def test_matches_function_performs_regex_matching_as_per_xpath_30_functions_spec():
html_body = """
<p>moe</p>
<p>larry</p>
<p>curly</p>"""
assert query_html_doc(html_body, '//p[matches(text(), "^l[ary]+")]/text()') == expected_result('larry')
assert query_html_doc(html_body, '//p[matches(text(), ".URL.", "i")]/text()') == expected_result('curly')
def test_tokenize_function_breaks_up_strings_as_per_xpath_30_functions_spec():
assert query_html_doc('', 'tokenize("Moe:Larry:..Curly", ":\.*")') == expected_result("""
Moe
Larry
Curly""")
assert query_html_doc('', 'tokenize("HaxtaXpatience", "x", "i")') == expected_result("""
Ha
ta
patience""")
assert query_html_doc('', 'count(tokenize("haxtaxstax", "x"))') == '4'
def test_escapes_work_in_string_literals():
assert query_html_doc('', '"foo bar"') == expected_result("""
foo
bar""")
assert query_html_doc('', "'foo bar'") == expected_result("""
foo
bar""")
assert query_html_doc('', '`foo bar`') == expected_result("""
foo
bar""")
def test_element_constructor_accepts_numbers_and_booleans():
assert query_html_doc('', 'element test { 98.6 }') == expected_result("""
<test>
98.6
</test>""")
assert query_html_doc('', 'element test { false() }') == expected_result("""
<test>
false
</test>""")
def test_if_then_else_works_with_node_sets():
html_body = """
<p>eekaboo</p>"""
assert query_html_doc(html_body, 'if (//p) then //p else 1 to 3') == expected_result("""
<p>
eekaboo
</p>""")
assert query_html_doc(html_body, 'if (//div) then //p else 1 to 3') == expected_result("""
1
2
3""")
def test_tokenize_function_breaks_up_strings_as_per_xpath_30_functions_spec():
assert query_html_doc(
'', 'tokenize("Moe:Larry:..Curly", ":\.*")') == expected_result("""
Moe
Larry
Curly""")
assert query_html_doc(
'', 'tokenize("HaxtaXpatience", "x", "i")') == expected_result("""
Ha
ta
patience""")
assert query_html_doc('', 'count(tokenize("haxtaxstax", "x"))') == '4'
def test_if_then_else_works_with_node_sets():
html_body = """
<p>eekaboo</p>"""
assert query_html_doc(
html_body, 'if (//p) then //p else 1 to 3') == expected_result("""
<p>
eekaboo
</p>""")
assert query_html_doc(
html_body, 'if (//div) then //p else 1 to 3') == expected_result("""
1
2
3""")
def test_matches_function_performs_regex_matching_as_per_xpath_30_functions_spec(
):
html_body = """
<p>moe</p>
<p>larry</p>
<p>curly</p>"""
assert query_html_doc(
html_body,
'//p[matches(text(), "^l[ary]+")]/text()') == expected_result('larry')
assert query_html_doc(
html_body,
'//p[matches(text(), ".URL.", "i")]/text()') == expected_result(
'curly')
def test_element_constructor_accepts_attributes_from_original_document_including_multi_values_like_classes():
html_body = """
<p class="one two" three="four">
contents
</p>"""
assert query_html_doc(html_body, 'element test { //p/@* }') == expected_result("""
<test class="one two" three="four">
</test>""")
assert query_html_doc(html_body, 'element test { //p/@three, //p }') == expected_result("""
<test three="four">
<p class="one two" three="four">
contents
</p>
</test>""")
def test_selects_the_chapter_children_of_the_context_node_that_have_one_or_more_title_children_with_string_value_equal_to_Introduction(
):
html = """
<context>
<chapter>
<title>Introduction</title>
</chapter>
<chapter>not selected</chapter>
<chapter>
<title>Author's Note</title>
</chapter>
<chapter>
<title>Introduction</title>
<content>Hello, I'm chapter.</content>
</chapter>
</context>"""
assert query_context_node(
html, 'chapter[title="Introduction"]') == expected_result("""
<chapter>
<title>
Introduction
</title>
</chapter>
<chapter>
<title>
Introduction
</title>
<content>
Hello, I'm chapter.
</content>
</chapter>""")
def test_selects_the_chapter_children_of_the_context_node_that_have_one_or_more_title_children(
):
html = """
<context>
<chapter>
<title>selected</title>
</chapter>
<chapter>
<not-title></not-title>
</chapter>
<chapter>
<title>also selected</title>
</chapter>
</context>"""
assert query_context_node(html, 'chapter[title]') == expected_result("""
<chapter>
<title>
selected
</title>
</chapter>
<chapter>
<title>
also selected
</title>
</chapter>""")
def test_selects_the_lang_attribute_of_the_parent_of_the_context_node():
html = """
<root lang="English">
<context></context>
</root>"""
soup = make_soup(html)
assert query_context_node(soup.root.context, '../@lang') == expected_result('lang="English"')
def test_selects_the_para_element_descendants_of_the_context_node():
html = """
<para>
<context>
<para>selected</para>
<not-para>not selected</not-para>
<para>
<para>also selected</para>
</para>
</context>
</para>"""
soup = make_soup(html)
assert query_context_node(soup.para.context,
'.//para') == expected_result("""
<para>
selected
</para>
<para>
<para>
also selected
</para>
</para>
<para>
also selected
</para>""")
def test_selects_all_the_item_elements_in_the_same_document_as_the_context_node_that_have_an_olist_parent(
):
html = """
<root>
<olist>no items</olist>
<item>not selected</item>
<context></context>
<olist>
<item>first</item>
</olist>
<item>
<olist>
<item>second</item>
<olist>
</item>
</root>"""
soup = make_soup(html)
assert query_context_node(soup.root.context,
'//olist/item') == expected_result("""
<item>
first
</item>
<item>
second
</item>""")
def test_comma_as_sequence_cat_operator_does_not_bind_at_end_of_return_clause():
assert query_html_doc('', 'for $x in (1 to 2) return $x, "!"') == expected_result("""
1
2
!""")
assert query_html_doc('', 'sum(for $x in //span return $x, "zero")') == 'zero'
assert query_html_doc('', 'sum(//span -> $_, "zero")') == 'zero'
def test_selects_the_chapter_children_of_the_context_node_that_have_one_or_more_title_children_with_string_value_equal_to_Introduction():
html = """
<context>
<chapter>
<title>Introduction</title>
</chapter>
<chapter>not selected</chapter>
<chapter>
<title>Author's Note</title>
</chapter>
<chapter>
<title>Introduction</title>
<content>Hello, I'm chapter.</content>
</chapter>
</context>"""
assert query_context_node(html, 'chapter[title="Introduction"]') == expected_result("""
<chapter>
<title>
Introduction
</title>
</chapter>
<chapter>
<title>
Introduction
</title>
<content>
Hello, I'm chapter.
</content>
</chapter>""")
def test_join_filter_joins_string_values_from_node_set():
html_body = """
<p>one</p>
<p>two</p>
<p>three</p>"""
assert query_html_doc(html_body,
'`${j:,://p}`') == expected_result('one,two,three')
def test_selects_the_fifth_para_child_of_the_context_node_if_that_child_has_a_type_attribute_with_value_warning(
):
html = """
<context>
<para>not selected</para>
<para>not selected</para>
<para>not selected</para>
<para>not selected</para>
<para type="error">selected</para>
</context>"""
assert query_context_node(
html, 'para[5][@type="warning"]') == expected_result("")
assert query_context_node(html.replace(
'error', 'warning'), 'para[5][@type="warning"]') == expected_result("""
<para type="warning">
selected
</para>""")
def test_selects_the_lang_attribute_of_the_parent_of_the_context_node():
html = """
<root lang="English">
<context></context>
</root>"""
soup = make_soup(html)
assert query_context_node(soup.root.context,
'../@lang') == expected_result('lang="English"')
def test_even_and_odd_functions_select_the_appropriate_elements_based_on_position():
html_body = """
<p>You</p>
<p>I</p>
<p>are</p>
<p>am</p>
<p>odd.</p>
<p>even.</p>"""
assert query_html_doc(html_body, '//p[even()]/text()') == expected_result("""
I
am
even.""")
assert query_html_doc(html_body, '//p[odd()]/text()') == expected_result("""
You
are
odd.""")
def test_element_node_becomes_normalized_text_contents_in_interpolated_string(
):
html_body = """
<p>
foo bar
</p>"""
assert query_html_doc(html_body,
'`-->${//p}<--`') == expected_result('-->foo bar<--')
def test_matches_function_extends_to_using_context_node_when_passed_no_input_string(
):
html_body = """
<p>bar</p>
<p>foo</p>"""
assert query_html_doc(
html_body, '//p[matches("^f.+")]/text()') == expected_result('foo')
def test_selects_the_context_node():
html = """
<context>
selected
</context>"""
assert query_context_node(html, '.') == expected_result("""
<context>
selected
</context>""")
def test_tolerates_latin_characters_in_comments(capsys, mocker):
mocker.patch('hq.hq.docopt').return_value = simulate_args_dict(expression='//comment()')
mocker.patch('sys.stdin.read').return_value = wrap_html_body(u"""
<!-- sacr\xe9 bleu! -->""")
main()
actual, _ = capture_console_output(capsys)
assert actual == expected_result(u'<!-- sacr\xe9 bleu! -->')
def test_iteration_using_for():
html_body = """
<p>one</p>
<p>two</p>
<p>three</p>"""
assert query_html_doc(html_body, 'for $x in //p return $x/text()') == expected_result("""
one
two
three""")
def test_union_decomposition_with_parentheses():
html_body = """
<h1>heading</h1>
<p>content</p>
<h1>another heading</h1>"""
assert query_html_doc(html_body, '(//h1 | //p) => ("fizz" | "buzz")') == expected_result("""
fizz
buzz
fizz""")
def test_selects_the_context_node():
html = """
<context>
selected
</context>"""
assert query_context_node(html, '.') == expected_result("""
<context>
selected
</context>""")
def test_union_decomposition_naked():
html_body = """
<h1>heading</h1>
<p>content</p>
<h1>another heading</h1>"""
assert query_html_doc(html_body, '(//h1 | //p) => `h1 $_` | `p $_`') == expected_result("""
h1 heading
p content
h1 another heading""")
def test_rooted_location_paths_work_with_both_kinds_of_slash():
html_body = """
<section>
<div>
<div>foo</div>
</div>
</section>
<section>
<div>
<div>bar</div>
</div>
</section>"""
assert query_html_doc(
html_body, 'for $x in //section return $x/div') == expected_result("""
<div>
<div>
foo
</div>
</div>
<div>
<div>
bar
</div>
</div>""")
assert query_html_doc(
html_body, 'for $x in //section return $x//div') == expected_result("""
<div>
<div>
foo
</div>
</div>
<div>
foo
</div>
<div>
<div>
bar
</div>
</div>
<div>
bar
</div>""")
def test_iteration_using_for():
html_body = """
<p>one</p>
<p>two</p>
<p>three</p>"""
assert query_html_doc(
html_body, 'for $x in //p return $x/text()') == expected_result("""
one
two
three""")
def test_abbreviated_flowr_provides_expected_iteration_variable_in_value_clause():
html_body = """
<p>one</p>
<p>two</p>
<p>three</p>"""
assert query_html_doc(html_body, '//p -> $_/text()') == expected_result("""
one
two
three""")
def test_union_decomposition_naked():
html_body = """
<h1>heading</h1>
<p>content</p>
<h1>another heading</h1>"""
assert query_html_doc(
html_body, '(//h1 | //p) => `h1 $_` | `p $_`') == expected_result("""
h1 heading
p content
h1 another heading""")
def test_selects_the_first_para_child_of_the_context_node():
html = """
<context>
<para>selected</para>
<para>not selected</para>
</context>"""
assert query_context_node(html, 'para[1]') == expected_result("""
<para>
selected
</para>""")
def test_comma_as_sequence_cat_operator_does_not_bind_at_end_of_return_clause(
):
assert query_html_doc(
'', 'for $x in (1 to 2) return $x, "!"') == expected_result("""
1
2
!""")
assert query_html_doc('',
'sum(for $x in //span return $x, "zero")') == 'zero'
assert query_html_doc('', 'sum(//span -> $_, "zero")') == 'zero'
def test_union_decomposition_with_parentheses():
html_body = """
<h1>heading</h1>
<p>content</p>
<h1>another heading</h1>"""
assert query_html_doc(
html_body, '(//h1 | //p) => ("fizz" | "buzz")') == expected_result("""
fizz
buzz
fizz""")
def test_tolerates_latin_characters_in_attribute_contents(capsys, mocker):
mocker.patch('hq.hq.docopt').return_value = simulate_args_dict(expression='//div/@role')
mocker.patch('sys.stdin.read').return_value = wrap_html_body(u"""
<div role="prim\xe4r">
</div>""")
main()
actual, _ = capture_console_output(capsys)
assert actual == expected_result(u'role="prim\xe4r"')
def test_tolerates_latin_characters_in_comments(capsys, mocker):
mocker.patch('hq.hq.docopt').return_value = simulate_args_dict(
expression='//comment()')
mocker.patch('sys.stdin.read').return_value = wrap_html_body(u"""
<!-- sacr\xe9 bleu! -->""")
main()
actual, _ = capture_console_output(capsys)
assert actual == expected_result(u'<!-- sacr\xe9 bleu! -->')
def test_selects_the_first_para_child_of_the_context_node():
html = """
<context>
<para>selected</para>
<para>not selected</para>
</context>"""
assert query_context_node(html, 'para[1]') == expected_result("""
<para>
selected
</para>""")
def test_rooted_location_paths_work_with_both_kinds_of_slash():
html_body = """
<section>
<div>
<div>foo</div>
</div>
</section>
<section>
<div>
<div>bar</div>
</div>
</section>"""
assert query_html_doc(html_body, 'for $x in //section return $x/div') == expected_result("""
<div>
<div>
foo
</div>
</div>
<div>
<div>
bar
</div>
</div>""")
assert query_html_doc(html_body, 'for $x in //section return $x//div') == expected_result("""
<div>
<div>
foo
</div>
</div>
<div>
foo
</div>
<div>
<div>
bar
</div>
</div>
<div>
bar
</div>""")
def test_tolerates_latin_characters_in_attribute_contents(capsys, mocker):
mocker.patch('hq.hq.docopt').return_value = simulate_args_dict(
expression='//div/@role')
mocker.patch('sys.stdin.read').return_value = wrap_html_body(u"""
<div role="prim\xe4r">
</div>""")
main()
actual, _ = capture_console_output(capsys)
assert actual == expected_result(u'role="prim\xe4r"')
def test_abbreviated_flowr_provides_expected_iteration_variable_in_value_clause(
):
html_body = """
<p>one</p>
<p>two</p>
<p>three</p>"""
assert query_html_doc(html_body, '//p -> $_/text()') == expected_result("""
one
two
three""")
def test_even_and_odd_functions_select_the_appropriate_elements_based_on_position(
):
html_body = """
<p>You</p>
<p>I</p>
<p>are</p>
<p>am</p>
<p>odd.</p>
<p>even.</p>"""
assert query_html_doc(html_body,
'//p[even()]/text()') == expected_result("""
I
am
even.""")
assert query_html_doc(html_body,
'//p[odd()]/text()') == expected_result("""
You
are
odd.""")
def test_selects_the_parent_of_the_context_node():
html = """
<root>
<context></context>
</root>"""
soup = make_soup(html)
assert query_context_node(html, '..') == expected_result("""
<root>
<context>
</context>
</root>""")
def test_selects_the_parent_of_the_context_node():
html = """
<root>
<context></context>
</root>"""
soup = make_soup(html)
assert query_context_node(html, '..') == expected_result("""
<root>
<context>
</context>
</root>""")
def test_union_decomposition_applies_first_matching_clause():
html_body = """
<div>div1</div>
<p>p1</p>
<div>
<p>p2</p>
</div>"""
query = '(//p | /html/body/div | /html/body//*) => "one" | "two" | "three"'
assert query_html_doc(html_body, query) == expected_result("""
two
one
two
one""")
def test_union_decomposition_applies_first_matching_clause():
html_body = """
<div>div1</div>
<p>p1</p>
<div>
<p>p2</p>
</div>"""
query = '(//p | /html/body/div | /html/body//*) => "one" | "two" | "three"'
assert query_html_doc(html_body, query) == expected_result("""
two
one
two
one""")
def test_selects_the_para_element_children_of_the_context_node():
html = """
<context>
<para>selected</para>
<not-para>not selected</not-para>
<para>also selected</para>
</context>"""
assert query_context_node(html, 'para') == expected_result("""
<para>
selected
</para>
<para>
also selected
</para>""")
def test_selects_all_the_employee_children_of_the_context_node_that_have_both_a_secretary_attribute_and_an_assistant_attribute():
html = """
<context>
<employee secretary="not selected"></employee>
<employee assistant="" secretary="">selected</employee>
<employee assistant="not selected"></employee>
<employee secretary="also" assistant="selected"></employee>
</context>"""
assert query_context_node(html, 'employee[@secretary and @assistant]') == expected_result("""
<employee assistant="" secretary="">
selected
</employee>
<employee assistant="selected" secretary="also">
</employee>""")
def test_tolerates_latin_characters_in_element_contents(capsys, mocker):
mocker.patch('hq.hq.docopt').return_value = simulate_args_dict(expression='//div')
mocker.patch('sys.stdin.read').return_value = wrap_html_body(u"""
<div>
T\xeate\xa0\xe0\xa0t\xeate
</div>""")
main()
actual, _ = capture_console_output(capsys)
assert actual == expected_result(u"""
<div>
T\xeate\xa0\xe0\xa0t\xeate
</div>""")
def test_selects_the_fifth_para_child_of_the_context_node_if_that_child_has_a_type_attribute_with_value_warning():
html = """
<context>
<para>not selected</para>
<para>not selected</para>
<para>not selected</para>
<para>not selected</para>
<para type="error">selected</para>
</context>"""
assert query_context_node(html, 'para[5][@type="warning"]') == expected_result("")
assert query_context_node(html.replace('error', 'warning'), 'para[5][@type="warning"]') == expected_result("""
<para type="warning">
selected
</para>""")
def test_selects_the_para_element_children_of_the_context_node():
html = """
<context>
<para>selected</para>
<not-para>not selected</not-para>
<para>also selected</para>
</context>"""
assert query_context_node(html, 'para') == expected_result("""
<para>
selected
</para>
<para>
also selected
</para>""")
def test_union_operator_produces_node_set_sorted_in_document_order():
html_body = """
<div>one</div>
<p>two</p>
<div>three</div>"""
assert query_html_doc(html_body, '//p | //div') == expected_result("""
<div>
one
</div>
<p>
two
</p>
<div>
three
</div>""")
def test_selects_all_para_children_of_the_context_node_that_have_a_type_attribute_with_value_warning():
html = """
<context>
<para>not selected</para>
<para type="warning">selected</para>
<para type="error">not selected</para>
<para type="warning">also selected</para>
</context>"""
assert query_context_node(html, 'para[@type="warning"]') == expected_result("""
<para type="warning">
selected
</para>
<para type="warning">
also selected
</para>""")
def test_selects_all_element_children_of_the_context_node():
html = """
<context>
<!-- comment -->
<element>selected</element>
non-selected text
<para>also selected</para>
</context>"""
assert query_context_node(html, '*') == expected_result("""
<element>
selected
</element>
<para>
also selected
</para>""")
def test_selects_all_para_grandchildren_of_the_context_node():
html = """
<context>
<para>
not selected
<para>selected</para>
<para>also selected</para>
</para>
</context>"""
assert query_context_node(html, '*/para') == expected_result("""
<para>
selected
</para>
<para>
also selected
</para>""")
def test_variables_before_for_have_global_scope_and_within_for_have_iteration_scope():
query = """
let $x := 2
let $z := $x
for $_ in (1, $x)
let $y := $_
let $x := $_
return ($x, $z, $x = $y)"""
assert query_html_doc('', ' '.join(query.split('\n'))) == expected_result("""
1
2
true
2
2
true""")
def test_selects_the_second_section_of_the_fifth_chapter_of_the_doc():
html = """
<doc>
<chapter>one</chapter>
<chapter>two</chapter>
<chapter>three</chapter>
<chapter>four</chapter>
<chapter>
<section>five point one</section>
<section>five point two</section>
</chapter>
</doc>"""
assert query_context_node(html, '/doc/chapter[5]/section[2]') == expected_result("""
<section>
five point two
</section>""")
