def test_numeric_exception(self):
bad_ids = [u"b002", u"d", u"c", u"a1"]
idsa = IDSortingAlgorithm(IDSortingAlgorithm.NUMERIC)
sids = idsa.sort(bad_ids)
self.assertTrue(sids == bad_ids)
def _read_unparsed(self, lines, parameters):
"""
Read text fragments from an unparsed format text file.
:param lines: the lines of the unparsed text file
:type lines: list of strings
:param parameters: additional parameters for parsing
(e.g., class/id regex strings)
:type parameters: dict
"""
#
# TODO better and/or parametric parsing,
# for example, removing tags but keeping text, etc.
#
self._log("Parsing fragments from unparsed text format")
pairs = []
# get filter attributes
attributes = dict()
if gc.PPN_JOB_IS_TEXT_UNPARSED_CLASS_REGEX in parameters:
class_regex_string = parameters[gc.PPN_JOB_IS_TEXT_UNPARSED_CLASS_REGEX]
if class_regex_string is not None:
self._log(["Regex for class: '%s'", class_regex_string])
class_regex = re.compile(r".*\b" + class_regex_string + r"\b.*")
attributes['class'] = class_regex
if gc.PPN_JOB_IS_TEXT_UNPARSED_ID_REGEX in parameters:
id_regex_string = parameters[gc.PPN_JOB_IS_TEXT_UNPARSED_ID_REGEX]
if id_regex_string is not None:
self._log(["Regex for id: '%s'", id_regex_string])
id_regex = re.compile(r".*\b" + id_regex_string + r"\b.*")
attributes['id'] = id_regex
# get id sorting algorithm
id_sort = IDSortingAlgorithm.UNSORTED
if gc.PPN_JOB_IS_TEXT_UNPARSED_ID_SORT in parameters:
id_sort = parameters[gc.PPN_JOB_IS_TEXT_UNPARSED_ID_SORT]
self._log(["Sorting text fragments using '%s'", id_sort])
# transform text in a soup object
self._log("Creating soup")
soup = BeautifulSoup.BeautifulSoup("\n".join(lines))
# extract according to class_regex and id_regex
text_from_id = dict()
ids = []
self._log(["Finding elements matching attributes '%s'", attributes])
nodes = soup.findAll(attrs=attributes)
for node in nodes:
try:
f_id = node['id']
f_text = node.text
text_from_id[f_id] = f_text
ids.append(f_id)
except KeyError:
self._log("KeyError while parsing a node", Logger.WARNING)
# sort by ID as requested
self._log("Sorting text fragments")
sorted_ids = IDSortingAlgorithm(id_sort).sort(ids)
# append to fragments
self._log("Appending fragments")
for key in sorted_ids:
pairs.append([key, [text_from_id[key]]])
self._create_text_fragments(pairs)
def test_numeric(self):
expected = [u"b001", u"a2", u"c03", u"d4"]
idsa = IDSortingAlgorithm(IDSortingAlgorithm.NUMERIC)
sids = idsa.sort(self.IDS)
self.assertTrue(sids == expected)
def test_lexicographic(self):
expected = [u"a2", u"b001", u"c03", u"d4"]
idsa = IDSortingAlgorithm(IDSortingAlgorithm.LEXICOGRAPHIC)
sids = idsa.sort(self.IDS)
self.assertTrue(sids == expected)
def test_unsorted(self):
expected = [u"b001", u"c03", u"d4", u"a2"]
idsa = IDSortingAlgorithm(IDSortingAlgorithm.UNSORTED)
sids = idsa.sort(self.IDS)
self.assertTrue(sids == expected)
def test_invalid_algorithm(self):
with self.assertRaises(ValueError):
idsa = IDSortingAlgorithm(u"foo")
def _read_unparsed(self, lines):
"""
Read text fragments from an unparsed format text file.
:param list lines: the lines of the unparsed text file
"""
from bs4 import BeautifulSoup
def filter_attributes():
""" Return a dict with the bs4 filter parameters """
attributes = {}
for attribute_name, filter_name in [
("class", gc.PPN_TASK_IS_TEXT_UNPARSED_CLASS_REGEX),
("id", gc.PPN_TASK_IS_TEXT_UNPARSED_ID_REGEX)
]:
if filter_name in self.parameters:
regex_string = self.parameters[filter_name]
if regex_string is not None:
self.log([
u"Regex for %s: '%s'", attribute_name, regex_string
])
regex = re.compile(r".*\b" + regex_string + r"\b.*")
attributes[attribute_name] = regex
return attributes
#
# TODO better and/or parametric parsing,
# for example, removing tags but keeping text, etc.
#
self.log(u"Parsing fragments from unparsed text format")
# transform text in a soup object
self.log(u"Creating soup")
soup = BeautifulSoup("\n".join(lines), "lxml")
# extract according to class_regex and id_regex
text_from_id = {}
ids = []
filter_attributes = filter_attributes()
self.log(
[u"Finding elements matching attributes '%s'", filter_attributes])
nodes = soup.findAll(attrs=filter_attributes)
for node in nodes:
try:
f_id = gf.safe_unicode(node["id"])
f_text = gf.safe_unicode(node.text)
text_from_id[f_id] = f_text
ids.append(f_id)
except KeyError:
self.log_warn(u"KeyError while parsing a node")
# sort by ID as requested
id_sort = gf.safe_get(dictionary=self.parameters,
key=gc.PPN_TASK_IS_TEXT_UNPARSED_ID_SORT,
default_value=IDSortingAlgorithm.UNSORTED,
can_return_none=False)
self.log([u"Sorting text fragments using '%s'", id_sort])
sorted_ids = IDSortingAlgorithm(id_sort).sort(ids)
# append to fragments
self.log(u"Appending fragments")
self._create_text_fragments([(key, [text_from_id[key]])
for key in sorted_ids])
