def test_class_paths_to_classes(self):
"""A list of class paths should get converted into a list of
classes"""
results = plugins.class_paths_to_classes(self._FORWARD)
self.assertEqual(results, [Meta, ExternalCitationParser])
results = plugins.class_paths_to_classes(self._BACKWARD)
self.assertEqual(results, [ExternalCitationParser, Meta])
def preprocess(self):
"""Unfortunately, the notice xml is often inaccurate. This function
attempts to fix some of those (general) flaws. For specific issues, we
tend to instead use the files in settings.LOCAL_XML_PATHS"""
for preprocessor in class_paths_to_classes(settings.PREPROCESSORS):
preprocessor().transform(self.xml)
return self
def preprocess(self):
"""Unfortunately, the notice xml is often inaccurate. This function
attempts to fix some of those (general) flaws. For specific issues, we
tend to instead use the files in settings.LOCAL_XML_PATHS"""
for preprocessor in class_paths_to_classes(settings.PREPROCESSORS):
preprocessor().transform(self.xml)
return self
def tree_and_builder(filename, title, checkpoint_path=None, doc_number=None):
"""Reads the regulation file and parses it. Returns the resulting tree as
well as a Builder object for further manipulation. Looks up the doc_number
if it's not provided"""
make_fake = doc_number is not None
if checkpoint_path is None:
checkpointer = NullCheckpointer()
else:
checkpointer = Checkpointer(checkpoint_path)
reg_text = ''
with codecs.open(filename, 'r', 'utf-8') as f:
reg_text = f.read()
file_digest = hashlib.sha256(reg_text.encode('utf-8')).hexdigest()
if reg_text[:1] == '<':
reg_xml = etree.fromstring(reg_text)
else:
raise ValueError("Building from text input is no longer supported")
for preprocessor in class_paths_to_classes(settings.PREPROCESSORS):
preprocessor().transform(reg_xml)
reg_tree = checkpointer.checkpoint(
"init-tree-" + file_digest,
lambda: xml_parser.reg_text.build_tree(reg_xml))
title_part = reg_tree.label_id()
if doc_number is None:
doc_number = checkpointer.checkpoint(
"doc-number-" + file_digest,
lambda: Builder.determine_doc_number(reg_xml, title, title_part))
if not doc_number:
raise ValueError("Could not determine document number")
checkpointer.suffix = ":".join(["", title_part, str(title), doc_number])
if make_fake:
fake_notice = notice_fake.build(
doc_number, notice_fake.effective_date_for(reg_xml), title,
title_part)
else:
fake_notice = None
builder = Builder(cfr_title=title,
cfr_part=title_part,
doc_number=doc_number,
checkpointer=checkpointer,
fake_notice=fake_notice)
return reg_tree, builder
def tree_and_builder(filename, title, checkpoint_path=None, doc_number=None):
"""Reads the regulation file and parses it. Returns the resulting tree as
well as a Builder object for further manipulation. Looks up the doc_number
if it's not provided"""
make_fake = doc_number is not None
if checkpoint_path is None:
checkpointer = NullCheckpointer()
else:
checkpointer = Checkpointer(checkpoint_path)
reg_text = ''
with codecs.open(filename, 'r', 'utf-8') as f:
reg_text = f.read()
file_digest = hashlib.sha256(reg_text.encode('utf-8')).hexdigest()
if reg_text[:1] == '<':
reg_xml = etree.fromstring(reg_text)
else:
raise ValueError("Building from text input is no longer supported")
for preprocessor in class_paths_to_classes(settings.PREPROCESSORS):
preprocessor().transform(reg_xml)
reg_tree = checkpointer.checkpoint(
"init-tree-" + file_digest,
lambda: xml_parser.reg_text.build_tree(reg_xml))
title_part = reg_tree.label_id()
if doc_number is None:
doc_number = checkpointer.checkpoint(
"doc-number-" + file_digest,
lambda: Builder.determine_doc_number(reg_xml, title, title_part))
if not doc_number:
raise ValueError("Could not determine document number")
checkpointer.suffix = ":".join(["", title_part, str(title), doc_number])
if make_fake:
fake_notice = notice_fake.build(
doc_number, notice_fake.effective_date_for(reg_xml), title,
title_part)
else:
fake_notice = None
builder = Builder(cfr_title=title,
cfr_part=title_part,
doc_number=doc_number,
checkpointer=checkpointer,
fake_notice=fake_notice)
return reg_tree, builder
