--- a/SMART/Java/Python/clusterize.py	Mon Apr 22 11:11:10 2013 -0400
+++ b/SMART/Java/Python/clusterize.py	Mon Apr 29 03:20:15 2013 -0400
@@ -31,7 +31,7 @@
 from commons.core.writer.WriterChooser import WriterChooser
 """Clusterize a set of transcripts"""
 
-import os
+import os, os.path, random
 from optparse import OptionParser
 from commons.core.parsing.ParserChooser import ParserChooser
 from commons.core.writer.Gff3Writer import Gff3Writer
@@ -39,127 +39,147 @@
 from SMART.Java.Python.ncList.NCListFilePickle import NCListFileUnpickle
 from SMART.Java.Python.ncList.FileSorter import FileSorter
 from SMART.Java.Python.misc.Progress import Progress
+from SMART.Java.Python.misc.UnlimitedProgress import UnlimitedProgress
 
 class Clusterize(object):
-        
-    def __init__(self, verbosity):
-        self.normalize         = False
-        self.presorted         = False
-        self.distance          = 1
-        self.colinear          = False
-        self.nbWritten         = 0
-        self.nbMerges          = 0
-        self.verbosity         = verbosity
-        self.splittedFileNames = {}
 
-    def __del__(self):
-        for fileName in self.splittedFileNames.values():
-            os.remove(fileName)
+	def __init__(self, verbosity):
+		self.normalize		 = False
+		self.presorted		 = False
+		self.distance		  = 1
+		self.colinear		  = False
+		self.nbWritten		 = 0
+		self.nbMerges		  = 0
+		self.verbosity		 = verbosity
+		self.splittedFileNames = {}
+
+	def __del__(self):
+		for fileName in self.splittedFileNames.values():
+			os.remove(fileName)
 
-    def setInputFile(self, fileName, format):
-        parserChooser = ParserChooser(self.verbosity)
-        parserChooser.findFormat(format)
-        self.parser = parserChooser.getParser(fileName)
-        self.sortedFileName = "%s_sorted.pkl" % (os.path.splitext(fileName)[0])
+	def setInputFile(self, fileName, format):
+		parserChooser = ParserChooser(self.verbosity)
+		parserChooser.findFormat(format)
+		self.parser = parserChooser.getParser(fileName)
+		self.sortedFileName = "%s_sorted_%d.pkl" % (os.path.splitext(fileName)[0], random.randint(1, 100000))
+		if "SMARTTMPPATH" in os.environ:
+			self.sortedFileName = os.path.join(os.environ["SMARTTMPPATH"], os.path.basename(self.sortedFileName))
 
-    def setOutputFileName(self, fileName, format="gff3", title="S-MART", feature="transcript", featurePart="exon"):
-        writerChooser = WriterChooser()
-        writerChooser.findFormat(format)
-        self.writer = writerChooser.getWriter(fileName)
-        self.writer.setTitle(title)
-        self.writer.setFeature(feature)
-        self.writer.setFeaturePart(featurePart)
+	def setOutputFileName(self, fileName, format="gff3", title="S-MART", feature="transcript", featurePart="exon"):
+		writerChooser = WriterChooser()
+		writerChooser.findFormat(format)
+		self.writer = writerChooser.getWriter(fileName)
+		self.writer.setTitle(title)
+		self.writer.setFeature(feature)
+		self.writer.setFeaturePart(featurePart)
 
-    def setDistance(self, distance):
-        self.distance = distance
+	def setDistance(self, distance):
+		self.distance = distance
 
-    def setColinear(self, colinear):
-        self.colinear = colinear
+	def setColinear(self, colinear):
+		self.colinear = colinear
 
-    def setNormalize(self, normalize):
-        self.normalize = normalize
-        
-    def setPresorted(self, presorted):
-        self.presorted = presorted
+	def setNormalize(self, normalize):
+		self.normalize = normalize
+		
+	def setPresorted(self, presorted):
+		self.presorted = presorted
 
-    def _sortFile(self):
-        fs = FileSorter(self.parser, self.verbosity-4)
-        fs.perChromosome(True)
-        fs.setPresorted(self.presorted)
-        fs.setOutputFileName(self.sortedFileName)
-        fs.sort()
-        self.splittedFileNames       = fs.getOutputFileNames()
-        self.nbElementsPerChromosome = fs.getNbElementsPerChromosome()
-        self.nbElements              = fs.getNbElements()
-        
-    def _iterate(self, chromosome):
-        progress    = Progress(self.nbElementsPerChromosome[chromosome], "Checking chromosome %s" % (chromosome), self.verbosity)
-        transcripts = []
-        parser      = NCListFileUnpickle(self.splittedFileNames[chromosome], self.verbosity)
-        for newTranscript in parser.getIterator():
-            newTranscripts = []
-            for oldTranscript in transcripts:
-                if self._checkOverlap(newTranscript, oldTranscript):
-                    self._merge(newTranscript, oldTranscript)
-                elif self._checkPassed(newTranscript, oldTranscript):
-                    self._write(oldTranscript)
-                else:
-                    newTranscripts.append(oldTranscript)
-            newTranscripts.append(newTranscript)
-            transcripts = newTranscripts
-            progress.inc()
-        for transcript in transcripts:
-            self._write(transcript)
-        progress.done()
+	def _sortFile(self):
+		if self.presorted:
+			return
+		fs = FileSorter(self.parser, self.verbosity-4)
+		fs.perChromosome(True)
+		fs.setPresorted(self.presorted)
+		fs.setOutputFileName(self.sortedFileName)
+		fs.sort()
+		self.splittedFileNames       = fs.getOutputFileNames()
+		self.nbElementsPerChromosome = fs.getNbElementsPerChromosome()
+		self.nbElements              = fs.getNbElements()
+		
+	def _iterate(self, chromosome):
+		if chromosome == None:
+			progress = UnlimitedProgress(10000, "Reading input file", self.verbosity)
+			parser   = self.parser
+		else:
+			progress = Progress(self.nbElementsPerChromosome[chromosome], "Checking chromosome %s" % (chromosome), self.verbosity)
+			parser   = NCListFileUnpickle(self.splittedFileNames[chromosome], self.verbosity)
+		transcripts     = []
+		self.nbElements = 0
+		for newTranscript in parser.getIterator():
+			newTranscripts = []
+			if newTranscript.__class__.__name__ == "Mapping":
+				newTranscript = newTranscript.getTranscript()
+			for oldTranscript in transcripts:
+				if self._checkOverlap(newTranscript, oldTranscript):
+					self._merge(newTranscript, oldTranscript)
+				elif self._checkPassed(newTranscript, oldTranscript):
+					self._write(oldTranscript)
+				else:
+					newTranscripts.append(oldTranscript)
+			newTranscripts.append(newTranscript)
+			transcripts = newTranscripts
+			self.nbElements += 1
+			progress.inc()
+		for transcript in transcripts:
+			self._write(transcript)
+		progress.done()
 
-    def _merge(self, transcript1, transcript2):
-        self.nbMerges += 1
-        transcript2.setDirection(transcript1.getDirection())
-        transcript1.merge(transcript2)
+	def _merge(self, transcript1, transcript2):
+		self.nbMerges += 1
+		transcript2.setDirection(transcript1.getDirection())
+		transcript1.merge(transcript2)
 
-    def _write(self, transcript):
-        self.nbWritten += 1
-        self.writer.addTranscript(transcript)
+	def _write(self, transcript):
+		self.nbWritten += 1
+		self.writer.addTranscript(transcript)
 
-    def _checkOverlap(self, transcript1, transcript2):
-        if self.colinear and transcript1.getDirection() != transcript2.getDirection():
-            return False
-        if transcript1.getDistance(transcript2) > self.distance:
-            return False
-        return True
+	def _checkOverlap(self, transcript1, transcript2):
+		if transcript1.getChromosome() != transcript2.getChromosome():
+			return False
+		if self.colinear and transcript1.getDirection() != transcript2.getDirection():
+			return False
+		if transcript1.getDistance(transcript2) > self.distance:
+			return False
+		return True
 
-    def _checkPassed(self, transcript1, transcript2):
-        return (transcript1.getDistance(transcript2) > self.distance)
+	def _checkPassed(self, transcript1, transcript2):
+		return ((transcript1.getChromosome() != transcript2.getChromosome()) or (transcript1.getDistance(transcript2) > self.distance))
 
-    def run(self):
-        self._sortFile()
-        for chromosome in sorted(self.splittedFileNames.keys()):
-            self._iterate(chromosome)
-        self.writer.close()
-        if self.verbosity > 0:
-            print "# input:   %d" % (self.nbElements)
-            print "# written: %d (%d%% overlaps)" % (self.nbWritten, 0 if (self.nbElements == 0) else ((float(self.nbWritten) / self.nbElements) * 100))
-            print "# merges:  %d" % (self.nbMerges)
-        
+	def run(self):
+		self._sortFile()
+		if self.presorted:
+			self._iterate(None)
+		else:
+			for chromosome in sorted(self.splittedFileNames.keys()):
+				self._iterate(chromosome)
+		self.writer.close()
+		if self.verbosity > 0:
+			print "# input:   %d" % (self.nbElements)
+			print "# written: %d (%d%% overlaps)" % (self.nbWritten, 0 if (self.nbElements == 0) else ((float(self.nbWritten) / self.nbElements) * 100))
+			print "# merges:  %d" % (self.nbMerges)
+		
 
 if __name__ == "__main__":
-    description = "Clusterize v1.0.3: clusterize the data which overlap. [Category: Merge]"
-    
-    parser = OptionParser(description = description)
-    parser.add_option("-i", "--input",     dest="inputFileName",  action="store",                     type="string", help="input file [compulsory] [format: file in transcript format given by -f]")
-    parser.add_option("-f", "--format",    dest="format",         action="store",                     type="string", help="format of file [format: transcript file format]")
-    parser.add_option("-o", "--output",    dest="outputFileName", action="store",                     type="string", help="output file [compulsory] [format: output file in transcript format given by -u]")
-    parser.add_option("-u", "--outputFormat", dest="outputFormat", action="store",     default="gff",             type="string", help="output file format [format: transcript file format]")
-    parser.add_option("-c", "--colinear",  dest="colinear",       action="store_true", default=False,                help="merge colinear transcripts only [format: bool] [default: false]")
-    parser.add_option("-d", "--distance",  dest="distance",       action="store",      default=0,     type="int",    help="max. distance between two transcripts to be merged [format: int] [default: 0]")
-    parser.add_option("-n", "--normalize", dest="normalize",      action="store_true", default=False,                help="normalize the number of reads per cluster by the number of mappings per read [format: bool] [default: false]")
-    parser.add_option("-v", "--verbosity", dest="verbosity",      action="store",      default=1,     type="int",    help="trace level [format: int] [default: 1]")
-    (options, args) = parser.parse_args()
-        
-    c = Clusterize(options.verbosity)
-    c.setInputFile(options.inputFileName, options.format)
-    c.setOutputFileName(options.outputFileName, options.outputFormat)
-    c.setColinear(options.colinear)
-    c.setDistance(options.distance)
-    c.setNormalize(options.normalize)
-    c.run()
+	description = "Clusterize v1.0.3: clusterize the data which overlap. [Category: Merge]"
+
+	parser = OptionParser(description = description)
+	parser.add_option("-i", "--input",     dest="inputFileName",  action="store",				     type="string", help="input file [compulsory] [format: file in transcript format given by -f]")
+	parser.add_option("-f", "--format",    dest="format",		 action="store",				     type="string", help="format of file [format: transcript file format]")
+	parser.add_option("-o", "--output",    dest="outputFileName", action="store",				     type="string", help="output file [compulsory] [format: output file in transcript format given by -u]")
+	parser.add_option("-u", "--outputFormat", dest="outputFormat", action="store",     default="gff",		     type="string", help="output file format [format: transcript file format]")
+	parser.add_option("-c", "--colinear",  dest="colinear",       action="store_true", default=False,				help="merge colinear transcripts only [format: bool] [default: false]")
+	parser.add_option("-d", "--distance",  dest="distance",       action="store",      default=0,     type="int",    help="max. distance between two transcripts to be merged [format: int] [default: 0]")
+	parser.add_option("-n", "--normalize", dest="normalize",      action="store_true", default=False,				help="normalize the number of reads per cluster by the number of mappings per read [format: bool] [default: false]")
+	parser.add_option("-s", "--sorted",    dest="sorted",		 action="store_true", default=False,				help="input is already sorted [format: bool] [default: false]")
+	parser.add_option("-v", "--verbosity", dest="verbosity",      action="store",      default=1,     type="int",    help="trace level [format: int] [default: 1]")
+	(options, args) = parser.parse_args()
+
+	c = Clusterize(options.verbosity)
+	c.setInputFile(options.inputFileName, options.format)
+	c.setOutputFileName(options.outputFileName, options.outputFormat)
+	c.setColinear(options.colinear)
+	c.setDistance(options.distance)
+	c.setNormalize(options.normalize)
+	c.setPresorted(options.sorted)
+	c.run()