Homec4science
Diffusion invenio-infoscience (iugw2017) Browse

R3600/modules/bibclassify/lib/bibclassify_acronym_analyzer.py
d30e56fa001aiugw2017

Last Change
Raw File

bibclassify_acronym_analyzer.py
View Options

# -*- coding: utf-8 -*-
##
## This file is part of Invenio.
## Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008 CERN.
##
## Invenio is free software; you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
## published by the Free Software Foundation; either version 2 of the
## License, or (at your option) any later version.
##
## Invenio is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with Invenio; if not, write to the Free Software Foundation, Inc.,
## 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
"""
Bibclassify acronym analyser.
"""
import re
ACRONYM_BRACKETS_REGEX = re.compile("[([] ?(([A-Z]\.?){2,})s? ?[)\]]")
DOTS_REGEX = re.compile("\.")
MAXIMUM_LEVEL = 2
STOPLIST = ("and", "of", "for", "the", "to", "do", "de", "theory",
"model", "radiation", "scheme", "representation")
# INTERFACE
def get_acronyms(fulltext):
"""Finds acronyms and expansions from the fulltext. If needed,
acronyms can already contain a dictionary of previously found
acronyms that will be merged with the current results."""
acronyms = {}
for m in ACRONYM_BRACKETS_REGEX.finditer(fulltext):
acronym = DOTS_REGEX.sub("", m.group(1))
potential_expansion = fulltext[m.start() - 80:m.start()].replace("\n",
" ")
# Strip
potential_expansion = re.sub("(\W).(\W)", "\1\2", potential_expansion)
potential_expansion = re.sub("(\w)\(s\)\W", "\1", potential_expansion)
potential_expansion = re.sub("""[^\w'"]+$""", "", potential_expansion)
potential_expansion = re.sub("[[(].+[\])]", "", potential_expansion)
potential_expansion = re.sub(" {2,}", " ", potential_expansion)
# LEVEL 0: expansion between quotes
# Double quotes
match = re.search(""""([^"]+)["]$""", potential_expansion)
if match is None:
# Single quotes
match = re.search("""'([^"]+)[']$""", potential_expansion)
if match is not None:
if acronym in match.group(1):
continue
pattern = ""
for char in acronym[:-1]:
pattern += "%s\w+\W*" % char
pattern += "%s\w+" % acronym[-1]
if re.search(pattern, match.group(1), re.I) is not None:
_add_expansion_to_acronym_dict(acronym, match.group(1), 0,
acronyms)
continue
pattern = "\W("
for char in acronym[:-1]:
pattern += "%s\w+\W+" % char
pattern += "%s\w+)$" % acronym[-1]
# LEVEL 1: expansion with uppercase initials
match = re.search(pattern, potential_expansion)
if match is not None:
_add_expansion_to_acronym_dict(acronym, match.group(1), 1,
acronyms)
continue
# LEVEL 2: expansion with initials
match = re.search(pattern, potential_expansion, re.I)
if match is not None:
_add_expansion_to_acronym_dict(acronym, match.group(1), 2,
acronyms)
continue
# LEVEL 3: expansion with initials and STOPLIST
potential_expansion_stripped = " ".join([word for word in
_words(potential_expansion) if word not in STOPLIST])
match = re.search(pattern, potential_expansion_stripped, re.I)
if match is not None:
first_expansion_word = re.search("\w+", match.group(1)).group()
start = potential_expansion.lower().rfind(first_expansion_word)
_add_expansion_to_acronym_dict(acronym,
potential_expansion[start:], 3, acronyms)
continue
# LEVEL 4: expansion with fuzzy initials and stoplist
reversed_words = _words(potential_expansion_stripped)
reversed_words.reverse()
reversed_acronym = list(acronym.lower())
reversed_acronym.reverse()
index0 = 0
index1 = 0
word = ""
try:
while index0 < len(reversed_acronym) and index1 < len(reversed_words):
word = reversed_words[index1]
if index0 + 1 < len(reversed_words):
next_word = reversed_words[index0 + 1]
else:
next_word = "_"
char = reversed_acronym[index0]
if index0 + 1 < len(reversed_acronym):
next_char = reversed_acronym[index0 + 1]
else:
next_char = "_"
if char == next_char and \
word.startswith(char) and \
word.count(char) > 1 and \
not next_word.startswith(char):
index0 += 2
index1 += 1
if word.startswith(char):
index0 += 1
index1 += 1
elif char in word and \
not word.endswith(char) and \
word.startswith(next_char):
index0 += 2
index1 += 1
else:
word = ""
break
except IndexError:
word = ""
if not word.startswith(char):
word = ""
if word:
start = potential_expansion.lower().rfind(word)
_add_expansion_to_acronym_dict(acronym,
potential_expansion[start:], 4, acronyms)
continue
# LEVEL 5: expansion with fuzzy initials
reversed_words = _words(potential_expansion.lower())
reversed_words.reverse()
reversed_acronym = list(acronym.lower())
reversed_acronym.reverse()
index0 = 0
index1 = 0
word = ""
try:
while index0 < len(reversed_acronym) and index1 < len(reversed_words):
word = reversed_words[index1]
if index0 + 1 < len(reversed_words):
next_word = reversed_words[index0 + 1]
else:
next_word = ""
char = reversed_acronym[index0]
if index0 + 1 < len(reversed_acronym):
next_char = reversed_acronym[index0 + 1]
else:
next_char = ""
if char == next_char and \
word.startswith(char) and \
word.count(char) > 1 and \
not next_word.startswith(char):
index0 += 2
index1 += 1
if word.startswith(char):
index0 += 1
index1 += 1
elif char in word and \
not word.endswith(char) and \
word.startswith(next_char):
index0 += 2
index1 += 1
else:
word = ""
break
except IndexError:
word = ""
if not word.startswith(char):
word = ""
if word:
start = potential_expansion.lower().rfind(word)
_add_expansion_to_acronym_dict(acronym,
potential_expansion[start:], 5, acronyms)
continue
return acronyms
# PRIVATE METHODS
def _words(expression):
"""Returns a list of words of the expression."""
return re.findall("\w+", expression.lower())
def _add_expansion_to_acronym_dict(acronym, expansion, level, dictionary):
"""Adds an acronym to the dictionary. Takes care of avoiding
duplicates and keeping the expansion marked with the best score."""
if len(acronym) >= len(expansion) or acronym in expansion:
return
for punctuation in re.findall("\W", expansion):
# The expansion contains non-basic punctuation. It is probable
# that it is invalid. Discard it.
if punctuation not in (",", " ", "-"):
return False
if acronym in dictionary:
add = True
for stored_expansion, stored_level in dictionary[acronym]:
if _equivalent_expansions(stored_expansion, expansion):
if level < stored_level:
dictionary[acronym].remove((stored_expansion, stored_level))
break
else:
add = False
if add:
dictionary[acronym].append((expansion, level))
return True
else:
dictionary.setdefault(acronym, []).append((expansion, level))
return True
return False
def _equivalent_expansions(expansion1, expansion2):
"""Compares 2 expansions."""
words1 = _words(expansion1)
words2 = _words(expansion2)
simplified_versions = []
if words1 == words2:
return True
for words in (words1, words2):
store = []
for word in words:
store.append(word[:5])
simplified_versions.append("".join(store))
return simplified_versions[0] == simplified_versions[1]
c4science · Help