def test_name_collisions(self):
with self.subTest(
"General and parameter entities occupy different namespaces"):
document = Document("")
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY % entity 'SOME TEXT'>
<!ENTITY entity 'SOME OTHER TEXT'>
]>""")
self.assertEqual(
"SOME TEXT",
document.parameter_entities["entity"].expansion_text)
self.assertEqual(
"SOME OTHER TEXT",
document.general_entities["entity"].expansion_text)
with self.subTest("General entities"):
document = Document("")
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY entity 'SOME TEXT'>
<!ENTITY entity 'SOME OTHER TEXT'>
]>""")
self.assertEqual(
"SOME TEXT",
document.general_entities["entity"].expansion_text)
with self.subTest("Parameter entities"):
document = Document("")
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY % entity 'SOME TEXT'>
<!ENTITY % entity 'SOME OTHER TEXT'>
]>""")
self.assertEqual(
"SOME TEXT",
document.parameter_entities["entity"].expansion_text)
def test_parameter_reference_within_markup(self):
with self.subTest("ELEMENT declaration"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration(
"""<!DOCTYPE root [
<!ENTITY % entity 'blah, blahblah'>'
<!ELEMENT root (%entity;)>
]>""")
with self.subTest("NOTATION declaration"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration(
"""<!DOCTYPE root [
<!ENTITY % entity 'SYSTEM \"blahblah\"'>'
<!NOTATION blah %entity;>
]>""")
with self.subTest("ATTLIST declaration"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration(
"""<!DOCTYPE root [
<!ENTITY % entity 'CDATA #REQUIRED'>'
<!ATTLIST root %entity;>
]>""")
def test_external_parameter_entity_definition(self):
with self.subTest("SYSTEM"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root [<!ENTITY % entity SYSTEM 'uri'>]>")
self.assertEqual("uri",
document.parameter_entities["entity"].system_URI)
self.assertEqual(Entity.Type.PARAMETER,
document.parameter_entities["entity"].type)
self.assertEqual(True,
document.parameter_entities["entity"].external)
self.assertEqual(True,
document.parameter_entities["entity"].parsed)
with self.subTest("PUBLIC"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root [<!ENTITY % entity PUBLIC 'uri1' 'uri2'>]>")
self.assertEqual("uri1",
document.parameter_entities["entity"].public_URI)
self.assertEqual("uri2",
document.parameter_entities["entity"].system_URI)
self.assertEqual(Entity.Type.PARAMETER,
document.parameter_entities["entity"].type)
self.assertEqual(True,
document.parameter_entities["entity"].external)
self.assertEqual(True,
document.parameter_entities["entity"].parsed)
def test_unparsed_entity_definition(self):
with self.subTest("SYSTEM"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root [<!ENTITY entity SYSTEM 'uri' NDATA notation>]>"
)
self.assertEqual("uri",
document.general_entities["entity"].system_URI)
self.assertEqual(Entity.Type.GENERAL,
document.general_entities["entity"].type)
self.assertEqual(True,
document.general_entities["entity"].external)
self.assertEqual(False, document.general_entities["entity"].parsed)
self.assertEqual("notation",
document.general_entities["entity"].notation)
with self.subTest("PUBLIC"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root [<!ENTITY entity PUBLIC 'uri1' 'uri2' NDATA notation>]>"
)
self.assertEqual("uri1",
document.general_entities["entity"].public_URI)
self.assertEqual("uri2",
document.general_entities["entity"].system_URI)
self.assertEqual(Entity.Type.GENERAL,
document.general_entities["entity"].type)
self.assertEqual(True,
document.general_entities["entity"].external)
self.assertEqual(False, document.general_entities["entity"].parsed)
self.assertEqual("notation",
document.general_entities["entity"].notation)
def test_root_name_parsing(self):
with self.subTest("With no subset"):
document = Document("")
document._Document__parse_doctype_declaration("<!DOCTYPE root>")
self.assertEqual("root", document.dtd_name)
with self.subTest("With empty internal subset"):
document = Document("")
document._Document__parse_doctype_declaration("<!DOCTYPE root []>")
self.assertEqual("root", document.dtd_name)
def test_without_xmldeclaration(self):
with self.subTest("Open tag"):
document = Document("<root></root>")
document.parse()
self.assertIsInstance(document.root, Element)
self.assertEqual(document.root.name, "root")
with self.subTest("Closed tag"):
document = Document("<root/>")
document.parse()
self.assertIsInstance(document.root, Element)
self.assertEqual(document.root.name, "root")
def test_forbidden_name_characters(self):
for char in ["\u00d7", "\u00f7", "\u037e", "\u2030"]:
with self.subTest(char):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration(
f"<!DOCTYPE root{char}>")
def test_missing_version(self):
with self.subTest("Encoding"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml encoding='utf-8'?>")
with self.subTest("Standalone"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml standalone='no'?>")
with self.subTest("Both"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml encoding='utf-8' standalone='yes'?>")
def test_forbidden_name_start_characters(self):
for char in ["-", ".", "0", "7", "\u0305"]:
with self.subTest(char):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration(
f"<!DOCTYPE {char}root>")
def test_no_text_in_internal_subset(self):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY % entity 'blah, blahblah'>'
some disallowed text
]>""")
def test_misordered_xml_declaration(self):
with self.subTest("Encoding -> Version"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml encoding='utf-8' version='1.0'?>")
with self.subTest("Version -> Standalone -> Encoding"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0' standalone='no' encoding='utf-8'?>")
with self.subTest("Standalone -> Version"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml standalone='no' version='1.0'?>")
def build_document(abstract, result):
reception_number = access_result_reception_number(result)
description_link = access_result_button(abstract, result)
return Document("document_number",
reception_number,
county=abstract.county,
description_link=description_link,
number_results=1)
def prepare_name_search():
name = title_strip(search_name)
if name == '':
name = request_new_name()
while request_yes_or_no(f'The current name to be searched is "{name}", is this correct?') is False:
clear_terminal()
name = request_new_name()
clear_terminal()
return Document(type="name", value=name)
def test_internal_parameter_entity_definition(self):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root [<!ENTITY % entity 'blah'>]>")
self.assertEqual("blah",
document.parameter_entities["entity"].expansion_text)
self.assertEqual(Entity.Type.PARAMETER,
document.parameter_entities["entity"].type)
self.assertEqual(False, document.parameter_entities["entity"].external)
def parse(xml: str) -> Document:
# Normalise whitespace
xml = xml.replace("\u000d\u000a", "\u000a")
xml = xml.replace("\u000d", "\u000a")
# Parse document
document = Document(xml)
document.parse()
return document
def test_standalone_parsing(self):
with self.subTest("With encoding"):
document = Document("")
document._Document__parse_xml_declaration(
"<?xml version='1.0' encoding='utf-8' standalone='no'?>")
self.assertEqual("1.0", document.version)
self.assertEqual("utf-8", document.encoding)
self.assertEqual(False, document.standalone)
with self.subTest("Without encoding"):
document = Document("")
document._Document__parse_xml_declaration(
"<?xml version='1.0' standalone='yes'?>")
self.assertEqual("1.0", document.version)
self.assertEqual(True, document.standalone)
with self.subTest("Invalid value"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0' standalone='maybe'?>")
def test_version_parsing(self):
with self.subTest("With space"):
document = Document("")
document._Document__parse_xml_declaration(
"<?xml version = '1.0' ?>")
self.assertEqual("1.0", document.version)
with self.subTest("Without space"):
document = Document("")
document._Document__parse_xml_declaration("<?xml version='1.0'?>")
self.assertEqual("1.0", document.version)
with self.subTest("Invalid version 2.0"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='2.0'?>")
with self.subTest("Valid version 1.1"):
document = Document("")
document._Document__parse_xml_declaration("<?xml version='1.1'?>")
self.assertEqual("1.1", document.version)
def test_external_declaration_parsing(self):
with self.subTest("SYSTEM"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root SYSTEM 'uri'>")
self.assertEqual("root", document.dtd_name)
self.assertEqual("uri", document.external_system_uri)
with self.subTest("PUBLIC"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root PUBLIC 'uri1' 'uri2'>")
self.assertEqual("root", document.dtd_name)
self.assertEqual("uri1", document.external_public_uri)
self.assertEqual("uri2", document.external_system_uri)
with self.subTest("SYSTEM with internal subset"):
document = Document("")
document._Document__parse_doctype_declaration(
"<!DOCTYPE root SYSTEM 'uri' []>")
self.assertEqual("root", document.dtd_name)
self.assertEqual("uri", document.external_system_uri)
def test_invalid_keyword(self):
with self.subTest("Versionn"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml versionn='1.0'?>")
with self.subTest("Versio"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml versio='1.0'?>")
with self.subTest("encodin"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0' encodin='utf-8?>")
with self.subTest("standalne"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0' standalne='no'?>")
def test_missing_whitespace(self):
with self.subTest("Before version"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xmlversion='1.0'?>")
with self.subTest("After version -> encoding"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0'encoding='utf-8'?>")
with self.subTest("After version -> standalone"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0'standalone='no'?>")
with self.subTest("After encoding -> standalone"):
document = Document("")
with self.assertRaises(XMLError):
document._Document__parse_xml_declaration(
"<?xml version='1.0' encoding='utf-8'standalone='no'?>")
def test_with_xmldeclaration(self):
document = Document(
"<?xml version='1.0' encoding='utf-8' standalone='yes'?> <root></root>"
)
document.parse()
self.assertEqual("1.0", document.version)
self.assertEqual("utf-8", document.encoding)
self.assertEqual(True, document.standalone)
self.assertIsInstance(document.root, Element)
self.assertEqual(document.root.name, "root")
def test_parameter_entity_expansion(self):
with self.subTest("In DTD"):
document = Document("")
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY % entity '<!ENTITY entity2 "SOME TEXT">'>
%entity;
]>""")
self.assertEqual(
"SOME TEXT",
document.general_entities["entity2"].expansion_text)
with self.subTest("In general entity"):
document = Document("")
document._Document__parse_doctype_declaration("""<!DOCTYPE root [
<!ENTITY % entity 'SOME TEXT'>
<!ENTITY entity2 'SOME %entity; OTHER TEXT'>
]>""")
self.assertEqual(
"SOME TEXT",
document.parameter_entities["entity"].expansion_text)
self.assertEqual(
"SOME SOME TEXT OTHER TEXT",
document.general_entities["entity2"].expansion_text)
def test_no_superfluous_characters(self):
with self.subTest("Before xml declaration"):
document = Document("some text<?xml version='1.0'?><root></root>")
with self.assertRaises(XMLError):
document.parse()
with self.subTest("Between xml declaration and dtd"):
document = Document(
"<?xml version='1.0'?>some text<!DOCTYPE root []><root></root>"
)
with self.assertRaises(XMLError):
document.parse()
with self.subTest("Between dtd and root element"):
document = Document(
"<?xml version='1.0'?><!DOCTYPE root []>some text<root></root>"
)
with self.assertRaises(XMLError):
document.parse()
with self.subTest("After root element"):
document = Document(
"<?xml version='1.0'?><!DOCTYPE root []><root></root>some text"
)
with self.assertRaises(XMLError):
document.parse()
def test_with_empty_dtd(self):
document = Document("""
<?xml version='1.0' encoding='utf-8' standalone='yes'?>
<!DOCTYPE root []>
<root></root>
""")
document.parse()
self.assertEqual("1.0", document.version)
self.assertEqual("utf-8", document.encoding)
self.assertEqual(True, document.standalone)
self.assertIsInstance(document.root, Element)
self.assertEqual(document.root.name, "root")
def index(files: List[str], empty_words: List[str]):
print("indexing...")
progress_bar = tqdm(total=len(files), unit="file")
for file_name in files:
doc = Document(file_name, {}, len(documents),
os.path.getsize(file_name))
documents.append(doc)
with open(file_name, encoding="utf-8", errors="ignore") as file:
tokens = tokenizar(file.read())
tokens = sacar_palabras_vacias(tokens, empty_words)
for token in tokens:
term = terms_doc_dic.get(token,
Term(token, set(), len(terms_doc_dic)))
terms_doc_dic[token] = term
doc.has_term(term)
term.found_in(doc)
progress_bar.update(1)
progress_bar.close()
def test_with_fluff(self):
document = Document("""
<?xml version='1.0' encoding='utf-8' standalone='yes'?>
<!DOCTYPE root []>
<?Target A rogue processing instruction?>
<!-- And even a comment! -->
<root></root>
<!-- Another comment -->
<?Target And more PIs?>
<?Target?>
""")
document.parse()
self.assertEqual("1.0", document.version)
self.assertEqual("utf-8", document.encoding)
self.assertEqual(True, document.standalone)
self.assertIsInstance(document.root, Element)
self.assertEqual(document.root.name, "root")
self.assertEqual(3, len(document.processing_instructions))
def test_only_one_root_element(self):
document = Document(
"<?xml version='1.0' encoding='utf-8' standalone='yes'?> <root1></root1> <root2></root2>"
)
with self.assertRaises(XMLError):
document.parse()
def transform_result_list(result_list):
document_list = []
for result in result_list:
document_list.append(Document(type="document_number", value=result))
return document_list
def test_encoding_parsing(self):
document = Document("")
document._Document__parse_xml_declaration(
"<?xml version='1.0' encoding='utf-8'?>")
self.assertEqual("1.0", document.version)
self.assertEqual("utf-8", document.encoding)
y_train = np.concatenate([y_train1, y_train2])
else:
data = pd.read_csv(dataset[args.dataset.lower()] + "train.csv")
X = np.array([unidecode(x) for x in data['x'].values])
y = np.array([int(i) - 1 for i in data['y'].values])
ids = range(0, len(X))
train, test = train_test_split(ids, test_size=0.1, random_state=42)
X_train, X_test, y_train, y_test = X[train], X[test], y[train], y[test]
hatebase = pd.read_csv('./data/hatebase/hatebase.csv').vocabulary.values
vectorizer = CountVectorizer(tokenizer=TreebankWordTokenizer().tokenize,
ngram_range=(1, 5),
preprocessor=Document().preprocess)
classifier = GridSearchCV(
LogisticRegression(penalty="l1", dual=False),
[{
'C': [0.0001, 0.001, 0.1, 1, 10, 100]
}] #range of C coefficients to try
)
X = vectorizer.fit_transform(X_train)
classifier.fit(X, y_train)
coeffs = classifier.best_estimator_.coef_
fn = [i.encode("utf-8") for i in vectorizer.get_feature_names()]
sfn = np.array(fn, dtype=np.str_)[coeffs.nonzero()[1]]
sfv = np.array(coeffs[coeffs.nonzero()], dtype=np.float32)