outf, "%S", line);
break;
}
/* Ignore existing relations: lines */
else if(ajStrMatchC(tok, "relations:"))
continue;
ajFmtPrintF(outf, "%S", line);
}
}
}
/* Free memory */
ajStrDel(&line);
ajStrDel(&tok);
ajStrDel(&acdtype);
ajListstrFreeData(&strlist);
return;
}
int main(int argc, char **argv)
{
AjPSeqout outseq = NULL;
AjPList list = NULL;
AjPSeq seq = NULL;
AjPStr insert = NULL;
AjPStr seqstr = NULL;
AjPStr* seqr = NULL;
AjPFile data = NULL;
ajint start = 0;
ajint length = 0;
ajint amount = 0;
ajint scmax = 0;
ajint extra = 0;
embInit("makeprotseq", argc, argv);
data = ajAcdGetInfile("pepstatsfile");
insert = ajAcdGetString("insert");
start = ajAcdGetInt("start");
length = ajAcdGetInt("length");
amount = ajAcdGetInt("amount");
outseq = ajAcdGetSeqoutall("outseq");
list = ajListstrNew();
/* this is checked by acd
if(amount <=0 || length <= 0)
ajFatal("Amount or length is 0 or less. "
"Unable to create any sequences"); */
/* if insert, make sure sequence is large enough */
if(ajStrGetLen(insert))
{
length -= ajStrGetLen(insert);
/* start= start <= 1 ? 0 : --start; */ /* checked in acd */
start--;
if(length <= 0)
ajFatal("Sequence smaller than inserted part. "
"Unable to create sequences.");
}
/* make the list of AjPStr to be used in sequence creation */
if(data)
{
ajDebug("Distribution datafile '%s' given checking type\n",
ajFileGetPrintnameC(data));
seqstr = ajStrNew();
ajReadlineTrim(data,&seqstr);
if(ajStrFindC(seqstr,"PEPSTATS") == 0)
{
makeprotseq_parse_pepstats(&list,data);
}
else
{
ajWarn("Not pepstats file. Making completely random sequences.");
makeprotseq_default_chars(&list);
}
ajStrDel(&seqstr);
ajFileClose(&data);
}
else
makeprotseq_default_chars(&list);
/* if insert, make sure type is correct */
/* typecheking code is not working, uncomment and test after it is
if(ajStrGetLen(insert))
{
seqstr = ajStrNew();
if(prot)
ajStrAssignC(&seqstr,"pureprotein");
if(!ajSeqTypeCheckS(&insert,seqstr))
ajFatal("Insert not the same sequence type as sequence itself.");
ajStrDel(&seqstr);
} */
/* array allows fast creation of a sequences */
scmax = (ajuint) ajListstrToarray(list,&seqr);
if(!scmax)
ajFatal("No strings in list. No characters to make the sequence.");
ajDebug("Distribution array done.\nscmax '%d', extra '%d', first '%S'\n",
scmax,extra,seqr[0]);
ajRandomSeed();
while(amount-- > 0)
{
seqstr = makeprotseq_random_sequence(seqr,scmax,length);
if(ajStrGetLen(insert))
ajStrInsertS(&seqstr,start,insert);
ajStrFmtLower(&seqstr);
seq = ajSeqNew();
ajSeqAssignSeqS(seq, seqstr);
ajSeqSetProt(seq);
ajSeqoutWriteSeq(outseq, seq);
ajSeqDel(&seq);
ajStrDel(&seqstr);
}
ajSeqoutClose(outseq);
ajSeqoutDel(&outseq);
ajListstrFreeData(&list);
ajStrDel(&insert);
AJFREE(seqr);
embExit();
return 0;
}
