bpo-28238: Implement "{*}tag" and "{ns}*" wildcard tag selection support for ElementPath, and extend the surrounding tests and docs. (GH-12997)
diff --git a/Lib/test/test_xml_etree.py b/Lib/test/test_xml_etree.py
index a59a11f..ca6862c 100644
--- a/Lib/test/test_xml_etree.py
+++ b/Lib/test/test_xml_etree.py
@@ -1137,16 +1137,21 @@
def test_xpath_tokenizer(self):
# Test the XPath tokenizer.
from xml.etree import ElementPath
- def check(p, expected):
+ def check(p, expected, namespaces=None):
self.assertEqual([op or tag
- for op, tag in ElementPath.xpath_tokenizer(p)],
+ for op, tag in ElementPath.xpath_tokenizer(p, namespaces)],
expected)
# tests from the xml specification
check("*", ['*'])
+ check("{ns}*", ['{ns}*'])
+ check("{}*", ['{}*'])
+ check("{*}tag", ['{*}tag'])
+ check("{*}*", ['{*}*'])
check("text()", ['text', '()'])
check("@name", ['@', 'name'])
check("@*", ['@', '*'])
+ check("@{ns}attr", ['@', '{ns}attr'])
check("para[1]", ['para', '[', '1', ']'])
check("para[last()]", ['para', '[', 'last', '()', ']'])
check("*/para", ['*', '/', 'para'])
@@ -1158,6 +1163,7 @@
check("//olist/item", ['//', 'olist', '/', 'item'])
check(".", ['.'])
check(".//para", ['.', '//', 'para'])
+ check(".//{*}tag", ['.', '//', '{*}tag'])
check("..", ['..'])
check("../@lang", ['..', '/', '@', 'lang'])
check("chapter[title]", ['chapter', '[', 'title', ']'])
@@ -1168,6 +1174,8 @@
check("{http://spam}egg", ['{http://spam}egg'])
check("./spam.egg", ['.', '/', 'spam.egg'])
check(".//{http://spam}egg", ['.', '//', '{http://spam}egg'])
+ check("./xsd:type", ['.', '/', '{http://www.w3.org/2001/XMLSchema}type'],
+ {'xsd': 'http://www.w3.org/2001/XMLSchema'})
def test_processinginstruction(self):
# Test ProcessingInstruction directly
@@ -2669,6 +2677,50 @@
self.assertEqual(len(root.findall(".//xx:b", namespaces=nsmap)), 2)
self.assertEqual(len(root.findall(".//b", namespaces=nsmap)), 1)
+ def test_findall_wildcard(self):
+ root = ET.XML('''
+ <a xmlns:x="X" xmlns:y="Y">
+ <x:b><c/></x:b>
+ <b/>
+ <c><x:b/><b/></c><y:b/>
+ </a>''')
+ root.append(ET.Comment('test'))
+
+ self.assertEqual(summarize_list(root.findall("{*}b")),
+ ['{X}b', 'b', '{Y}b'])
+ self.assertEqual(summarize_list(root.findall("{*}c")),
+ ['c'])
+ self.assertEqual(summarize_list(root.findall("{X}*")),
+ ['{X}b'])
+ self.assertEqual(summarize_list(root.findall("{Y}*")),
+ ['{Y}b'])
+ self.assertEqual(summarize_list(root.findall("{}*")),
+ ['b', 'c'])
+ self.assertEqual(summarize_list(root.findall("{}b")), # only for consistency
+ ['b'])
+ self.assertEqual(summarize_list(root.findall("{}b")),
+ summarize_list(root.findall("b")))
+ self.assertEqual(summarize_list(root.findall("{*}*")),
+ ['{X}b', 'b', 'c', '{Y}b'])
+ # This is an unfortunate difference, but that's how find('*') works.
+ self.assertEqual(summarize_list(root.findall("{*}*") + [root[-1]]),
+ summarize_list(root.findall("*")))
+
+ self.assertEqual(summarize_list(root.findall(".//{*}b")),
+ ['{X}b', 'b', '{X}b', 'b', '{Y}b'])
+ self.assertEqual(summarize_list(root.findall(".//{*}c")),
+ ['c', 'c'])
+ self.assertEqual(summarize_list(root.findall(".//{X}*")),
+ ['{X}b', '{X}b'])
+ self.assertEqual(summarize_list(root.findall(".//{Y}*")),
+ ['{Y}b'])
+ self.assertEqual(summarize_list(root.findall(".//{}*")),
+ ['c', 'b', 'c', 'b'])
+ self.assertEqual(summarize_list(root.findall(".//{}b")), # only for consistency
+ ['b', 'b'])
+ self.assertEqual(summarize_list(root.findall(".//{}b")),
+ summarize_list(root.findall(".//b")))
+
def test_bad_find(self):
e = ET.XML(SAMPLE_XML)
with self.assertRaisesRegex(SyntaxError, 'cannot use absolute path'):
diff --git a/Lib/xml/etree/ElementPath.py b/Lib/xml/etree/ElementPath.py
index b670d58..cfe72f2 100644
--- a/Lib/xml/etree/ElementPath.py
+++ b/Lib/xml/etree/ElementPath.py
@@ -99,13 +99,70 @@
parent_map[e] = p
return parent_map
+
+
+def _is_wildcard_tag(tag):
+ return tag[:3] == '{*}' or tag[-2:] == '}*'
+
+
+def _prepare_tag(tag):
+ _isinstance, _str = isinstance, str
+ if tag == '{*}*':
+ # Same as '*', but no comments or processing instructions.
+ # It can be a surprise that '*' includes those, but there is no
+ # justification for '{*}*' doing the same.
+ def select(context, result):
+ for elem in result:
+ if _isinstance(elem.tag, _str):
+ yield elem
+ elif tag == '{}*':
+ # Any tag that is not in a namespace.
+ def select(context, result):
+ for elem in result:
+ el_tag = elem.tag
+ if _isinstance(el_tag, _str) and el_tag[0] != '{':
+ yield elem
+ elif tag[:3] == '{*}':
+ # The tag in any (or no) namespace.
+ suffix = tag[2:] # '}name'
+ no_ns = slice(-len(suffix), None)
+ tag = tag[3:]
+ def select(context, result):
+ for elem in result:
+ el_tag = elem.tag
+ if el_tag == tag or _isinstance(el_tag, _str) and el_tag[no_ns] == suffix:
+ yield elem
+ elif tag[-2:] == '}*':
+ # Any tag in the given namespace.
+ ns = tag[:-1]
+ ns_only = slice(None, len(ns))
+ def select(context, result):
+ for elem in result:
+ el_tag = elem.tag
+ if _isinstance(el_tag, _str) and el_tag[ns_only] == ns:
+ yield elem
+ else:
+ raise RuntimeError(f"internal parser error, got {tag}")
+ return select
+
+
def prepare_child(next, token):
tag = token[1]
- def select(context, result):
- for elem in result:
- for e in elem:
- if e.tag == tag:
- yield e
+ if _is_wildcard_tag(tag):
+ select_tag = _prepare_tag(tag)
+ def select(context, result):
+ def select_child(result):
+ for elem in result:
+ yield from elem
+ return select_tag(context, select_child(result))
+ else:
+ if tag[:2] == '{}':
+ tag = tag[2:] # '{}tag' == 'tag'
+ def select(context, result):
+ for elem in result:
+ for e in elem:
+ if e.tag == tag:
+ yield e
return select
def prepare_star(next, token):
@@ -130,11 +187,24 @@
tag = token[1]
else:
raise SyntaxError("invalid descendant")
- def select(context, result):
- for elem in result:
- for e in elem.iter(tag):
- if e is not elem:
- yield e
+
+ if _is_wildcard_tag(tag):
+ select_tag = _prepare_tag(tag)
+ def select(context, result):
+ def select_child(result):
+ for elem in result:
+ for e in elem.iter():
+ if e is not elem:
+ yield e
+ return select_tag(context, select_child(result))
+ else:
+ if tag[:2] == '{}':
+ tag = tag[2:] # '{}tag' == 'tag'
+ def select(context, result):
+ for elem in result:
+ for e in elem.iter(tag):
+ if e is not elem:
+ yield e
return select
def prepare_parent(next, token):