Source Code for Module TEES.Utils.InteractionXML.CopyParse

  1  try: 
  2      import xml.etree.cElementTree as ET 
  3  except ImportError: 
  4      import cElementTree as ET 
  5  import Utils.ElementTreeUtils as ETUtils 
  6  import sys 
  7  import CorpusElements 
  8  from optparse import OptionParser 
  9   
 10 -def copyParse(input, source, output, parse, tokenization): 
 11      print >> sys.stderr, "Loading input file", input 
 12      inputTree = ETUtils.ETFromObj(input) 
 13      inputRoot = inputTree.getroot() 
 14      print >> sys.stderr, "Loading source:", 
 15      sourceElements = CorpusElements.loadCorpus(source, parse, tokenization) 
 16      sourceSentencesByText = {} 
 17      for sentence in sourceElements.sentences: 
 18          sentenceText = sentence.sentence.get("text") 
 19          #assert not sourceSentencesByText.has_key(sentenceText) 
 20          if sourceSentencesByText.has_key(sentenceText): 
 21              print >> sys.stderr, "Duplicate text", sentence.sentence.get("id"), sourceSentencesByText[sentenceText].sentence.get("id")  
 22          sourceSentencesByText[sentenceText] = sentence 
 23      parsesCopied = [0,0] 
 24      tokenizationsCopied = [0,0] 
 25      for sentence in inputRoot.getiterator("sentence"): 
 26          parsesCopied[1] += 1 
 27          tokenizationsCopied[1] += 1 
 28          #sourceSentence = sourceElements.sentencesByOrigId[sentence.attrib["origId"]] 
 29          if not sourceSentencesByText.has_key(sentence.get("text")): 
 30              print >> sys.stderr, "Warning, no text found for sentence", sentence.get("id") 
 31              continue 
 32          sourceSentence = sourceSentencesByText[sentence.get("text")] 
 33          # Create analyses element (if needed) 
 34          targetAnalysesElement = sentence.find("sentenceanalyses") 
 35          if targetAnalysesElement == None: 
 36              targetAnalysesElement = ET.Element("sentenceanalyses") 
 37              sentence.append(targetAnalysesElement) 
 38          # Create parses element (if needed) 
 39          targetParsesElement = targetAnalysesElement.find("parses") 
 40          if targetParsesElement == None: 
 41              targetParsesElement = ET.Element("parses") 
 42              targetAnalysesElement.append(targetParsesElement) 
 43          # Check whether parse already exists 
 44          targetParseElements = targetParsesElement.findall("parse") 
 45          newParse = None 
 46          for parseElement in targetParseElements: 
 47              if parseElement.get("parser") == parse: 
 48                  newParse = parseElement 
 49                  break 
 50          # Copy parse if it doesn't 
 51          if newParse == None and sourceSentence.parseElement != None: 
 52              targetParsesElement.append(sourceSentence.parseElement) 
 53              parsesCopied[0] += 1 
 54           
 55          # Create tokenizations element (if needed) 
 56          targetTokenizationsElement = targetAnalysesElement.find("tokenizations") 
 57          if targetTokenizationsElement == None: 
 58              targetTokenizationsElement = ET.Element("tokenizations") 
 59              targetAnalysesElement.append(targetTokenizationsElement) 
 60          # Check whether tokenization already exists 
 61          targetTokenizationElements = targetTokenizationsElement.findall("tokenization") 
 62          newTokenization = None 
 63          for tokenizationElement in targetTokenizationElements: 
 64              if tokenizationElement.attrib["tokenizer"] == newParse.attrib["tokenizer"]: 
 65                  newTokenization = tokenizationElement 
 66                  break 
 67          # Copy parse if it doesn't 
 68          if newTokenization == None and sourceSentence.tokenizationElement != None: 
 69              targetTokenizationsElement.append(sourceSentence.tokenizationElement) 
 70              tokenizationsCopied[0] += 1 
 71       
 72      print >> sys.stderr, "Copied parse elements", parsesCopied 
 73      print >> sys.stderr, "Copied tokenization elements", tokenizationsCopied 
 74       
 75      if output != None:        
 76          print >> sys.stderr, "Writing output to", output 
 77          ETUtils.write(inputTree, output) 
 78      return inputTree 
 79           
 80  if __name__=="__main__": 
 81      print >> sys.stderr, "##### Copy Parse #####" 
 82      # Import Psyco if available 
 83      try: 
 84          import psyco 
 85          psyco.full() 
 86          print >> sys.stderr, "Found Psyco, using" 
 87      except ImportError: 
 88          print >> sys.stderr, "Psyco not installed" 
 89   
 90      optparser = OptionParser(usage="%prog [options]\nCreate an html visualization for a corpus.") 
 91      optparser.add_option("-i", "--input", default=None, dest="input", help="Corpus in analysis format", metavar="FILE") 
 92      optparser.add_option("-s", "--source", default=None, dest="source", help="Corpus in analysis format", metavar="FILE") 
 93      optparser.add_option("-o", "--output", default=None, dest="output", help="Corpus in analysis format", metavar="FILE") 
 94      optparser.add_option("-t", "--tokenization", default=None, dest="tokenization", help="Tokenization element name") 
 95      optparser.add_option("-p", "--parse", default=None, dest="parse", help="Parse element name") 
 96      (options, args) = optparser.parse_args() 
 97      assert(options.input != None) 
 98      assert(options.source != None) 
 99      assert(options.output != None) 
100      copyParse(options.input, options.source, options.output, options.parse, options.tokenization) 
101