version 1.7.2.6, 2007/12/03 21:30:19
|
version 1.7.2.7, 2007/12/11 17:27:36
|
Line 28 komma_exception="([^sStThH]),"
|
Line 28 komma_exception="([^sStThH]),"
|
# grapheme boundaries |
# grapheme boundaries |
#graphemeBounds="\{|\}|<|>|\(|\)|-|_|\#|,|\||\]|\[|\!|\?" |
#graphemeBounds="\{|\}|<|>|\(|\)|-|_|\#|,|\||\]|\[|\!|\?" |
graphemeBounds="\{|\}|<|>|-|_|\#|,|\]|\[|\!|\?|\"" |
graphemeBounds="\{|\}|<|>|-|_|\#|,|\]|\[|\!|\?|\"" |
|
graphemeIgnore="" |
# for words |
# for words |
#wordBounds="<|>|\(|\)|_|\#|,|\||\]|\[|\!|\?" |
#wordBounds="<|>|\(|\)|_|\#|,|\||\]|\[|\!|\?" |
wordBounds="<|>|_|\#|,|\]|\[|\!|\?|\"" |
wordBounds="_|,|\"" |
|
wordIgnore="<|>|\#|\||\]|\[|\!|\?" |
|
|
class cdliSplitter: |
class cdliSplitter: |
"""base class for splitter. |
"""base class for splitter. |
Line 40 class cdliSplitter:
|
Line 41 class cdliSplitter:
|
|
|
default_encoding = "utf-8" |
default_encoding = "utf-8" |
bounds=graphemeBounds |
bounds=graphemeBounds |
|
ignore=graphemeIgnore |
indexName="cdliSplitter" |
indexName="cdliSplitter" |
|
|
|
|
Line 79 class cdliSplitter:
|
Line 81 class cdliSplitter:
|
|
|
# delete kommata except kommata relevant for graphemes |
# delete kommata except kommata relevant for graphemes |
txt = re.sub(komma_exception,r"\1",txt) |
txt = re.sub(komma_exception,r"\1",txt) |
|
# replace letters to be ignored |
|
txt = re.sub(self.ignore,'',txt) |
# replace word boundaries by spaces |
# replace word boundaries by spaces |
txt = re.sub(self.bounds,' ',txt) |
txt = re.sub(self.bounds,' ',txt) |
# split words |
# split words |
Line 94 class cdliSplitter:
|
Line 98 class cdliSplitter:
|
|
|
class graphemeSplitter(cdliSplitter): |
class graphemeSplitter(cdliSplitter): |
bounds=graphemeBounds |
bounds=graphemeBounds |
|
ignore=graphemeIgnore |
indexName="graphemeSplitter" |
indexName="graphemeSplitter" |
|
|
class wordSplitter(cdliSplitter): |
class wordSplitter(cdliSplitter): |
bounds=wordBounds |
bounds=wordBounds |
|
ignore=wordIgnore |
indexName="wordSplitter" |
indexName="wordSplitter" |
|
|
try: |
try: |