def build_reg_text_tree(text, part):
"""Build up the whole tree from the plain text of a single regulation. This
only builds the regulation text part, and does not include appendices or
the supplement. """
title, body = utils.title_body(text)
label = [str(part)]
subparts_list = []
subpart_locations = subparts(body)
if subpart_locations:
pre_subpart = body[:subpart_locations[0][0]]
first_emptypart, children_text = build_subparts_tree(
pre_subpart, part, build_empty_part)
if pre_subpart.strip() and first_emptypart.children:
subparts_list.append(first_emptypart)
else:
children_text = pre_subpart
for start, end in subpart_locations:
subpart_body = body[start:end]
subpart, _ = build_subparts_tree(
subpart_body, part, lambda p: build_subpart(subpart_body, p))
subparts_list.append(subpart)
else:
emptypart, children_text = build_subparts_tree(body, part,
build_empty_part)
if emptypart.children:
subparts_list.append(emptypart)
else:
return struct.Node(text, [build_empty_part(part)], label, title)
return struct.Node(children_text, subparts_list, label, title)
def build_reg_text_tree(text, part):
"""Build up the whole tree from the plain text of a single regulation. This
only builds the regulation text part, and does not include appendices or
the supplement. """
title, body = utils.title_body(text)
label = [str(part)]
subparts_list = []
subpart_locations = subparts(body)
if subpart_locations:
pre_subpart = body[:subpart_locations[0][0]]
first_emptypart, children_text = build_subparts_tree(
pre_subpart, part, build_empty_part)
if pre_subpart.strip() and first_emptypart.children:
subparts_list.append(first_emptypart)
else:
children_text = pre_subpart
for start, end in subpart_locations:
subpart_body = body[start:end]
subpart, _ = build_subparts_tree(
subpart_body, part, lambda p: build_subpart(subpart_body, p))
subparts_list.append(subpart)
else:
emptypart, children_text = build_subparts_tree(
body, part, build_empty_part)
if emptypart.children:
subparts_list.append(emptypart)
else:
return struct.Node(
text, [build_empty_part(part)], label, title)
return struct.Node(children_text, subparts_list, label, title)
def segment_tree(text, part, parent_label):
"""Build a tree representing the interpretation of a section, paragraph,
or appendix."""
title, body = utils.title_body(text)
exclude = [(s, e) for _, s, e in comment_citation.scanString(body)]
label = text_to_label(title, part)
return interpParser.build_tree(body, 1, exclude, label, title)
def segment_tree(text, part, parent_label):
"""Build a tree representing the interpretation of a section, paragraph,
or appendix."""
title, body = utils.title_body(text)
exclude = [(pc.full_start, pc.full_end) for pc in
internal_citations(body, Label(part=parent_label[0]))]
label = merge_labels(text_to_labels(title, Label(part=part, comment=True)))
return interpParser.build_tree(body, 1, exclude, label, title)
def segment_tree(text, part, parent_label):
"""Build a tree representing the interpretation of a section, paragraph,
or appendix."""
title, body = utils.title_body(text)
exclude = [(pc.full_start, pc.full_end) for pc in
internal_citations(body, Label(part=parent_label[0]))]
label = merge_labels(text_to_labels(title, Label(part=part, comment=True)))
return interpParser.build_tree(body, 1, exclude, label, title)
def build_section_tree(text, part):
"""Construct the tree for a whole section. Assumes the section starts
with an identifier"""
title, text = utils.title_body(text)
exclude = [(pc.full_start, pc.full_end) for pc in
internal_citations(text, Label(part=part))]
section = re.search(r'%d\.(\d+)\b' % part, title).group(1)
label = [str(part), section]
p_tree = regParser.build_tree(
text, exclude=exclude, label=label, title=title)
return p_tree
def build(text, part):
"""Create a tree representing the whole interpretation."""
part = str(part)
title, body = utils.title_body(text)
segments = segment_by_header(body, part)
if segments:
children = [segment_tree(body[s:e], part, [part]) for s, e in segments]
return Node(body[:segments[0][0]], treeify(children),
[part, Node.INTERP_MARK], title, Node.INTERP)
else:
return Node(body, [], [part, Node.INTERP_MARK], title, Node.INTERP)
def build_section_tree(text, part):
"""Construct the tree for a whole section. Assumes the section starts
with an identifier"""
title, text = utils.title_body(text)
exclude = [(pc.full_start, pc.full_end)
for pc in internal_citations(text, Label(part=part))]
section = re.search(r'%d\.(\d+)\b' % part, title).group(1)
label = [str(part), section]
p_tree = regParser.build_tree(text,
exclude=exclude,
label=label,
title=title)
return p_tree
def build_section_tree(text, part):
"""Construct the tree for a whole section. Assumes the section starts
with an identifier"""
title, text = utils.title_body(text)
exclude = [(start, end) for _, start, end in
regtext_citation.scanString(text)]
exclude += [(start, end) for _, start, end in
appendix_citation.scanString(text)]
section = re.search(r'%d\.(\d+)\b' % part, title).group(1)
label = [str(part), section]
p_tree = regParser.build_tree(
text, exclude=exclude, label=label, title=title)
return p_tree
def build(text, part):
"""Create a tree representing the whole interpretation."""
part = str(part)
title, body = utils.title_body(text)
segments = segment_by_header(body, part)
if segments:
children = [segment_tree(body[s:e], part, [part]) for s, e in segments]
return Node(
body[:segments[0][0]], treeify(children),
[part, Node.INTERP_MARK], title, Node.INTERP)
else:
return Node(
body, [], [part, Node.INTERP_MARK], title,
Node.INTERP)
def trees_from(text, part, parent_label):
"""Build a tree for the appendix section. It will have children for each
appendix. Text is the text of the entire regulation, while part is the
regulation's part (e.g. 1520.)"""
children = []
for begin, end in carving.appendices(text):
title, appendix = utils.title_body(text[begin:end])
appendix_letter = carving.get_appendix_letter(title, part)
label = parent_label + [appendix_letter]
sections = carving.appendix_sections(appendix, appendix_letter)
if sections:
child = paragraph_tree(
appendix_letter, sections, appendix, label, title)
else:
child = generic_tree(appendix, label, title)
children.append(child)
return children
def trees_from(text, part, parent_label):
"""Build a tree for the appendix section. It will have children for each
appendix. Text is the text of the entire regulation, while part is the
regulation's part (e.g. 1520.)"""
children = []
for begin, end in carving.appendices(text):
title, appendix = utils.title_body(text[begin:end])
appendix_letter = carving.get_appendix_letter(title, part)
label = parent_label + [appendix_letter]
sections = carving.appendix_sections(appendix, appendix_letter)
if sections:
child = paragraph_tree(appendix_letter, sections, appendix, label,
title)
else:
child = generic_tree(appendix, label, title)
children.append(child)
return children
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 paragraph_tree(appendix_letter, sections, text, label, title=None):
"""Use the paragraph parser to parse through each section in this
appendix."""
if not sections:
return Node(text, label=label, title=title, node_type=Node.APPENDIX)
children = []
for begin, end in sections:
seg_title, section_text = utils.title_body(text[begin:end])
sec_num = carving.get_appendix_section_number(
seg_title, appendix_letter)
exclude = [(pc.full_start, pc.full_end) for pc in
internal_citations(section_text, Label(part=label[0]))]
child = parParser.build_tree(
section_text, exclude=exclude, label=label + [sec_num],
title=seg_title)
children.append(child)
return Node(text[:sections[0][0]], children, label, title, Node.APPENDIX)
def test_title_body_normal_case(self):
title = "This is a title"
body = "Here is text that follows\nnewlines\n\n\nabout in the body"
self.assertEqual((title, "\n" + body),
utils.title_body(title + "\n" + body))
def test_title_body_title_only(self):
text = "This is some long, long title with no body"
self.assertEqual((text, ""), utils.title_body(text))
def test_title_body_title_only(self):
text = "This is some long, long title with no body"
self.assertEqual((text, ""), utils.title_body(text))
def test_title_body_normal_case(self):
title = "This is a title"
body = "Here is text that follows\nnewlines\n\n\nabout in the body"
self.assertEqual((title, "\n" + body),
utils.title_body(title + "\n" + body))