Return to cdliSplitter.py CVS log

Up to [Repository] / cdli

Annotation of cdli/cdliSplitter.py, revision 1.7.2.11

1.1       dwinter     1: """
1.7.2.2   casties     2: CDLI word and grapheme splitter
1.1       dwinter     3: """
                      4: 
                      5: from Products.ZCTextIndex.PipelineFactory import element_factory
                      6: 
                      7: import re
1.5       dwinter     8: import logging
1.1       dwinter     9: 
                     10: def getSupportedEncoding(encodings):
                     11:     for encoding in encodings:
                     12:         try:
                     13:             unicode('A', encoding)
                     14:             return encoding
                     15:         except:
                     16:             pass
                     17:     return 'utf-8'
                     18: 
                     19: 
                     20: 
                     21: """beta of a fulltext splitter for cdli
                     22: 
                     23: """
1.7       casties    24: ignoreLines=['$','@','#','&','>']
1.1       dwinter    25: separators=['']
1.7.2.2   casties    26: # kommas relevant for graphemes will not be deleted
1.7.2.9   casties    27: komma_exception="([^sStThH]),"
                     28: komma_exceptionex=re.compile(komma_exception)
1.7.2.2   casties    29: # grapheme boundaries
1.7.2.6   casties    30: #graphemeBounds="\{|\}|<|>|\(|\)|-|_|\#|,|\||\]|\[|\!|\?"
                     31: graphemeBounds="\{|\}|<|>|-|_|\#|,|\]|\[|\!|\?|\""
1.7.2.11! casties    32: graphemeIgnore="<|>|\#|\||\]|\[|\!|\?\*|;"
1.7.2.2   casties    33: # for words 
1.7.2.6   casties    34: #wordBounds="<|>|\(|\)|_|\#|,|\||\]|\[|\!|\?"
1.7.2.7   casties    35: wordBounds="_|,|\""
1.7.2.11! casties    36: wordIgnore="<|>|\#|\||\]|\[|\!|\?\*|;"
1.5       dwinter    37:            
                     38: class cdliSplitter:
1.7.2.2   casties    39:     """base class for splitter. 
                     40:     the difference between word and grapheme splitter 
                     41:     is the word boundary list."""
1.5       dwinter    42:     
1.1       dwinter    43:     default_encoding = "utf-8"
1.7.2.2   casties    44:     bounds=graphemeBounds
1.7.2.9   casties    45:     boundsex=re.compile(graphemeBounds)
1.7.2.7   casties    46:     ignore=graphemeIgnore
1.7.2.9   casties    47:     ignorex=re.compile(graphemeIgnore)
1.5       dwinter    48:     indexName="cdliSplitter"
                     49:     
1.2       dwinter    50:     
1.1       dwinter    51:     def process(self, lst):
1.7.2.2   casties    52:         """gets a list of strings and returns a list of words"""
                     53:         
1.7.2.4   casties    54:         logging.debug("cdliSplitter: %s"%self.indexName) 
1.1       dwinter    55:         result = []
1.2       dwinter    56:         pNum=None
                     57:         lineNum=None
1.5       dwinter    58:     
1.1       dwinter    59:         for t in lst:
1.7.2.2   casties    60:             # normalise line breaks
                     61:             t.replace("\r","\n")
                     62:             # split lines
                     63:             for s in t.split("\n"):
                     64:                 if isinstance(s, str): 
                     65:                     # not unicode
                     66:                     s = unicode(s, self.default_encoding, 'replace')
                     67:          
                     68:                 if (s!=''):
                     69:                     if s[0]=='&': 
                     70:                         # store pNum
                     71:                         pNum=s[1:8]
1.7.2.3   casties    72:                         logging.debug("%s processing: %s"%(self.indexName,pNum))
1.5       dwinter    73:                         
1.7.2.2   casties    74:                     elif not (s[0] in ignoreLines):
                     75:                         # regular line
                     76:                         lineparts=s.split(".")
                     77:                         if len(lineparts)==1: 
                     78:                             # no line number
                     79:                             txt=s
                     80:                         else:
                     81:                             #store line number
                     82:                             txt=lineparts[1]
                     83:                             lineNum=lineparts[0] 
                     84:                             
                     85:                         # delete kommata except kommata relevant for graphemes
1.7.2.9   casties    86:                         txt = komma_exceptionex.sub(r"\1",txt)
1.7.2.2   casties    87:                         # replace word boundaries by spaces
1.7.2.9   casties    88:                         txt = self.boundsex.sub(' ',txt)
                     89:                         # replace letters to be ignored
                     90:                         txt = self.ignorex.sub('',txt)
1.7.2.2   casties    91:                         # split words
                     92:                         words = txt.split(" ")
                     93:                         for w in words:
                     94:                             w=w.strip()
                     95:                             if not (w==''):
                     96:                                 result.append(w)
                     97: 
1.7.2.8   casties    98:         #logging.debug("split '%s' into %s"%(lst,repr(result)))
1.5       dwinter    99:         return result
1.2       dwinter   100: 
                    101: 
1.5       dwinter   102: class graphemeSplitter(cdliSplitter):
1.7.2.2   casties   103:     bounds=graphemeBounds
1.7.2.9   casties   104:     boundsex=re.compile(graphemeBounds)
1.7.2.7   casties   105:     ignore=graphemeIgnore
1.7.2.9   casties   106:     ignorex=re.compile(graphemeIgnore)
1.5       dwinter   107:     indexName="graphemeSplitter"
                    108:     
                    109: class wordSplitter(cdliSplitter):
1.7.2.2   casties   110:     bounds=wordBounds
1.7.2.9   casties   111:     boundsex=re.compile(wordBounds)
1.7.2.7   casties   112:     ignore=wordIgnore
1.7.2.9   casties   113:     ignorex=re.compile(wordIgnore)
1.5       dwinter   114:     indexName="wordSplitter"
                    115:       
                    116: try:
                    117:     element_factory.registerFactory('Word Splitter',
                    118:           'CDLI grapheme splitter', graphemeSplitter)
                    119: except:
                    120:     # in case the splitter is already registered, ValueError is raised
                    121:     pass
1.2       dwinter   122: 
1.5       dwinter   123: try:
                    124:     element_factory.registerFactory('Word Splitter',
                    125:           'CDLI word splitter', wordSplitter)
                    126: except:
                    127:     # in case the splitter is already registered, ValueError is raised
                    128:     pass
1.1       dwinter   129: