Homec4science
Diffusion invenio-infoscience (iugw2017) Browse

R3600/modules/bibindex/lib/bibindex_engine_tokenizer_unit_tests.py
d071120788eciugw2017

Last Change
Raw File

bibindex_engine_tokenizer_unit_tests.py
View Options

# -*- coding:utf-8 -*-
##
## This file is part of Invenio.
## Copyright (C) 2010, 2011, 2012, 2013 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.
"""bibindex_engine_tokenizer_tests - unit tests for tokenizers
There should always be at least one test class for each class in b_e_t.
"""
from invenio.testutils import InvenioTestCase
from invenio.testutils import make_test_suite, run_test_suite
from invenio.bibindex_engine_utils import load_tokenizers
_TOKENIZERS = load_tokenizers()
class TestAuthorTokenizerScanning(InvenioTestCase):
"""Test BibIndex name tokenization"""
def setUp(self):
self.tokenizer = _TOKENIZERS["BibIndexAuthorTokenizer"]()
self.scan = self.tokenizer.scan_string_for_phrases
def test_bifnt_scan_single(self):
"""BibIndexAuthorTokenizer - scanning single names like 'Dido'"""
teststr = "Dido"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'], 'lastnames': ['Dido'], 'nonlastnames': [], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_simple_western_forward(self):
"""BibIndexAuthorTokenizer - scanning simple Western-style: first last"""
teststr = "Ringo Starr"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'], 'lastnames': ['Starr'], 'nonlastnames': ['Ringo'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_simple_western_reverse(self):
"""BibIndexAuthorTokenizer - scanning simple Western-style: last, first"""
teststr = "Starr, Ringo"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'], 'lastnames': ['Starr'], 'nonlastnames': ['Ringo'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_multiname_forward(self):
"""BibIndexAuthorTokenizer - scanning multiword: first middle last"""
teststr = "Michael Edward Peskin"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Peskin'], 'nonlastnames': ['Michael', 'Edward'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_multiname_dotcrammed(self):
"""BibIndexAuthorTokenizer - scanning multiword: f.m. last"""
teststr = "M.E. Peskin"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Peskin'], 'nonlastnames': ['M', 'E'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_multiname_dotcrammed_reversed(self):
"""BibIndexAuthorTokenizer - scanning multiword: last, f.m."""
teststr = "Peskin, M.E."
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Peskin'], 'nonlastnames': ['M', 'E'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_multiname_dashcrammed(self):
"""BibIndexAuthorTokenizer - scanning multiword: first-middle last"""
teststr = "Jean-Luc Picard"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Picard'], 'nonlastnames': ['Jean', 'Luc'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_multiname_dashcrammed_reversed(self):
"""BibIndexAuthorTokenizer - scanning multiword: last, first-middle"""
teststr = "Picard, Jean-Luc"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Picard'], 'nonlastnames': ['Jean', 'Luc'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_compound_lastname_dashes(self):
"""BibIndexAuthorTokenizer - scanning multiword: first middle last-last"""
teststr = "Cantina Octavia Jones-Smith"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Jones', 'Smith'], 'nonlastnames': ['Cantina', 'Octavia'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_compound_lastname_dashes_reverse(self):
"""BibIndexAuthorTokenizer - scanning multiword: last-last, first middle"""
teststr = "Jones-Smith, Cantina Octavia"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Jones', 'Smith'], 'nonlastnames': ['Cantina', 'Octavia'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_compound_lastname_reverse(self):
"""BibIndexAuthorTokenizer - scanning compound last: last last, first"""
teststr = "Alvarez Gaume, Joachim"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Alvarez', 'Gaume'], 'nonlastnames': ['Joachim'], 'titles': [], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_titled(self):
"""BibIndexAuthorTokenizer - scanning title-bearing: last, first, title"""
teststr = "Epstein, Brian, The Fifth Beatle"
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Epstein'], 'nonlastnames': ['Brian'], 'titles': ['The Fifth Beatle'], 'raw' : teststr}
self.assertEqual(output, anticipated)
def test_bifnt_scan_wildly_interesting(self):
"""BibIndexAuthorTokenizer - scanning last last last, first first, title, title"""
teststr = "Ibanez y Gracia, Maria Luisa, II., ed."
output = self.scan(teststr)
anticipated = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Ibanez', 'y', 'Gracia'], 'nonlastnames': ['Maria', 'Luisa'], 'titles': ['II.', 'ed.'], 'raw' : teststr}
self.assertEqual(output, anticipated)
class TestAuthorTokenizerTokens(InvenioTestCase):
"""Test BibIndex name variant token generation from scanned and tagged sets"""
def setUp(self):
self.tokenizer = _TOKENIZERS["BibIndexAuthorTokenizer"]()
self.get_index_tokens = self.tokenizer.parse_scanned_for_phrases
def test_bifnt_tokenize_single(self):
"""BibIndexAuthorTokenizer - tokens for single-word name
Ronaldo
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Ronaldo'], 'nonlastnames': [], 'titles': [], 'raw' : 'Ronaldo'}
output = self.get_index_tokens(tagged_data)
anticipated = ['Ronaldo']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_simple_forward(self):
"""BibIndexAuthorTokenizer - tokens for first last
Ringo Starr
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Starr'], 'nonlastnames': ['Ringo'], 'titles': [], 'raw' : 'Ringo Starr'}
output = self.get_index_tokens(tagged_data)
anticipated = ['R Starr', 'Ringo Starr', 'Starr, R', 'Starr, Ringo']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_simple_reverse(self):
"""BibIndexAuthorTokenizer - tokens for last, first
Starr, Ringo
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Starr'], 'nonlastnames': ['Ringo'], 'titles': [], 'raw' : 'Starr, Ringo'}
output = self.get_index_tokens(tagged_data)
anticipated = ['R Starr', 'Ringo Starr', 'Starr, R', 'Starr, Ringo']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_twoname_forward(self):
"""BibIndexAuthorTokenizer - tokens for first middle last
Michael Edward Peskin
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Peskin'], 'nonlastnames': ['Michael', 'Edward'], 'titles': [], 'raw' : 'Michael Edward Peskin'}
output = self.get_index_tokens(tagged_data)
anticipated = ['E Peskin', 'Edward Peskin', 'M E Peskin', 'M Edward Peskin', 'M Peskin',
'Michael E Peskin', 'Michael Edward Peskin', 'Michael Peskin',
'Peskin, E', 'Peskin, Edward', 'Peskin, M',
'Peskin, M E', 'Peskin, M Edward', 'Peskin, Michael',
'Peskin, Michael E', 'Peskin, Michael Edward']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_compound_last(self):
"""BibIndexAuthorTokenizer - tokens for last last, first
Alvarez Gaume, Joachim
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Alvarez', 'Gaume'], 'nonlastnames': ['Joachim'], 'titles': [], 'raw' : 'Alvarez Gaume, Joachim'}
output = self.get_index_tokens(tagged_data)
anticipated = ['Alvarez Gaume, J', 'Alvarez Gaume, Joachim', 'Alvarez, J', 'Alvarez, Joachim', 'Gaume, J',
'Gaume, Joachim', 'J Alvarez', 'J Alvarez Gaume', 'J Gaume', 'Joachim Alvarez',
'Joachim Alvarez Gaume', 'Joachim Gaume']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_titled(self):
"""BibIndexAuthorTokenizer - tokens for last, first, title
Epstein, Brian, The Fifth Beatle
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Epstein'], 'nonlastnames': ['Brian'], 'titles': ['The Fifth Beatle'], 'raw' : 'Epstein, Brian, The Fifth Beatle'}
output = self.get_index_tokens(tagged_data)
anticipated = ['B Epstein', 'B Epstein, The Fifth Beatle', 'Brian Epstein',
'Brian Epstein, The Fifth Beatle', 'Epstein, B', 'Epstein, B, The Fifth Beatle',
'Epstein, Brian', 'Epstein, Brian, The Fifth Beatle']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_wildly_interesting(self):
"""BibIndexAuthorTokenizer - tokens for last last last, first first, title, title
Ibanez y Gracia, Maria Luisa, II, (ed.)
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Ibanez', 'y', 'Gracia'], 'nonlastnames': ['Maria', 'Luisa'], 'titles': ['II', '(ed.)'], 'raw' : 'Ibanez y Gracia, Maria Luisa, II, (ed.)'}
output = self.get_index_tokens(tagged_data)
anticipated = ['Gracia, L', 'Gracia, Luisa', 'Gracia, M', 'Gracia, M L', 'Gracia, M Luisa',
'Gracia, Maria', 'Gracia, Maria L', 'Gracia, Maria Luisa',
'Ibanez y Gracia, L', 'Ibanez y Gracia, L, II',
'Ibanez y Gracia, Luisa', 'Ibanez y Gracia, Luisa, II',
'Ibanez y Gracia, M', 'Ibanez y Gracia, M L', 'Ibanez y Gracia, M L, II',
'Ibanez y Gracia, M Luisa', 'Ibanez y Gracia, M Luisa, II',
'Ibanez y Gracia, M, II',
'Ibanez y Gracia, Maria',
'Ibanez y Gracia, Maria L', 'Ibanez y Gracia, Maria L, II',
'Ibanez y Gracia, Maria Luisa', 'Ibanez y Gracia, Maria Luisa, II',
'Ibanez y Gracia, Maria, II',
'Ibanez, L', 'Ibanez, Luisa',
'Ibanez, M', 'Ibanez, M L', 'Ibanez, M Luisa', 'Ibanez, Maria',
'Ibanez, Maria L', 'Ibanez, Maria Luisa', 'L Gracia', 'L Ibanez',
'L Ibanez y Gracia', 'L Ibanez y Gracia, II', 'Luisa Gracia', 'Luisa Ibanez',
'Luisa Ibanez y Gracia', 'Luisa Ibanez y Gracia, II', 'M Gracia',
'M Ibanez', 'M Ibanez y Gracia', 'M Ibanez y Gracia, II', 'M L Gracia',
'M L Ibanez', 'M L Ibanez y Gracia', 'M L Ibanez y Gracia, II',
'M Luisa Gracia', 'M Luisa Ibanez', 'M Luisa Ibanez y Gracia', 'M Luisa Ibanez y Gracia, II',
'Maria Gracia',
'Maria Ibanez', 'Maria Ibanez y Gracia', 'Maria Ibanez y Gracia, II',
'Maria L Gracia', 'Maria L Ibanez', 'Maria L Ibanez y Gracia', 'Maria L Ibanez y Gracia, II',
'Maria Luisa Gracia', 'Maria Luisa Ibanez', 'Maria Luisa Ibanez y Gracia',
'Maria Luisa Ibanez y Gracia, II']
self.assertEqual(output, anticipated)
def test_bifnt_tokenize_multimiddle_forward(self):
"""BibIndexAuthorTokenizer - tokens for first middle middle last
W K H Panofsky
"""
tagged_data = {'TOKEN_TAG_LIST': ['lastnames', 'nonlastnames', 'titles', 'raw'],
'lastnames': ['Panofsky'], 'nonlastnames': ['W', 'K', 'H'], 'titles': [], 'raw' : 'W K H Panofsky'}
output = self.get_index_tokens(tagged_data)
anticipated = ['H Panofsky', 'K H Panofsky', 'K Panofsky', 'Panofsky, H', 'Panofsky, K',
'Panofsky, K H', 'Panofsky, W', 'Panofsky, W H', 'Panofsky, W K',
'Panofsky, W K H', 'W H Panofsky',
'W K H Panofsky', 'W K Panofsky', 'W Panofsky']
self.assertEqual(output, anticipated)
def test_tokenize(self):
"""BibIndexAuthorTokenizer - check tokenize_for_phrases()
Ringo Starr
"""
teststr = "Ringo Starr"
output = self.tokenizer.tokenize_for_phrases(teststr)
anticipated = ['R Starr', 'Ringo Starr', 'Starr, R', 'Starr, Ringo']
self.assertEqual(output, anticipated)
class TestExactAuthorTokenizer(InvenioTestCase):
"""Test exact author name tokenizer."""
def setUp(self):
"""setup"""
self.tokenizer = _TOKENIZERS["BibIndexExactAuthorTokenizer"]()
self.tokenize = self.tokenizer.tokenize_for_phrases
def test_exact_author_name_tokenizer_bare(self):
"""BibIndexExactNameTokenizer - bare name"""
self.assertEqual(self.tokenize('John Doe'),
['John Doe'])
def test_exact_author_name_tokenizer_dots(self):
"""BibIndexExactNameTokenizer - name with dots"""
self.assertEqual(self.tokenize('J. Doe'),
['J Doe'])
self.assertEqual(self.tokenize('J.R. Doe'),
['J R Doe'])
self.assertEqual(self.tokenize('J. R. Doe'),
['J R Doe'])
def test_exact_author_name_tokenizer_trailing_dots(self):
"""BibIndexExactNameTokenizer - name with trailing dots"""
self.assertEqual(self.tokenize('Doe, J'),
['Doe, J'])
self.assertEqual(self.tokenize('Doe, J.'),
['Doe, J'])
def test_exact_author_name_tokenizer_hyphens(self):
"""BibIndexExactNameTokenizer - name with hyphens"""
self.assertEqual(self.tokenize('Doe, Jean-Pierre'),
['Doe, Jean Pierre'])
class TestCJKTokenizer(InvenioTestCase):
"""Tests for CJK Tokenizer which splits CJK words into characters and treats
every single character as a word"""
@classmethod
def setUp(self):
self.tokenizer = _TOKENIZERS["BibIndexCJKTokenizer"]()
def test_tokenize_for_words_phrase_galaxy(self):
"""tokenizing phrase: galaxy s4据信"""
phrase = "galaxy s4据信"
result = self.tokenizer.tokenize_for_words(phrase)
self.assertEqual(sorted(['galaxy','s4','据','信']), sorted(result))
def test_tokenize_for_words_phrase_with_special_punctuation(self):
"""tokenizing phrase: 马英九：台湾民"""
phrase = u"马英九：台湾民"
result = self.tokenizer.tokenize_for_words(phrase)
self.assertEqual(sorted(['马','英','九','台','湾','民']), sorted(result))
def test_tokenize_for_words_phrase_with_special_punctuation_two(self):
"""tokenizing phrase: 色的“刀子嘴”"""
phrase = u"色的“刀子嘴”"
result = self.tokenizer.tokenize_for_words(phrase)
self.assertEqual(sorted(['色','的','刀','子','嘴']), sorted(result))
def test_tokenize_for_words_simple_phrase(self):
"""tokenizing phrase: 春眠暁覚"""
self.assertEqual(sorted(self.tokenizer.tokenize_for_words(u'春眠暁覚')), sorted(['春', '眠', '暁', '覚']))
def test_tokenize_for_words_mixed_phrase(self):
"""tokenizing phrase: 春眠暁ABC覚"""
self.assertEqual(sorted(self.tokenizer.tokenize_for_words(u'春眠暁ABC覚')), sorted(['春', '眠', '暁', 'abc', '覚']))
def test_tokenize_for_words_phrase_with_comma(self):
"""tokenizing phrase: 春眠暁, 暁"""
phrase = u"春眠暁, 暁"
self.assertEqual(sorted(self.tokenizer.tokenize_for_words(phrase)), sorted(['春','眠','暁']))
TEST_SUITE = make_test_suite(TestAuthorTokenizerScanning,
TestAuthorTokenizerTokens,
TestExactAuthorTokenizer,
TestCJKTokenizer)
if __name__ == '__main__':
run_test_suite(TEST_SUITE)
c4science · Help