def test_match_labels_and_changes_candidate(self):
labels_amended = [
Amendment('POST', '200-2'),
Amendment('PUT', '200-2-a-1-i')
]
n1 = Node('n2', label=['200', '2'])
n2 = Node('n2a', label=['200', '2', 'i'])
n1.children = [n2]
root = Node('root', label=['200'], children=[n1])
amend_map = changes.match_labels_and_changes(labels_amended, root)
self.assertTrue(amend_map['200-2-a-1-i'][0]['candidate'])
self.assertTrue(amend_map['200-2-a-1-i'][0]['node'].label_id(),
'200-2-a-1-i')
def build_supplement_tree(reg_part, node):
""" Build the tree for the supplement section. """
title = get_app_title(node)
root = Node(node_type=Node.INTERP,
label=[reg_part, Node.INTERP_MARK],
title=title)
return parse_from_xml(root, node.getchildren())
def test_sub_comment(self):
text = "refer to comment 36(a)(2)-3 of thing"
result = self.parser.process(Node(text, label=['222', '87']))
self.assertEqual(1, len(result))
self.assertEqual(['222', '36', 'a', '2', Node.INTERP_MARK, '3'],
result[0]['citation'])
offsets = result[0]['offsets'][0]
self.assertEqual('36(a)(2)-3', text[offsets[0]:offsets[1]])
def test_resolve_candidates_double_delete(self):
"""In the unfortunate case where *two* candidates are wrong make
sure we don't blow up"""
amend_map = {}
n1 = Node('n1', label=['200', '1', 'i'])
n2 = Node('n2', label=['200', '1', 'i'])
amend_map['200-1-a-i'] = [{
'node': n1,
'candidate': True
}, {
'node': n2,
'candidate': True
}]
amend_map['200-1-i'] = []
changes.resolve_candidates(amend_map, warn=False)
self.assertEqual(1, len(amend_map.keys()))
def test_single_section_citation(self):
""" Ensure that offsets work correctly in a simple single section
citation case. """
text = u"date in § 1005.20(h)(1) must disclose"
citations = self.parser.process(Node(text, label=['1005', '6']))
c = citations[0]
self.assertEquals(text[c['offsets'][0][0]:c['offsets'][0][1]],
u'1005.20(h)(1)')
def test_comment_header(self):
text = "See comment 32(b)(3) blah blah"
result = self.parser.process(Node(text, label=['222', '87']))
self.assertEqual(1, len(result))
self.assertEqual(['222', '32', 'b', '3', Node.INTERP_MARK],
result[0]['citation'])
offsets = result[0]['offsets'][0]
self.assertEqual('32(b)(3)', text[offsets[0]:offsets[1]])
def hed(self, part, text):
"""HD with an HED source indicates the root of the appendix"""
n = Node(node_type=Node.APPENDIX,
label=[part, self.appendix_letter],
title=text)
self.m_stack.push_last((0, n))
self.paragraph_counter = 0
self.depth = 0
def test_process_cfr(self):
m = Meta(None, cfr_title=3)
result = m.process(Node(label=['a']))
self.assertEqual(1, len(result))
self.assertTrue('cfr_title_number' in result[0])
self.assertEqual(3, result[0]['cfr_title_number'])
self.assertTrue('cfr_title_text' in result[0])
self.assertEqual('The President', result[0]['cfr_title_text'])
def graphic(self, xml_node):
self.paragraph_counter += 1
gid = xml_node.xpath('./GID')[0].text
text = ''
n = Node(text,
node_type=Node.APPENDIX,
label=['p' + str(self.paragraph_counter)])
self.nodes.append(n)
def test_build_tree_exclude(self):
"""Paragraph tree should not split on exclude areas."""
ref = "Ref (b)(2)"
text = "This (a) is a good (1) %s test (2) no?" % ref
ref_pos = text.find(ref)
self.assertEqual(
self.regParser.build_tree(text,
exclude=[(ref_pos, ref_pos + len(ref))]),
Node("This ",
children=[
Node("(a) is a good ",
label=['a'],
children=[
Node("(1) %s test " % ref, label=['a', '1']),
Node("(2) no?", label=['a', '2'])
])
]))
def test_cfr(self):
"""Ensure that we successfully parse CFR references."""
node = Node("Ref 1: 12 CFR part 1026. " + "Ref 2: 12 CFR 1026.13.")
parser = external_citations.ExternalCitationParser(None)
citations = parser.process(node)
self.assertEqual(2, len(citations))
self.assertEqual("CFR", citations[0]['citation_type'])
self.assertEqual("CFR", citations[1]['citation_type'])
def test_process_subparagraph_of_referenced_text():
root = Node(
label=['100'],
children=[
Node("\n\n\n",
node_type=Node.INTERP,
label=['100', '11', 'a', Node.INTERP_MARK],
children=[
Node("Interp11a1",
node_type=Node.INTERP,
label=['100', '11', 'a', '1', Node.INTERP_MARK])
])
])
interp = Interpretations(root)
interp.pre_process()
assert interp.process(Node(label=['100', '11', 'a'])) is None
assert interp.process(Node(label=['100', '11', 'a', '1'])) is not None
def test_build_hierarchy(self):
"""Nodes should be returned at the provided depths"""
root = Node(label=['root'])
nodes = [Node(label=['a']), Node(label=['1']), Node(label=['2']),
Node(label=['i']), Node(label=['b']), Node(label=['c'])]
depths = [ParAssignment(mtypes.lower, 0, 0),
ParAssignment(mtypes.ints, 0, 1),
ParAssignment(mtypes.ints, 1, 1),
ParAssignment(mtypes.roman, 0, 2),
ParAssignment(mtypes.lower, 1, 0),
ParAssignment(mtypes.lower, 2, 0)]
result = _ExampleProcessor().build_hierarchy(root, nodes, depths)
self.assertEqual(result.label, ['root'])
self.assertEqual(len(result.children), 3)
a, b, c = result.children
self.assertEqual(a.label, ['root', 'a'])
self.assertEqual(len(a.children), 2)
self.assertEqual(b.label, ['root', 'b'])
self.assertEqual(len(b.children), 0)
self.assertEqual(c.label, ['root', 'c'])
self.assertEqual(len(c.children), 0)
a1, a2 = a.children
self.assertEqual(a1.label, ['root', 'a', '1'])
self.assertEqual(len(a1.children), 0)
self.assertEqual(a2.label, ['root', 'a', '2'])
self.assertEqual(len(a2.children), 1)
self.assertEqual(a2.children[0].label, ['root', 'a', '2', 'i'])
def test_toc_with_subparts(self):
c1 = Node(label=['205', '2'], title='Authority and Purpose')
c2 = Node(label=['205', '3'], title='Definitions')
c3 = Node(label=['205', '4'], title='Coverage')
s1 = Node(children=[c1, c2, c3],
label=['205', 'Subpart', 'A'],
title='First Subpart',
node_type=Node.SUBPART)
c4 = Node(label=['205', '5'], title='Fifth Title')
s2 = Node(children=[c4],
label=['205', 'Subpart', 'B'],
title='Second Subpart',
node_type=Node.SUBPART)
n = Node(children=[s1, s2], label=['205'])
parser = table_of_contents.TableOfContentsLayer(None)
toc = parser.process(n)
self.assertEqual(len(toc), 2)
self.assertEqual(toc[0]['index'], ['205', 'Subpart', 'A'])
self.assertEqual(toc[1]['index'], ['205', 'Subpart', 'B'])
toc = parser.process(s1)
self.assertEqual(len(toc), 3)
self.assertEqual(toc[0]['index'], ['205', '2'])
self.assertEqual(toc[1]['index'], ['205', '3'])
self.assertEqual(toc[2]['index'], ['205', '4'])
toc = parser.process(s2)
self.assertEqual(len(toc), 1)
self.assertEqual(toc[0]['index'], ['205', '5'])
def test_process_extra(self):
settings.META = {'some': 'setting', 'then': 42}
m = Meta(None, cfr_title=19, version=None)
result = m.process(Node(label=['a']))
self.assertEqual(1, len(result))
self.assertTrue('some' in result[0])
self.assertEqual('setting', result[0]['some'])
self.assertTrue('then' in result[0])
self.assertEqual(42, result[0]['then'])
def test_overwrite_marker(self):
n = Node(
'3. Interpretation paragraph text.',
label=['205', '2', 'a', 'Interp', '3'],
node_type=Node.INTERP)
changed = compiler.overwrite_marker(n, '2')
self.assertTrue('2.' in changed.text)
self.assertFalse('3.' in changed.text)
def test_keyterm_definition(self):
node = Node("(a) Terminator means I'll be back",
label=['101', '22', 'a'])
node.tagged_text = """(a) <E T="03">Terminator</E> means I'll be """
node.tagged_text += 'back'
kt = KeyTerms(None)
results = kt.process(node)
self.assertEqual(results, None)
node = Node("(1) Act means pretend", label=['101', '22', 'a', '1'])
node.tagged_text = """(1) <E T="03">Act</E> means pretend"""
node = Node(
"(1) Act means the Truth in Lending Act (15 U.S.C. 1601 et seq.).",
label=['1026', '2', 'a', '1'])
node.tagged_text = """(1) <E T="03">Act</E> means the Truth in Lending Act (15 U.S.C. 1601 <E T="03">et seq.</E>)."""
kt = KeyTerms(None)
results = kt.process(node)
self.assertEqual(results, None)
def test_reserve_add_new(self):
root = self.tree_with_paragraphs()
reg_tree = compiler.RegulationTree(root)
n2ai = Node('[Reserved]', label=['205', '2', 'a', '1'])
reg_tree.reserve('205-2-a-1', n2ai)
self.assertNotEqual(reg_tree.tree, root)
reserved_node = find(reg_tree.tree, '205-2-a-1')
self.assertEqual(reserved_node.text, '[Reserved]')
def test_get_section_parent(self):
root = self.tree_with_subparts()
section = Node('section', label=['205', '3'], node_type=Node.REGTEXT)
subpart = find(root, '205-Subpart-B')
subpart.children = [section]
reg_tree = compiler.RegulationTree(root)
parent = reg_tree.get_section_parent(section)
self.assertEqual(parent.label_id(), '205-Subpart-B')
def test_sub_comment2(self):
text = "See comment 3(b)(1)-1.v."
result = self.parser.process(Node(text, label=['222', '87']))
self.assertEqual(1, len(result))
self.assertEqual(['222', '3', 'b', '1', Node.INTERP_MARK, '1', 'v'],
result[0]['citation'])
offsets = result[0]['offsets'][0]
# Note the final period is not included
self.assertEqual('3(b)(1)-1.v', text[offsets[0]:offsets[1]])
def test_single_paragraph(self):
""" Ensure that offsets work correctly in a simple single paragraph
citation. """
text = 'The requirements in paragraph (a)(4)(iii) of'
citations = self.parser.process(Node(text, label=['1005', '6']))
c = citations[0]
self.assertEquals(text[c['offsets'][0][0]:c['offsets'][0][1]],
u'(a)(4)(iii)')
self.assertEquals(['1005', '6', 'a', '4', 'iii'], c['citation'])
def test_match_labels_and_changes_reserve():
labels_amended = [Amendment('RESERVE', '200-?-2-a')]
amend_map = changes.match_labels_and_changes(
labels_amended, _section_node())
assert set(amend_map.keys()) == {'200-2-a'}
amendments = amend_map['200-2-a']
assert amendments[0]['action'] == 'RESERVE'
assert amendments[0]['node'] == Node('n2a', label=['200', '2', 'a'])
def test_process_subscript(self):
node = Node("This is a_{subscript}. And then a_{subscript} again")
result = formatting.Formatting(None).process(node)
self.assertEqual(1, len(result))
result = result[0]
self.assertEqual(result['text'], "a_{subscript}")
self.assertEqual(result['locations'], [0, 1])
self.assertEqual(result['subscript_data'],
{'variable': 'a', 'subscript': 'subscript'})
def generic_tree(text, label, title=None):
"""Use the "generic" parser to build a tree. The "generic" parser simply
splits on Title Case and treats body text as the node content."""
segments = generic.segments(text)
if not segments:
return Node(text, label=label, title=title, node_type=Node.APPENDIX)
children = []
for index, seg in enumerate(segments):
start, end = seg
seg_title, body = utils.title_body(text[start:end])
label_character = letter_for(index)
children.append(
Node(body, label=(
label + [label_character]),
title=seg_title, node_type=Node.APPENDIX))
return Node(text[:segments[0][0]], children, label, title, Node.APPENDIX)
def test_determine_scope_this_part(self):
"""Definitions scoped to a part also cover the interpretations for
that part"""
self.add_nodes(1)
self.stack.add(
1,
Node('For the purposes of this part, blah blah',
label=['1001', '3']))
self.assert_scope(('1001', ), ('1001', Node.INTERP_MARK))
def test_process_format(self):
node = Node("")
g = Graphics(None)
self.expect_http(uri='http://example.com/ER22MY13.257-1.gif',
method='HEAD')
self.expect_http(uri='http://example.com/ER22MY13.257-1.thumb.gif',
method='HEAD')
self.assertEqual(1, len(g.process(node)))
def test_determine_scope_this_paragraph(self):
"""Paragraph scope is tied to the paragraph that determined it.
Previous paragraph scopes won't apply to adjacent children"""
self.add_nodes(2)
self.stack.add(2, Node('For the purposes of this section, blah blah',
label=['1000', '3', 'd']))
self.stack.add(3, Node('For the purposes of this paragraph, blah blah',
label=['1000', '3', 'd', '5']))
self.assert_scope(('1000', '3', 'd', '5'),
('1000', '3', 'd', '5', Node.INTERP_MARK))
self.stack.add(3, Node(label=['1002', '3', 'd', '6']))
self.assert_scope(('1000', '3'), ('1000', '3', Node.INTERP_MARK))
self.stack.add(3, Node('Blah as used in this paragraph, blah blah',
label=['1000', '3', 'd', '7']))
self.assert_scope(('1000', '3', 'd', '7'),
('1000', '3', 'd', '7', Node.INTERP_MARK))
def build_supplement_tree(reg_part, node):
""" Build the tree for the supplement section. """
title = get_app_title(node)
node = spaces_then_remove(deepcopy(node), 'PRTPAGE')
root = Node(node_type=Node.INTERP,
label=[reg_part, Node.INTERP_MARK],
title=title)
return parse_from_xml(root, node.getchildren())
def derive_nodes(self, xml, processor=None):
texts = ["```" + xml.get('LANGUAGE', 'code')]
for child in xml:
text = tree_utils.get_node_text(child).strip()
if text:
texts.append(text)
texts.append("```")
return [Node("\n".join(texts), label=[mtypes.MARKERLESS])]
def test_pre_process_defined_twice(self):
tree = Node(u"The term “lol” means laugh out loud. " +
u"How do you pronounce “lol”, though?",
label=['1212', '5'])
t = Terms(tree)
t.pre_process()
self.assertEqual(t.layer['referenced']['lol:1212-5']['position'],
(10, 13))