AjBool ensListUintSortDescendingUnique(AjPList list)
{
if (!list)
return ajFalse;
ajListSortUnique(list,
&listUintCompareDescending,
&listUintDelete);
return ajTrue;
}
static void jaspscan_ParseInput(const AjPStr dir, const AjPStr jaspdir,
const AjPStr mats, const AjPStr excl,
ajuint *recurs, AjPList ret)
{
ajuint nm = 0;
ajuint ne = 0;
AjPStr *carr = NULL;
AjPStr *earr = NULL;
AjPFile inf = NULL;
AjPStr line = NULL;
AjPStr comm = NULL;
AjPStr val = NULL;
ajuint i;
ajuint j;
char c;
ajuint rlen = 0;
if(*recurs > JASPSCAN_RECURS)
ajFatal("Too many recursion levels in matrix list files");
line = ajStrNew();
comm = ajStrNew();
if(mats)
{
nm = ajArrCommaList(mats,&carr);
for(i=0; i < nm; ++i)
{
if(ajStrGetCharFirst(carr[i]) != '@')
ajStrFmtUpper(&carr[i]);
if(ajStrMatchC(carr[i],"ALL"))
{
jaspscan_GetFileList(dir, jaspdir, "*", ret);
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
}
else if(ajStrGetCharFirst(carr[i]) == '@')
{
ajStrTrimStartC(&carr[i],"@");
inf = ajFileNewInNameS(carr[i]);
if(!inf)
ajFatal("Cannot open list file %S",carr[i]);
while(ajReadlineTrim(inf,&line))
{
ajStrRemoveWhite(&line);
c = ajStrGetCharFirst(line);
if(c == '#' || c== '!')
continue;
if(ajStrGetLen(comm))
ajStrAppendC(&comm,",");
ajStrFmtUpper(&line);
ajStrAppendS(&comm,line);
}
*recurs += 1;
jaspscan_ParseInput(dir,jaspdir,comm,NULL,recurs,ret);
*recurs -= 1;
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
ajFileClose(&inf);
}
else
{
jaspscan_GetFileList(dir,jaspdir,ajStrGetPtr(carr[i]),ret);
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
}
}
for(i=0; i < nm; ++i)
ajStrDel(&carr[i]);
AJFREE(carr);
}
if(excl)
{
ne = ajArrCommaList(excl,&earr);
for(i=0; i < ne; ++i)
{
if(ajStrGetCharFirst(earr[i]) != '@')
ajStrFmtUpper(&earr[i]);
if(ajStrGetCharFirst(earr[i]) == '@')
{
ajStrTrimStartC(&earr[i],"@");
inf = ajFileNewInNameS(earr[i]);
if(!inf)
ajFatal("Cannot open list file %S",earr[i]);
while(ajReadlineTrim(inf,&line))
{
ajStrRemoveWhite(&line);
c = ajStrGetCharFirst(line);
if(c == '#' || c== '!')
continue;
if(ajStrGetLen(comm))
ajStrAppendC(&comm,",");
ajStrFmtUpper(&line);
ajStrAppendS(&comm,line);
}
*recurs += 1;
jaspscan_ParseInput(dir,jaspdir,NULL,comm,recurs,ret);
*recurs -= 1;
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
ajFileClose(&inf);
}
else
{
ajStrAssignS(&line,earr[i]);
ajStrAppendC(&line,J_EXT);
rlen = ajListGetLength(ret);
for(j=0; j < rlen; ++j)
{
ajListPop(ret,(void **)&val);
if(ajStrSuffixS(val,line))
ajStrDel(&val);
else
ajListPushAppend(ret,(void *)val);
}
}
}
for(i=0; i < ne; ++i)
ajStrDel(&earr[i]);
AJFREE(earr);
}
ajStrDel(&line);
ajStrDel(&comm);
return;
}
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;
}
int main(int argc, char** argv)
{
AjPFile outf = NULL;
AjPFile cachef = NULL;
AjIList iterator = NULL;
AjPList aliases = NULL;
AjPList dbas = NULL;
AjPList species = NULL;
AjPStr alias = NULL;
AjPStr dbname = NULL;
AjPStr spname = NULL;
AjPStr svrname = NULL;
AjPStr svrurl = NULL;
AjPStr dbcurl = NULL;
AjPTime svrtime = NULL;
EnsEDatabaseadaptorGroup dbag = ensEDatabaseadaptorGroupNULL;
EnsPDatabaseadaptor dba = NULL;
EnsPDatabaseconnection dbc = NULL;
embInit("cacheensembl", argc, argv);
ensInit();
svrname = ajAcdGetString("servername");
outf = ajAcdGetOutfile("outfile");
cachef = ajAcdGetOutfile("cachefile");
dbcurl = ajStrNew();
svrurl = ajStrNew();
dbname = ajStrNew();
ajNamSvrGetUrl(svrname, &svrurl);
if(!svrurl)
ajFatal("Could not resolve server name '%S'.", svrname);
dbc = ensDatabaseconnectionNewUrl(svrurl);
ensRegistryLoadDatabaseconnection(dbc);
ensDatabaseconnectionDel(&dbc);
/* Write the server file header. */
svrtime = ajTimeNewTodayFmt("cachefile");
ajFmtPrintF(cachef, "# %S %D\n", ajFileGetNameS(cachef), svrtime);
ajTimeDel(&svrtime);
ajFmtPrintF(cachef,
"# Automatically generated by cacheensembl "
"for server '%S'.\n\n",
svrname);
/*
** Get all Ensembl Database Adaptor objects and write them as
** EMBOSS Database definitions.
*/
aliases = ajListstrNew();
dbas = ajListNew();
species = ajListstrNew();
ensRegistryRetrieveAllSpecies(species);
while(ajListstrPop(species, &spname))
{
ensRegistryGetAllDatabaseadaptors(ensEDatabaseadaptorGroupNULL,
spname,
dbas);
while(ajListPop(dbas, (void**) &dba))
{
dbag = ensDatabaseadaptorGetGroup(dba);
if(dbag == ensEDatabaseadaptorGroupNULL)
{
ajDebug("cacheensembl main got unexpected "
"Ensembl Database Adaptor Group %d.\n",
dbag);
continue;
}
ajStrAssignS(&dbname, ensDatabaseadaptorGetSpecies(dba));
if(dbag != ensEDatabaseadaptorGroupCore)
{
ajStrAppendC(&dbname, "_");
ajStrAppendC(&dbname, ensDatabaseadaptorGroupToChar(dbag));
}
dbc = ensDatabaseadaptorGetDatabaseconnection(dba);
ensDatabaseconnectionFetchUrl(dbc, &dbcurl);
if(outf)
ajFmtPrintF(outf, "%S\n", dbname);
ajFmtPrintF(cachef, "DBNAME %S [\n", dbname);
ajFmtPrintF(cachef, " release: \"%s\"\n", ensSoftwareGetVersion());
ajFmtPrintF(cachef, " server: \"%S\"\n", svrname);
ajFmtPrintF(cachef, " url: \"%S\"\n", dbcurl);
ajFmtPrintF(cachef, "]\n");
ajFmtPrintF(cachef, "\n");
if(dbag != ensEDatabaseadaptorGroupCore)
continue;
ensRegistryAliasFetchAllbySpecies(
ensDatabaseadaptorGetSpecies(dba),
aliases);
/*
** Format all aliases to lower case,
** sort them alphabetically and remove duplicates.
*/
iterator = ajListIterNew(aliases);
while(!ajListIterDone(iterator))
{
alias = ajListstrIterGet(iterator);
ajStrFmtLower(&alias);
}
ajListIterDel(&iterator);
ajListSortUnique(aliases,
cacheensembl_stringcompare,
cacheensembl_stringdelete);
alias = NULL;
if(ajListGetLength(aliases) > 0)
{
while(ajListstrPop(aliases, &alias))
{
/*
** Reject any aliases with other than alpha-numeric
** characters like white space.
*/
if(ajStrIsAlnum(alias))
ajFmtPrintF(cachef,
"ALIAS %S %S\n",
alias,
ensDatabaseadaptorGetSpecies(dba));
ajStrDel(&alias);
}
ajFmtPrintF(cachef, "\n");
}
/* Ensembl Database Adaptor objects *must not* be deleted. */
}
ajStrDel(&spname);
}
ajListstrFree(&aliases);
ajListFree(&dbas);
ajStrDel(&dbcurl);
ajStrDel(&svrurl);
ajStrDel(&dbname);
ajStrDel(&svrname);
ajFileClose(&outf);
ajFileClose(&cachef);
embExit();
return EXIT_SUCCESS;
}
