def test_add_html_classes():
doc = Document("<html><body/></html>")
transformation = Transformation(
Rule("body", lib.add_html_classes("transformed")))
result = transformation(doc).root
assert result[0].attributes["class"] == "transformed"
doc = Document('<html><body class="loaded" /></html>')
result = transformation(doc).root
assert all(x in result[0].attributes["class"]
for x in ("transformed", "loaded"))
transformation = Transformation(
Rule("body", lib.add_html_classes("transformed",
"and_something_else")))
result = transformation(doc).root
assert all(x in result[0].attributes["class"]
for x in ("and_something_else", "loaded", "transformed"))
transformation = Transformation(
Rule("body", lib.add_html_classes(Ref("html_classes"))),
context={"html_classes": ["transformed", "and_something_else"]},
)
result = transformation(doc).root
assert all(x in result[0].attributes["class"]
for x in ("and_something_else", "loaded", "transformed"))
def test_wikipedia_example_1():
expected = Document("""
<root>
<name username="******">John</name>
<name username="******">Morka</name>
</root>
""")
def extract_person(node: TagNode):
return node.attributes["username"], first(
node.css_select("name")).full_text
def append_person(previous_result, result: TagNode):
result.append_child(
result.new_tag_node(
"name",
attributes={"username": previous_result[0]},
children=[previous_result[1]],
))
transformation = Transformation(
Rule("person", (extract_person, append_person)),
result_object="context.result",
context={"result": new_tag_node("root")},
)
# that's four lines less LOC than the XSLT implementation
assert equal_subtree(transformation(wp_document.root), expected.root)
def transform_toc(routes, toc_order, src_to_title, root_dir, epub_zip,
section_title, filepath):
context = locals().copy()
context.pop('epub_zip', None)
context.pop('toc_order', None)
transformation = Transformation(inxs.lib.init_elementmaker(
name='elmaker', ),
make_toc_skeleton,
Rule('navPoint', set_titles),
Rule(
Any(
MatchesAttributes({'href': None}),
MatchesAttributes({'src': None}),
),
route_url,
),
make_toc,
list_contents,
indent,
result_object='context.html',
context=context)
with epub_zip.open(os.path.join(root_dir, filepath)) as doc_xml:
parser = etree.XMLParser(remove_blank_text=True)
doc_tree = etree.parse(doc_xml, parser)
root = doc_tree.getroot()
return transformation(
root,
src_to_title=src_to_title,
toc_order=toc_order,
)
def test_strip_namespace():
namespace = 'http://www.example.org/ns/'
e = builder.ElementMaker(namespace=namespace, nsmap={'x': namespace})
root = e.div()
t = Transformation(Rule(namespace, lib.strip_namespace))
result = t(root)
assert result.tag == 'div'
def test_attributes(constraint):
document = Document('<root><a b="x"/><a b="y"/></root>')
transformation = Transformation(Rule(constraint, lib.set_text("x")))
result = transformation(document).root
assert not result._data_node._exists
assert result[0].full_text == "x", str(result)
assert not len(result[1])
def test_merge():
source = etree.fromstring('<root><a><bb/></a><b/></root>')
destination = etree.fromstring('<root><a><aa/></a><b/><c/></root>')
expected = etree.fromstring('<root><a><aa/><bb/></a><b/><c/></root>')
transformation = Transformation(lib.merge('source'),
context={'source': source})
equal_subtree(transformation(destination), expected)
def test_xpath(xpath):
document = Document("<root><a/><b/></root>")
transformation = Transformation(Rule(xpath, lib.set_text("x")))
result = transformation(document).root
assert not result._data_node._exists
assert first(result.css_select("a")).full_text == "x"
assert first(result.css_select("b")).full_text == ""
def test_strip_namespace():
namespace = "http://www.example.org/ns/"
register_namespace("x", namespace)
root = new_tag_node("div", namespace=namespace)
transformation = Transformation(Rule(namespace, lib.remove_namespace))
result = transformation(root)
assert result.qualified_name == "div"
def test_OneOf():
document = Document('<root x="x"><a x="x"/><b x="x"/></root>')
transformation = Transformation(
Rule(OneOf("a", "b", {"x": "x"}), lib.set_text("x")))
result = transformation(document).root
assert result[0] == "x"
assert all(x.full_text == "" for x in result.child_nodes(is_tag_node))
def test_is_root_condition():
transformation = Transformation(
Rule(("a", "/"), lib.append("basket")),
result_object="context.basket",
context={"basket": []},
)
result = transformation(Document("<a><a/></a>"))
assert len(result) == 1
def test_attributes_re_key(constraint, expected):
document = Document('<root><item1 default-source="x"/>'
'<item2 default-value="y"/><item3/></root>')
transformation = Transformation(
Rule(constraint, (lib.get_localname, lib.append("result"))),
context={"result": []},
result_object="context.result",
)
assert transformation(document) == expected
def test_css_selector(selector, expected):
document = Document(
'<section xmlns="foo"><table><head>Table Header</head></table>'
'<cb type="start">X</cb><row>#</row></section>')
transformation = Transformation(
Rule(selector, (lib.get_text, lib.put_variable("result"))),
result_object="context.result",
)
assert transformation(document) == expected
def test_aborts():
def put_foo(context):
context.foo = "foo"
transformation = Transformation(
AbortTransformation, put_foo, result_object="context"
)
result = transformation(Document("<root/>"))
assert getattr(result, "foo", None) is None
def test_config_is_immutable():
trnsfmtn = Transformation(
lib.put_variable("test", "result"),
name="strip_surrounding_content",
result_object="context.test",
)
document = Document("<root/>")
assert trnsfmtn(document) == "result"
assert trnsfmtn(document) == "result"
def test_common_conditions():
document = Document(
'<root><a href="foo"/><a id="bar"/><a href="peng"/></root>')
transformation = Transformation(
Rule("*", (lib.get_attribute("href"), lib.append("references"))),
common_rule_conditions={"href": None},
context={"references": []},
result_object="context.references",
)
assert transformation(document) == ["foo", "peng"]
def linkfix_document(routes, filepath, currentpath, stats, fallback_url):
context = locals().copy()
context.pop('stats', None)
transformation = Transformation(
Rule([has_link, has_path_url], check_and_fix_link),
context=context,
)
with open(currentpath, mode='rb') as doc:
doc_tree = html5.parse(doc.read(), fallback_encoding='utf-8')
with requests.Session() as s:
return transformation(doc_tree, session=s, stats=stats)
def test_subtransformation():
subtransformation = Transformation(Rule("*", lib.set_localname("pablo")))
transformation = Transformation(
lib.f(id, Ref("root")),
lib.put_variable("source_id"),
subtransformation,
lib.f(id, Ref("root")),
lib.put_variable("result_id"),
lib.debug_symbols("source_id", "result_id"),
Rule(
Not(If(Ref("source_id"), operator.eq, Ref("result_id"))),
(
lib.debug_message("NO!"),
lib.debug_symbols("root"),
lib.set_localname("neruda"),
AbortRule,
),
),
)
doc = Document("<augustus />")
assert doc.root.local_name == "augustus"
result = transformation(doc)
assert result.root.local_name == "pablo"
def test_SkipToNextElement():
def more_complicated_test(node: TagNode):
# well, supposedly
if "x" not in node.attributes:
raise SkipToNextNode
if int(node.attributes["x"]) % 2:
raise SkipToNextNode
return node.local_name
transformation = Transformation(
Rule("*", (more_complicated_test, lib.append("evens"))),
context={"evens": []},
result_object="context.evens",
)
doc = Document('<root><a x="1"/><b x="2"/><c x="3"/><d x="4"/></root>')
assert transformation(doc) == ["b", "d"]
def test_If():
def return_zero():
return 0
def return_one():
return 1
transformation = Transformation(
Rule(If(0, operator.eq, 0), lib.put_variable("a")),
Rule(Not(If(return_zero, operator.eq, return_one)),
lib.put_variable("b")),
result_object="context",
)
result = transformation(Document("<root/>"))
assert hasattr(result, "a")
assert hasattr(result, "b")
def test_wikipedia_example_2():
expected = Document("""
<html xmlns="http://www.w3.org/1999/xhtml">
<head> <title>Testing XML Example</title> </head>
<body>
<h1>Persons</h1>
<ul>
<li>Ismincius, Morka</li>
<li>Smith, John</li>
</ul>
</body>
</html>
""")
def generate_skeleton(context):
context.html = new_tag_node(
"html",
namespace="http://www.w3.org/1999/xhtml",
children=(
tag("head", tag("title", "Testing XML Example")),
tag("body", (tag("h1", "Persons"), tag("ul"))),
),
)
def extract_person(node: TagNode, persons):
persons.append((
first(node.css_select("name")).full_text,
first(node.css_select("family-name")).full_text,
))
def list_persons(previous_result, html: TagNode):
first(html.css_select("html|body html|ul")).append_child(
*(html.new_tag_node("li", children=[f"{x[1]}, {x[0]}"])
for x in previous_result))
transformation = Transformation(
generate_skeleton,
Rule("person", extract_person),
lib.sort("persons", itemgetter(1)),
list_persons,
result_object="context.html",
context={"persons": []},
)
# that's four lines more LOC than the XSLT implementation
assert equal_subtree(transformation(wp_document.root), expected.root)
def transform_document(routes, root_dir, epub_zip, filepath, currentpath,
stats, fallback_url):
context = locals().copy()
context.pop('epub_zip', None)
transformation = Transformation(
add_re_namespace,
Rule("title", remove_from_tree),
Rule(has_link, route_url),
Rule(MatchesXPath(sutta_ref_xpath), link_sutta_references),
context=context,
)
with epub_zip.open(os.path.join(root_dir, filepath)) as doc_xml:
doc_tree = html5.parse(doc_xml.read(), fallback_encoding='utf-8')
with requests.Session() as s:
return transformation(doc_tree, session=s)
def test_grouped_steps():
def append_to_list(value):
def appender(list):
list.append(value)
return appender
stpgrp_c = (append_to_list(3), append_to_list(4))
stpgrp_b = (append_to_list(2), stpgrp_c, append_to_list(5))
stpgrp_a = append_to_list(1)
transformation = Transformation(
append_to_list(0),
stpgrp_a,
stpgrp_b,
context={"list": []},
result_object="context.list",
)
result = transformation(Document("<root/>"))
for exp, val in enumerate(result):
assert exp == val
def test_wikipedia_example_2():
expected = parse("""
<html xmlns="http://www.w3.org/1999/xhtml">
<head> <title>Testing XML Example</title> </head>
<body>
<h1>Persons</h1>
<ul>
<li>Ismincius, Morka</li>
<li>Smith, John</li>
</ul>
</body>
</html>
""")
def generate_skeleton(context, e):
context.html = e.html(e.head(e.title('Testing XML Example')),
e.body(e.h1('Persons'), e.ul()))
context.persons_list = context.html.xpath('./body/ul',
smart_prefix=True)[0]
def extract_person(element, persons):
persons.append(
(element.find('name').text, element.find('family-name').text))
def list_persons(previous_result, persons_list, e):
persons_list.extend(e.li(f'{x[1]}, {x[0]}') for x in previous_result)
transformation = Transformation(
lib.init_elementmaker(namespace='http://www.w3.org/1999/xhtml'),
generate_skeleton,
Rule('person', extract_person),
lib.sorter('persons', itemgetter(1)),
list_persons,
result_object='context.html',
context={'persons': []})
# that's eight (or not counting line-breaks: thirteen) lines less sloc
# than the XSLT implementation
assert equal_subtree(transformation(wp_document), expected)
def test_wikipedia_example_1():
expected = parse("""
<root>
<name username="******">John</name>
<name username="******">Morka</name>
</root>
""")
def extract_person(element):
return element.attrib['username'], element.find('name').text
def append_person(previous_result, target):
element = etree.SubElement(target, 'name',
{'username': previous_result[0]})
element.text = previous_result[1]
transformation = Transformation(Rule('person',
(extract_person, append_person)),
result_object='context.target',
context={'target': etree.Element('root')})
# that's five (or not counting line-breaks: eight) lines less sloc than the XSLT implementation
assert equal_subtree(transformation(wp_document), expected)
def test_any_element():
document = Document("<root><a/><b/></root>")
transformation = Transformation(Rule("*", lib.set_text("x")))
for node in transformation(document).root.child_nodes():
assert node[0] == "x"
Not,
Rule,
Transformation,
)
__all__ = []
# reduce_whitespaces
def _reduce_whitespace_handler(node: TagNode):
for child in node.child_nodes(is_text_node, recurse=True):
child.content = utils.reduce_whitespaces(node.content, strip="")
reduce_whitespaces = Transformation(Rule("/", _reduce_whitespace_handler))
"""
Normalizes any whitespace character in text nodes to a simple space and reduces
consecutive ones to one. Leading or tailing whitespaces are not stripped away.
"""
__all__.append("reduce_whitespaces")
# remove_empty_nodes
remove_empty_nodes = Transformation(
Rule(Not(lib.has_children, lib.has_text, lib.has_attributes, "/"),
lib.remove_node),
name="remove_empty_nodes",
traversal_order=(TRAVERSE_DEPTH_FIRST | TRAVERSE_LEFT_TO_RIGHT
| TRAVERSE_BOTTOM_TO_TOP),
)
mods_name = Transformation(
Once(
("roleTerm", {
"type": "text"
}, lib.text_equals("author")),
(
lib.make_node(local_name="author", namespace=TEI_NAMESPACE),
as_result,
lib.put_variable("role", "author"),
),
),
Once(
("namePart", {
"type": "family"
}),
(lib.get_text, lib.f(tag, "surname", prev), lib.append("result")),
),
Once(
("namePart", {
"type": "given"
}),
(lib.get_text, lib.f(tag, "forename", prev), lib.append("result")),
),
Once(
If(Ref("role"), op.eq, "author"),
(
lib.append(
"outer_context.titleStmt", Ref("result"), copy_node=True),
lib.append("outer_context.biblFull_titleStmt",
Ref("result"),
copy_node=True),
),
),
common_rule_conditions=MODS_NAMESPACE,
copy=False,
result_object=None,
)
def test_invalid_input_raises_type_error():
transformation = Transformation()
with raises(TypeError):
transformation({})
def test_set_text():
node = new_tag_node("pre")
transformation = Transformation(lib.put_variable("x", "Hello world."),
Rule("/", lib.set_text(Ref("x"))))
assert str(transformation(node)) == "<pre>Hello world.</pre>"
""" This module contains transformations that are supposedly of common interest. """
from inxs import \
(TRAVERSE_DEPTH_FIRST, TRAVERSE_BOTTOM_TO_TOP, TRAVERSE_LEFT_TO_RIGHT,
lib, Not, Rule, Transformation)
from inxs.lxml_utils import remove_element
__all__ = []
def _append_tail_to_previous_in_stream(element):
if not element.tail:
return
previous = element.getprevious()
if previous is None:
element.getparent().text += element.tail
elif previous.tail is None:
previous.tail = element.tail
else:
previous.tail += element.tail
remove_empty_elements = Transformation(
Rule(Not(lib.has_children, lib.has_text, lib.has_attributes, '/'),
(_append_tail_to_previous_in_stream, remove_element)),
name='remove_empty_elements',
traversal_order=TRAVERSE_DEPTH_FIRST | TRAVERSE_LEFT_TO_RIGHT
| TRAVERSE_BOTTOM_TO_TOP)
""" Removes elements without attributes, text, tail and children from the (sub-)tree. """
__all__.append('remove_empty_elements')