def test_compress_context_in_tokenlists():
tokenized = [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph.make(part='123', section='23', paragraph='a'),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph.make(part='123',
section='23',
paragraphs=['a', '1']),
tokens.Paragraph.make(paragraphs=[None, None, 'i']),
tokens.Paragraph.make(section='23', paragraph='b')
])
]
assert amdparser.compress_context_in_tokenlists(tokenized) == [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph.make(part='123', section='23', paragraph='a'),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph.make(part='123',
section='23',
paragraphs=['a', '1']),
tokens.Paragraph.make(part='123',
section='23',
paragraphs=['a', '1', 'i']),
tokens.Paragraph.make(part='123', section='23', paragraph='b')
])
]
def test_compress_context_in_tokenlists(self):
tokenized = [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph(part='123', section='23', paragraph='a'),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph(part='123',
section='23',
paragraphs=['a', '1']),
tokens.Paragraph(paragraphs=[None, None, 'i']),
tokens.Paragraph(section='23', paragraph='b')
])
]
converted = amdparser.compress_context_in_tokenlists(tokenized)
self.assertEqual(converted, [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph(part='123', section='23', paragraph='a'),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph(
part='123', section='23', paragraphs=['a', '1']),
tokens.Paragraph(
part='123', section='23', paragraphs=['a', '1', 'i']),
tokens.Paragraph(part='123', section='23', paragraph='b')
])
])
def test_multiple_moved_not_even_number_of_elements_on_either_side():
tokenized = [
tokens.TokenList([
tokens.Paragraph.make(part='444', sub='1'),
tokens.Paragraph.make(part='444', sub='2')
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([tokens.Paragraph.make(part='444', sub='3')])
]
assert tokenized == amdparser.multiple_moves(tokenized)
def test_contains_one_tokenlist(self):
token_list = self.list_of_tokens()
designate_token_2 = tokens.Verb(tokens.Verb.DESIGNATE, True)
tokenized = [tokens.TokenList(token_list), designate_token_2]
self.assertTrue(contains_one_tokenlist(tokenized))
tokenized = [
tokens.TokenList(token_list), designate_token_2,
tokens.TokenList(token_list)
]
self.assertFalse(contains_one_tokenlist(tokenized))
def test_context_to_paragraph_exceptions2():
tokenized = [
tokens.Verb(tokens.Verb.PUT, active=True),
tokens.Context(['2']),
tokens.TokenList([tokens.Paragraph.make(part='3')])
]
assert tokenized == amdparser.context_to_paragraph(tokenized)
def test_separate_tokenlist():
tokenized = [
tokens.Context(['1']),
tokens.TokenList([
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Context([None, '2'])
]),
tokens.Paragraph.make(sub='3'),
tokens.TokenList([tokens.Paragraph.make(section='b')])
]
assert amdparser.separate_tokenlist(tokenized) == [
tokens.Context(['1']),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Context([None, '2']),
tokens.Paragraph.make(sub='3'),
tokens.Paragraph.make(section='b')
]
def test_multiple_intro_texts(self):
text = ('the introductory text of paragraphs (a)(5)(ii) and '
'(d)(5)(ii)')
self.assertEqual(
parse_text(text),
[tokens.TokenList([
tokens.Paragraph(paragraphs=['a', '5', 'ii'], field='text'),
tokens.Paragraph(paragraphs=['d', '5', 'ii'], field='text'),
])])
def test_separate_tokenlist(self):
tokenized = [
tokens.Context(['1']),
tokens.TokenList([
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Context([None, '2'])
]),
tokens.Paragraph([None, '3']),
tokens.TokenList([tokens.Paragraph([None, None, 'b'])])
]
converted = separate_tokenlist(tokenized)
self.assertEqual(converted, [
tokens.Context(['1']),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Context([None, '2']),
tokens.Paragraph([None, '3']),
tokens.Paragraph([None, None, 'b'])
])
def text_example9(self):
text = u"3. Amend § 5397.31 to revise paragraphs (a)(3)(ii), "
text += "(a)(3)(iii), and (b)(3); and add paragraphs (a)(3)(iv), "
text += "(a)(5)(iv), and (b)(2)(vii) to read as follows:"
result = parse_text(text)
self.assertEqual(result, [
tokens.Context(['5397', None, '31']),
tokens.Verb(tokens.Verb.PUT, active=True),
tokens.TokenList([
tokens.Paragraph([None, None, None, 'a', '3', 'ii']),
tokens.Paragraph([None, None, None, 'a', '3', 'iii']),
tokens.Paragraph([None, None, None, 'b', '3'])
]),
tokens.Verb(tokens.Verb.POST, active=True),
tokens.TokenList([
tokens.Paragraph([None, None, None, 'a', '3', 'iv']),
tokens.Paragraph([None, None, None, 'a', '5', 'iv']),
tokens.Paragraph([None, None, None, 'b', '2', 'vii'])
]),
])
def test_example16(self):
text = " A-30(a), A-30(b), A-30(c), A-30(d) are added"
result = parse_text(text)
self.assertEqual(result, [
tokens.TokenList([
tokens.Paragraph([None, "Appendix:A", "30(a)"]),
tokens.Paragraph([None, "Appendix:A", "30(b)"]),
tokens.Paragraph([None, "Appendix:A", "30(c)"]),
tokens.Paragraph([None, "Appendix:A", "30(d)"]),
]),
tokens.Verb(tokens.Verb.POST, active=False),
])
def test_tokenlist_iteratable(self):
token_list = tokens.TokenList([
tokens.Paragraph([1005, None, 1]),
tokens.Paragraph([1005, None, 2]),
tokens.Paragraph([1005, None, 3]),
])
count = 1
for t in token_list:
self.assertEqual(t.label, [1005, None, count])
count += 1
self.assertEqual(count, 4)
def test_multiple_moves_success():
tokenized = [
tokens.TokenList([
tokens.Paragraph.make(part='444', sub='1'),
tokens.Paragraph.make(part='444', sub='2')
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([
tokens.Paragraph.make(part='444', sub='3'),
tokens.Paragraph.make(part='444', sub='4')
])
]
tokenized = amdparser.multiple_moves(tokenized)
assert tokenized == [
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph.make(part='444', sub='1'),
tokens.Paragraph.make(part='444', sub='3'),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph.make(part='444', sub='2'),
tokens.Paragraph.make(part='444', sub='4')
]
def test_example16(self):
text = " A-30(a), A-30(b), A-30(c), A-30(d) are added"
result = parse_text(text)
self.assertEqual(result, [
tokens.TokenList([
tokens.Paragraph(appendix='A', section='30(a)'),
tokens.Paragraph(appendix='A', section='30(b)'),
tokens.Paragraph(appendix='A', section='30(c)'),
tokens.Paragraph(appendix='A', section='30(d)')
]),
tokens.Verb(tokens.Verb.POST, active=False),
])
def test_tokenlist_iteratable(self):
token_list = tokens.TokenList([
tokens.Paragraph(part=1005, section=1),
tokens.Paragraph(part=1005, section=2),
tokens.Paragraph(part=1005, section=3)
])
count = 1
for t in token_list:
self.assertEqual(t.label, [1005, None, count])
count += 1
self.assertEqual(count, 4)
def test_example18(self):
text = 'Section 106.52(b)(1)(ii)(A) and (B) is revised'
text += ' to read as follows'
result = parse_text(text)
self.assertEqual(result, [
tokens.TokenList([
tokens.Paragraph(['106', None, '52', 'b', '1', 'ii', 'A']),
tokens.Paragraph([None, None, None, None, None, None, 'B']),
]),
tokens.Verb(tokens.Verb.PUT, active=False)
])
def test_multiple_moves(self):
tokenized = [
tokens.TokenList([
tokens.Paragraph(['444', '1']),
tokens.Paragraph(['444', '2'])
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([
tokens.Paragraph(['444', '3']),
tokens.Paragraph(['444', '4'])
])
]
tokenized = multiple_moves(tokenized)
self.assertEqual(tokenized, [
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph(['444', '1']),
tokens.Paragraph(['444', '3']),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph(['444', '2']),
tokens.Paragraph(['444', '4'])
])
# Not even number of elements on either side
tokenized = [
tokens.TokenList([
tokens.Paragraph(['444', '1']),
tokens.Paragraph(['444', '2'])
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([tokens.Paragraph(['444', '3'])])
]
self.assertEqual(tokenized, multiple_moves(tokenized))
# Paragraphs on either side of a move
tokenized = [
tokens.Paragraph(['444', '1']),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.Paragraph(['444', '3'])
]
self.assertEqual(tokenized, multiple_moves(tokenized))
def test_multiple_moves(self):
tokenized = [
tokens.TokenList([
tokens.Paragraph(part='444', sub='1'),
tokens.Paragraph(part='444', sub='2')
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([
tokens.Paragraph(part='444', sub='3'),
tokens.Paragraph(part='444', sub='4')
])
]
tokenized = amdparser.multiple_moves(tokenized)
self.assertEqual(tokenized, [
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph(part='444', sub='1'),
tokens.Paragraph(part='444', sub='3'),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph(part='444', sub='2'),
tokens.Paragraph(part='444', sub='4')
])
# Not even number of elements on either side
tokenized = [
tokens.TokenList([
tokens.Paragraph(part='444', sub='1'),
tokens.Paragraph(part='444', sub='2')
]),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.TokenList([tokens.Paragraph(part='444', sub='3')])
]
self.assertEqual(tokenized, amdparser.multiple_moves(tokenized))
# Paragraphs on either side of a move
tokenized = [
tokens.Paragraph(part='444', sub='1'),
tokens.Verb(tokens.Verb.MOVE, active=False),
tokens.Paragraph(part='444', sub='3')
]
self.assertEqual(tokenized, amdparser.multiple_moves(tokenized))
def test_example13(self):
text = "h. Under Section 6363.36, add comments 36(a), 36(b) and "
text += "36(d)."
result = parse_text(text)
self.assertEqual(result, [
tokens.Context(['6363', None, '36'], certain=True),
tokens.Verb(tokens.Verb.POST, active=True),
# We assume that lists of comments are not context
tokens.TokenList([
tokens.Paragraph(is_interp=True, section='36', paragraph=p)
for p in ('(a)', '(b)', '(d)')
])
])
def test_example4(self):
text = "b. Add Model Forms E-11 through E-15."
result = parse_text(text)
self.assertEqual(result, [
tokens.Verb(tokens.Verb.POST, active=True),
tokens.TokenList([
tokens.Paragraph([None, 'Appendix:E', '11']),
tokens.Paragraph([None, 'Appendix:E', '12']),
tokens.Paragraph([None, 'Appendix:E', '13']),
tokens.Paragraph([None, 'Appendix:E', '14']),
tokens.Paragraph([None, 'Appendix:E', '15'])
])
])
def test_example4(self):
text = "b. Add Model Forms E-11 through E-15."
result = parse_text(text)
self.assertEqual(result, [
tokens.Verb(tokens.Verb.POST, active=True),
tokens.TokenList([
tokens.Paragraph(appendix='E', section='11'),
tokens.Paragraph(appendix='E', section='12'),
tokens.Paragraph(appendix='E', section='13'),
tokens.Paragraph(appendix='E', section='14'),
tokens.Paragraph(appendix='E', section='15')
])
])
def test_example12(self):
text = "comment 31(b), amend paragraph 31(b)(2) by adding "
text += "paragraphs 4 through 6;"
result = parse_text(text)
self.assertEqual(result, [
tokens.Context([None, 'Interpretations', '31', '(b)']),
tokens.Context([None, 'Interpretations', '31', '(b)(2)']),
tokens.Verb(tokens.Verb.POST, active=True),
tokens.TokenList([
tokens.Paragraph([None, 'Interpretations', None, None, '4']),
tokens.Paragraph([None, 'Interpretations', None, None, '5']),
tokens.Paragraph([None, 'Interpretations', None, None, '6'])
])
])
def test_compress_context_in_tokenlists(self):
tokenized = [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph(['123', None, '23', 'a']),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph(['123', None, '23', 'a', '1']),
tokens.Paragraph([None, None, None, None, None, 'i']),
tokens.Paragraph([None, None, '23', 'b'])
])
]
converted = compress_context_in_tokenlists(tokenized)
self.assertEqual(converted, [
tokens.Context(['123', 'Interpretations']),
tokens.Paragraph(['123', None, '23', 'a']),
tokens.Verb(tokens.Verb.PUT, True),
tokens.TokenList([
tokens.Verb(tokens.Verb.POST, True),
tokens.Paragraph(['123', None, '23', 'a', '1']),
tokens.Paragraph(['123', None, '23', 'a', '1', 'i']),
tokens.Paragraph(['123', None, '23', 'b'])
])
])
def test_example15(self):
text = "paragraphs (a)(1)(iii), (a)(1)(iv)(B), (c)(2) introductory "
text += 'text and (c)(2)(ii)(A)(<E T="03">2</E>) redesignating '
text += "paragraph (c)(2)(iii) as paragraph (c)(2)(iv),"
result = parse_text(text)
expected = [tokens.TokenList([
tokens.Paragraph(paragraphs=['a', '1', 'iii']),
tokens.Paragraph(paragraphs=['a', '1', 'iv', 'B']),
tokens.Paragraph(paragraphs=['c', '2'],
field=tokens.Paragraph.TEXT_FIELD),
tokens.Paragraph(paragraphs=['c', '2', 'ii', 'A'])]),
tokens.Verb(tokens.Verb.MOVE, active=True),
tokens.Paragraph(paragraphs=['c', '2', 'iii']),
tokens.Paragraph(paragraphs=['c', '2', 'iv'])]
self.assertEqual(result, expected)
def test_example11(self):
text = u"Amend § 1005.36 to revise the section heading and "
text += "paragraphs (a) and (b), and to add paragraph (d) to read "
text += "as follows:"
result = parse_text(text)
self.assertEqual(result, [
tokens.Context(['1005', None, '36']),
tokens.Verb(tokens.Verb.PUT, active=True),
tokens.Paragraph([], field=tokens.Paragraph.HEADING_FIELD),
tokens.AndToken(),
tokens.TokenList([tokens.Paragraph(paragraph='a'),
tokens.Paragraph(paragraph='b')]),
tokens.AndToken(),
tokens.Verb(tokens.Verb.POST, active=True),
tokens.Paragraph(paragraph='d'),
])
def test_context_to_paragraph_exceptions(self):
tokenized = [
tokens.Verb(tokens.Verb.PUT, active=True),
tokens.Context(['2']),
tokens.Paragraph(['3'])
]
converted = context_to_paragraph(tokenized)
self.assertEqual(tokenized, converted)
tokenized = [
tokens.Verb(tokens.Verb.PUT, active=True),
tokens.Context(['2']),
tokens.TokenList([tokens.Paragraph(['3'])])
]
converted = context_to_paragraph(tokenized)
self.assertEqual(tokenized, converted)
def test_example8(self):
text = "2. On page 8765 through 8767, in Appendix A to Part 1234,"
text += "Model Forms A-15 through A-19 are corrected to read as "
text += "follows:"
result = parse_text(text)
self.assertEqual(result, [
tokens.Context(['1234', 'Appendix:A'], certain=True),
tokens.TokenList([
tokens.Paragraph([None, 'Appendix:A', '15']),
tokens.Paragraph([None, 'Appendix:A', '16']),
tokens.Paragraph([None, 'Appendix:A', '17']),
tokens.Paragraph([None, 'Appendix:A', '18']),
tokens.Paragraph([None, 'Appendix:A', '19'])
]),
tokens.Verb(tokens.Verb.PUT, active=False),
])
def test_example6(self):
"""Although this includes the term 'Commentary', we assume these are
not interpretations and handle the problem of merging later"""
text = u"a. Add new Commentary for §§ 6363.30, 6363.31, 6363.32,"
text += " 6363.33, 6363.34, 6363.35, and 6363.36."
result = parse_text(text)
self.assertEqual(result, [
tokens.Verb(tokens.Verb.POST, active=True),
tokens.TokenList([
tokens.Paragraph(['6363', None, '30']),
tokens.Paragraph(['6363', None, '31']),
tokens.Paragraph(['6363', None, '32']),
tokens.Paragraph(['6363', None, '33']),
tokens.Paragraph(['6363', None, '34']),
tokens.Paragraph(['6363', None, '35']),
tokens.Paragraph(['6363', None, '36']),
])
])
def curried(match=None):
pars = []
matches = [match.head] + list(match.tail)
for match in matches:
match_as_list = listify(match)
next_par = tokens.Paragraph(match_as_list)
next_lab = next_par.label
if match[-1] == 'text' or force_text_field:
next_par.field = tokens.Paragraph.TEXT_FIELD
if match.through:
# Iterate through, creating paragraph tokens
prev_lab = pars[-1].label
if '(' in prev_lab[-1] and '(' in next_lab[-1]:
pars.extend(_through_paren(prev_lab, next_lab))
elif len(prev_lab) == 3:
pars.extend(_through_sect(prev_lab, next_lab))
elif len(prev_lab) > 3:
pars.extend(_through_paragraph(prev_lab, next_lab))
pars.append(next_par)
return tokens.TokenList(pars)
def paragraph_token_list(self):
paragraph_tokens = [
tokens.Paragraph(['200', '1', 'a']),
tokens.Paragraph(['200', '1', 'b'])
]
return tokens.TokenList(paragraph_tokens)
def _paragraph_token_list():
return tokens.TokenList([
tokens.Paragraph.make(part='200', sub='1', section='a'),
tokens.Paragraph.make(part='200', sub='1', section='b')
])