Return to cdliSplitter.py CVS log

Up to [Repository] / cdli

Annotation of cdli/cdliSplitter.py, revision 1.7.2.6

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
                     27: komma_exception="([^sStThH])," 
                     28: # grapheme boundaries
1.7.2.6 ! casties    29: #graphemeBounds="\{|\}|<|>|\(|\)|-|_|\#|,|\||\]|\[|\!|\?"
        !            30: graphemeBounds="\{|\}|<|>|-|_|\#|,|\]|\[|\!|\?|\""
1.7.2.2   casties    31: # for words 
1.7.2.6 ! casties    32: #wordBounds="<|>|\(|\)|_|\#|,|\||\]|\[|\!|\?"
        !            33: wordBounds="<|>|_|\#|,|\]|\[|\!|\?|\""
1.7.2.2   casties    34: 
1.5       dwinter    35:            
                     36: class cdliSplitter:
1.7.2.2   casties    37:     """base class for splitter. 
                     38:     the difference between word and grapheme splitter 
                     39:     is the word boundary list."""
1.5       dwinter    40:     
1.1       dwinter    41:     default_encoding = "utf-8"
1.7.2.2   casties    42:     bounds=graphemeBounds
1.5       dwinter    43:     indexName="cdliSplitter"
                     44:     
1.2       dwinter    45:     
1.1       dwinter    46:     def process(self, lst):
1.7.2.2   casties    47:         """gets a list of strings and returns a list of words"""
                     48:         
1.7.2.4   casties    49:         logging.debug("cdliSplitter: %s"%self.indexName) 
1.1       dwinter    50:         result = []
1.2       dwinter    51:         pNum=None
                     52:         lineNum=None
1.5       dwinter    53:     
1.1       dwinter    54:         for t in lst:
1.7.2.2   casties    55:             # normalise line breaks
                     56:             t.replace("\r","\n")
                     57:             # split lines
                     58:             for s in t.split("\n"):
                     59:                 if isinstance(s, str): 
                     60:                     # not unicode
                     61:                     s = unicode(s, self.default_encoding, 'replace')
                     62:          
                     63:                 if (s!=''):
                     64:                     if s[0]=='&': 
                     65:                         # store pNum
                     66:                         pNum=s[1:8]
1.7.2.3   casties    67:                         logging.debug("%s processing: %s"%(self.indexName,pNum))
1.5       dwinter    68:                         
1.7.2.2   casties    69:                     elif not (s[0] in ignoreLines):
                     70:                         # regular line
                     71:                         lineparts=s.split(".")
                     72:                         if len(lineparts)==1: 
                     73:                             # no line number
                     74:                             txt=s
                     75:                         else:
                     76:                             #store line number
                     77:                             txt=lineparts[1]
                     78:                             lineNum=lineparts[0] 
                     79:                             
                     80:                         # delete kommata except kommata relevant for graphemes
                     81:                         txt = re.sub(komma_exception,r"\1",txt)
                     82:                         # replace word boundaries by spaces
                     83:                         txt = re.sub(self.bounds,' ',txt)
                     84:                         # split words
                     85:                         words = txt.split(" ")
                     86:                         for w in words:
                     87:                             w=w.strip()
                     88:                             if not (w==''):
                     89:                                 result.append(w)
                     90: 
1.7.2.6 ! casties    91:         logging.debug("split '%s' into %s"%(lst,repr(result)))
1.5       dwinter    92:         return result
1.2       dwinter    93: 
                     94: 
1.5       dwinter    95: class graphemeSplitter(cdliSplitter):
1.7.2.2   casties    96:     bounds=graphemeBounds
1.5       dwinter    97:     indexName="graphemeSplitter"
                     98:     
                     99: class wordSplitter(cdliSplitter):
1.7.2.2   casties   100:     bounds=wordBounds
1.5       dwinter   101:     indexName="wordSplitter"
                    102:       
                    103: try:
                    104:     element_factory.registerFactory('Word Splitter',
                    105:           'CDLI grapheme splitter', graphemeSplitter)
                    106: except:
                    107:     # in case the splitter is already registered, ValueError is raised
                    108:     pass
1.2       dwinter   109: 
1.5       dwinter   110: try:
                    111:     element_factory.registerFactory('Word Splitter',
                    112:           'CDLI word splitter', wordSplitter)
                    113: except:
                    114:     # in case the splitter is already registered, ValueError is raised
                    115:     pass
1.1       dwinter   116: