static String HHVM_STATIC_METHOD(Locale, lookup, CArrRef langtag,
const String& locale,
bool canonicalize, const String& def) {
String locname(localeOrDefault(locale), CopyString);
std::vector<std::pair<String,String>> cur_arr;
for (ArrayIter iter(langtag); iter; ++iter) {
auto val = iter.second();
if (!val.isString()) {
s_intl_error->setError(U_ILLEGAL_ARGUMENT_ERROR, "lookup_loc_range: "
"locale array element is not a string");
return def;
}
String normalized(val.toString(), CopyString);
normalize_for_match(normalized);
if (canonicalize) {
normalized = get_icu_value(normalized, LOC_CANONICALIZE);
if (normalized.isNull()) {
s_intl_error->setError(U_ILLEGAL_ARGUMENT_ERROR, "lookup_loc_range: "
"unable to canonicalize lang_tag");
return def;
}
normalize_for_match(normalized);
}
cur_arr.push_back(std::make_pair(val,normalized));
}
if (canonicalize) {
locname = get_icu_value(locname, LOC_CANONICALIZE);
if (locname.isNull()) {
s_intl_error->setError(U_ILLEGAL_ARGUMENT_ERROR, "lookup_loc_range: "
"unable to canonicalize loc_range");
return def;
}
}
normalize_for_match(locname);
while (locname.size() > 0) {
for (auto &p : cur_arr) {
if (locname.same(p.second)) {
return canonicalize ? p.second : p.first;
}
}
locname = locale_suffix_strip(locname);
}
return def;
}
static Variant get_icu_display_value(const String &locale,
const String &disp_locale,
LocaleTag tag) {
String locname(locale);
if (tag != LOC_DISPLAY) {
int ofs = getGrandfatheredOffset(locale);
if (ofs >= 0) {
if (tag == LOC_LANG) {
locname = getGrandfatheredPreferred(ofs);
} else {
return false;
}
}
}
int32_t (*ulocfunc)(const char *loc, const char *dloc,
UChar *dest, int32_t destcap, UErrorCode *err);
switch (tag) {
case LOC_LANG: ulocfunc = uloc_getDisplayLanguage; break;
case LOC_SCRIPT: ulocfunc = uloc_getDisplayScript; break;
case LOC_REGION: ulocfunc = uloc_getDisplayCountry; break;
case LOC_VARIANT: ulocfunc = uloc_getDisplayVariant; break;
case LOC_DISPLAY: ulocfunc = uloc_getDisplayName; break;
default:
assert(false);
return false;
}
icu::UnicodeString buf;
auto ubuf = buf.getBuffer(64);
do {
UErrorCode error = U_ZERO_ERROR;
int32_t len = ulocfunc(locname.c_str(), disp_locale.c_str(),
ubuf, buf.getCapacity(), &error);
if (error != U_BUFFER_OVERFLOW_ERROR &&
error != U_STRING_NOT_TERMINATED_WARNING) {
if (U_FAILURE(error)) {
s_intl_error->setError(error, "locale_get_display_%s : unable to "
"get locale %s",
LocaleName(tag).c_str(),
LocaleName(tag).c_str());
return false;
}
buf.releaseBuffer(len);
error = U_ZERO_ERROR;
String out(u8(buf, error));
if (U_FAILURE(error)) {
s_intl_error->setError(error, "Unable to convert result from "
"locale_get_display_%s to UTF-8",
LocaleName(tag).c_str());
return false;
}
return out;
}
if (len <= buf.getCapacity()) {
// Avoid infinite loop
buf.releaseBuffer(0);
s_intl_error->setError(U_INTERNAL_PROGRAM_ERROR,
"Got invalid response from ICU");
return false;
}
// Grow the buffer to sufficient size
buf.releaseBuffer(0);
ubuf = buf.getBuffer(len);
} while (true);
not_reached();
return false;
}
int main(int argc, char * const argv[]) {
GArgs args(argc, argv,
"debug;merge;cluster-only;help;force-exons;no-pseudo;MINCOV=MINPID=hvOUNHWCVJMKQNSXTDAPRZFGLEm:g:i:r:s:t:a:b:o:w:x:y:d:");
args.printError(USAGE, true);
if (args.getOpt('h') || args.getOpt("help")) {
GMessage("%s",USAGE);
exit(1);
}
debugMode=(args.getOpt("debug")!=NULL);
decodeChars=(args.getOpt('D')!=NULL);
forceExons=(args.getOpt("force-exons")!=NULL);
NoPseudo=(args.getOpt("no-pseudo")!=NULL);
mRNAOnly=(args.getOpt('O')==NULL);
//sortByLoc=(args.getOpt('S')!=NULL);
addDescr=(args.getOpt('A')!=NULL);
verbose=(args.getOpt('v')!=NULL);
wCDSonly=(args.getOpt('C')!=NULL);
validCDSonly=(args.getOpt('V')!=NULL);
altPhases=(args.getOpt('H')!=NULL);
fmtGTF=(args.getOpt('T')!=NULL); //switch output format to GTF
bothStrands=(args.getOpt('B')!=NULL);
fullCDSonly=(args.getOpt('J')!=NULL);
spliceCheck=(args.getOpt('N')!=NULL);
bool matchAllIntrons=(args.getOpt('K')==NULL);
bool fuzzSpan=(args.getOpt('Q')!=NULL);
if (args.getOpt('M') || args.getOpt("merge")) {
doCluster=true;
doCollapseRedundant=true;
}
else {
if (!matchAllIntrons || fuzzSpan) {
GMessage("%s",USAGE);
GMessage("Error: -K or -Q options require -M/--merge option!\n");
exit(1);
}
}
if (args.getOpt("cluster-only")) {
doCluster=true;
doCollapseRedundant=false;
if (!matchAllIntrons || fuzzSpan) {
GMessage("%s",USAGE);
GMessage("Error: -K or -Q options have no effect with --cluster-only.\n");
exit(1);
}
}
if (fullCDSonly) validCDSonly=true;
if (verbose) {
fprintf(stderr, "Command line was:\n");
args.printCmdLine(stderr);
}
fullattr=(args.getOpt('F')!=NULL);
if (args.getOpt('G')==NULL)
noExonAttr=!fullattr;
else {
noExonAttr=true;
fullattr=true;
}
if (NoPseudo && !fullattr) {
noExonAttr=true;
fullattr=true;
}
ensembl_convert=(args.getOpt('L')!=NULL);
if (ensembl_convert) {
fullattr=true;
noExonAttr=false;
//sortByLoc=true;
}
mergeCloseExons=(args.getOpt('Z')!=NULL);
multiExon=(args.getOpt('U')!=NULL);
writeExonSegs=(args.getOpt('W')!=NULL);
tracklabel=args.getOpt('t');
GFastaDb gfasta(args.getOpt('g'));
//if (gfasta.fastaPath!=NULL)
// sortByLoc=true; //enforce sorting by chromosome/contig
GStr s=args.getOpt('i');
if (!s.is_empty()) maxintron=s.asInt();
FILE* f_repl=NULL;
s=args.getOpt('d');
if (!s.is_empty()) {
if (s=="-") f_repl=stdout;
else {
f_repl=fopen(s.chars(), "w");
if (f_repl==NULL) GError("Error creating file %s\n", s.chars());
}
}
rfltWithin=(args.getOpt('R')!=NULL);
s=args.getOpt('r');
if (!s.is_empty()) {
s.trim();
if (s[0]=='+' || s[0]=='-') {
rfltStrand=s[0];
s.cut(0,1);
}
int isep=s.index(':');
if (isep>0) { //gseq name given
if (rfltStrand==0 && (s[isep-1]=='+' || s[isep-1]=='-')) {
isep--;
rfltStrand=s[isep];
s.cut(isep,1);
}
if (isep>0)
rfltGSeq=Gstrdup((s.substr(0,isep)).chars());
s.cut(0,isep+1);
}
GStr gsend;
char slast=s[s.length()-1];
if (rfltStrand==0 && (slast=='+' || slast=='-')) {
s.chomp(slast);
rfltStrand=slast;
}
if (s.index("..")>=0) gsend=s.split("..");
else gsend=s.split('-');
if (!s.is_empty()) rfltStart=(uint)s.asInt();
if (!gsend.is_empty()) {
rfltEnd=(uint)gsend.asInt();
if (rfltEnd==0) rfltEnd=MAX_UINT;
}
} //gseq/range filtering
else {
if (rfltWithin)
GError("Error: option -R requires -r!\n");
//if (rfltWholeTranscript)
// GError("Error: option -P requires -r!\n");
}
s=args.getOpt('m');
if (!s.is_empty()) {
FILE* ft=fopen(s,"r");
if (ft==NULL) GError("Error opening reference table: %s\n",s.chars());
loadRefTable(ft, reftbl);
fclose(ft);
}
s=args.getOpt('s');
if (!s.is_empty()) {
FILE* fsize=fopen(s,"r");
if (fsize==NULL) GError("Error opening info file: %s\n",s.chars());
loadSeqInfo(fsize, seqinfo);
fclose(fsize);
}
openfw(f_out, args, 'o');
//if (f_out==NULL) f_out=stdout;
if (gfasta.fastaPath==NULL && (validCDSonly || spliceCheck || args.getOpt('w')!=NULL || args.getOpt('x')!=NULL || args.getOpt('y')!=NULL))
GError("Error: -g option is required for options -w, -x, -y, -V, -N, -M !\n");
openfw(f_w, args, 'w');
openfw(f_x, args, 'x');
openfw(f_y, args, 'y');
if (f_y!=NULL || f_x!=NULL) wCDSonly=true;
//useBadCDS=useBadCDS || (fgtfok==NULL && fgtfbad==NULL && f_y==NULL && f_x==NULL);
int numfiles = args.startNonOpt();
//GList<GffObj> gfkept(false,true); //unsorted, free items on delete
int out_counter=0; //number of records printed
while (true) {
GStr infile;
if (numfiles) {
infile=args.nextNonOpt();
if (infile.is_empty()) break;
if (infile=="-") { f_in=stdin; infile="stdin"; }
else
if ((f_in=fopen(infile, "r"))==NULL)
GError("Error: cannot open input file %s!\n",infile.chars());
}
else
infile="-";
GffLoader gffloader(infile.chars());
gffloader.transcriptsOnly=mRNAOnly;
gffloader.fullAttributes=fullattr;
gffloader.noExonAttrs=noExonAttr;
gffloader.mergeCloseExons=mergeCloseExons;
gffloader.showWarnings=(args.getOpt('E')!=NULL);
gffloader.noPseudo=NoPseudo;
gffloader.load(g_data, &validateGffRec, doCluster, doCollapseRedundant,
matchAllIntrons, fuzzSpan, forceExons);
if (doCluster)
collectLocusData(g_data);
if (numfiles==0) break;
}
GStr loctrack("gffcl");
if (tracklabel) loctrack=tracklabel;
g_data.setSorted(&gseqCmpName);
GffPrintMode exonPrinting;
if (fmtGTF) {
exonPrinting = pgtfAny;
} else {
exonPrinting = forceExons ? pgffBoth : pgffAny;
}
bool firstGff3Print=!fmtGTF;
if (doCluster) {
//grouped in loci
for (int g=0;g<g_data.Count();g++) {
GenomicSeqData* gdata=g_data[g];
int gfs_i=0;
for (int l=0;l<gdata->loci.Count();l++) {
GffLocus& loc=*(gdata->loci[l]);
//check all non-replaced transcripts in this locus:
int numvalid=0;
int idxfirstvalid=-1;
for (int i=0;i<loc.rnas.Count();i++) {
GffObj& t=*(loc.rnas[i]);
if (f_out) {
while (gfs_i<gdata->gfs.Count() && gdata->gfs[gfs_i]->start<=t.start) {
GffObj& gfst=*(gdata->gfs[gfs_i]);
if ((gfst.udata&4)==0) { //never printed
gfst.udata|=4;
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
if (gfst.exons.Count()==0 && gfst.children.Count()==0 && forceExons)
gfst.addExon(gfst.start,gfst.end);
gfst.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
++gfs_i;
}
}
GTData* tdata=(GTData*)(t.uptr);
if (tdata->replaced_by!=NULL) {
if (f_repl && (t.udata & 8)==0) {
//t.udata|=8;
fprintf(f_repl, "%s", t.getID());
GTData* rby=tdata;
while (rby->replaced_by!=NULL) {
fprintf(f_repl," => %s", rby->replaced_by->getID());
rby->rna->udata|=8;
rby=(GTData*)(rby->replaced_by->uptr);
}
fprintf(f_repl, "\n");
}
continue;
}
if (process_transcript(gfasta, t)) {
t.udata|=4; //tag it as valid
numvalid++;
if (idxfirstvalid<0) idxfirstvalid=i;
}
}
if (f_out && numvalid>0) {
GStr locname("RLOC_");
locname.appendfmt("%08d",loc.locus_num);
if (!fmtGTF) {
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
fprintf(f_out,"%s\t%s\tlocus\t%d\t%d\t.\t%c\t.\tID=%s;locus=%s",
loc.rnas[0]->getGSeqName(), loctrack.chars(), loc.start, loc.end, loc.strand,
locname.chars(), locname.chars());
//const char* loc_gname=loc.getGeneName();
if (loc.gene_names.Count()>0) { //print all gene names associated to this locus
fprintf(f_out, ";genes=%s",loc.gene_names.First()->name.chars());
for (int i=1;i<loc.gene_names.Count();i++) {
fprintf(f_out, ",%s",loc.gene_names[i]->name.chars());
}
}
if (loc.gene_ids.Count()>0) { //print all GeneIDs names associated to this locus
fprintf(f_out, ";geneIDs=%s",loc.gene_ids.First()->name.chars());
for (int i=1;i<loc.gene_ids.Count();i++) {
fprintf(f_out, ",%s",loc.gene_ids[i]->name.chars());
}
}
fprintf(f_out, ";transcripts=%s",loc.rnas[idxfirstvalid]->getID());
for (int i=idxfirstvalid+1;i<loc.rnas.Count();i++) {
fprintf(f_out, ",%s",loc.rnas[i]->getID());
}
fprintf(f_out, "\n");
}
//now print all valid, non-replaced transcripts in this locus:
for (int i=0;i<loc.rnas.Count();i++) {
GffObj& t=*(loc.rnas[i]);
GTData* tdata=(GTData*)(t.uptr);
if (tdata->replaced_by!=NULL || ((t.udata & 4)==0)) continue;
t.addAttr("locus", locname.chars());
out_counter++;
if (fmtGTF) t.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
else {
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
//print the parent first, if any
if (t.parent!=NULL && ((t.parent->udata & 4)==0)) {
GTData* pdata=(GTData*)(t.parent->uptr);
if (pdata && pdata->geneinfo!=NULL)
pdata->geneinfo->finalize();
t.parent->addAttr("locus", locname.chars());
t.parent->printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
t.parent->udata|=4;
}
t.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
}
} //have valid transcripts to print
}//for each locus
//print the rest of the isolated pseudo/gene/region features not printed yet
if (f_out) {
while (gfs_i<gdata->gfs.Count()) {
GffObj& gfst=*(gdata->gfs[gfs_i]);
if ((gfst.udata&4)==0) { //never printed
gfst.udata|=4;
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
if (gfst.exons.Count()==0 && gfst.children.Count()==0 && forceExons)
gfst.addExon(gfst.start,gfst.end);
gfst.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
++gfs_i;
}
}
} //for each genomic sequence
}
else {
//not grouped into loci, print the rnas with their parents, if any
int numvalid=0;
for (int g=0;g<g_data.Count();g++) {
GenomicSeqData* gdata=g_data[g];
int gfs_i=0;
for (int m=0;m<gdata->rnas.Count();m++) {
GffObj& t=*(gdata->rnas[m]);
if (f_out) {
while (gfs_i<gdata->gfs.Count() && gdata->gfs[gfs_i]->start<=t.start) {
GffObj& gfst=*(gdata->gfs[gfs_i]);
if ((gfst.udata&4)==0) { //never printed
gfst.udata|=4;
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
if (gfst.exons.Count()==0 && gfst.children.Count()==0 && forceExons)
gfst.addExon(gfst.start,gfst.end);
gfst.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
++gfs_i;
}
}
GTData* tdata=(GTData*)(t.uptr);
if (tdata->replaced_by!=NULL) continue;
if (process_transcript(gfasta, t)) {
t.udata|=4; //tag it as valid
numvalid++;
if (f_out) {
if (tdata->geneinfo) tdata->geneinfo->finalize();
out_counter++;
if (fmtGTF) t.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
else {
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
//print the parent first, if any
if (t.parent!=NULL && ((t.parent->udata & 4)==0)) {
GTData* pdata=(GTData*)(t.parent->uptr);
if (pdata && pdata->geneinfo!=NULL)
pdata->geneinfo->finalize();
t.parent->printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
t.parent->udata|=4;
}
t.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
}//GFF/GTF output requested
} //valid transcript
} //for each rna
//print the rest of the isolated pseudo/gene/region features not printed yet
if (f_out) {
while (gfs_i<gdata->gfs.Count()) {
GffObj& gfst=*(gdata->gfs[gfs_i]);
if ((gfst.udata&4)==0) { //never printed
gfst.udata|=4;
if (firstGff3Print) { printGff3Header(f_out, args);firstGff3Print=false; }
if (gfst.exons.Count()==0 && gfst.children.Count()==0 && forceExons)
gfst.addExon(gfst.start,gfst.end);
gfst.printGxf(f_out, exonPrinting, tracklabel, NULL, decodeChars);
}
++gfs_i;
}
}
} //for each genomic seq
} //not clustered
if (f_repl && f_repl!=stdout) fclose(f_repl);
seqinfo.Clear();
//if (faseq!=NULL) delete faseq;
//if (gcdb!=NULL) delete gcdb;
GFREE(rfltGSeq);
FRCLOSE(f_in);
FWCLOSE(f_out);
FWCLOSE(f_w);
FWCLOSE(f_x);
FWCLOSE(f_y);
}
