import os
import re
import shutil
import tempfile
import textwrap
import unittest
from package_marble_build_artifacts import ManualParser, main
EXPECT_CSP_META = ''
EXPECT_SCRIPT = (
''
)
EXPECT_CHROME_ASSETS = "chrome://browser/content/aboutmanual/assets"
class TestSingleHTML(unittest.TestCase):
top_dir: str | None = None
html_path: str | None = None
all_locales = ["en", "ar"]
# Set the TestCase.maxDiff to a larger number to see the full HTML.
maxDiff = 2000
@classmethod
def setUpClass(cls) -> None:
cls.top_dir = tempfile.mkdtemp()
try:
os.chdir(cls.top_dir)
html_dir = os.path.join(cls.top_dir, "html/html")
assets1_dir = os.path.join(cls.top_dir, "html/assets1")
assets2_sub_dir = os.path.join(cls.top_dir, "assets2/sub")
os.makedirs(html_dir)
os.makedirs(assets1_dir)
os.makedirs(assets2_sub_dir)
cls.html_path = os.path.join(html_dir, "test.html")
for filename in (
"html/html/neighbour1.svg",
"html/html/neighbour2.svg",
"html/html/neighbour1.css",
"html/html/neighbour2.css",
"html/assets1/image1.png",
"html/assets1/style1.css",
"assets2/image2.png",
"assets2/style2.css",
"assets2/sub/image2.png",
"assets2/sub/style2.css",
"html/assets1/image@2x.png",
"html/assets1/🦭.svg",
"html/assets1/~.svg",
):
# Create an empty file.
open(os.path.join(cls.top_dir, filename), "w", encoding="utf-8").close()
except:
shutil.rmtree(cls.top_dir)
raise
@classmethod
def tearDownClass(cls) -> None:
assert cls.top_dir is not None
shutil.rmtree(cls.top_dir)
@staticmethod
def dedent(in_str: str) -> str:
out_str = textwrap.dedent(in_str)
if out_str.startswith("\n"):
out_str = out_str[1:]
return out_str
def assert_html_out(
self, in_html: str, locale: str, expect_html: str, expect_assets: dict[str, str]
) -> None:
assert self.html_path is not None
assert self.top_dir is not None
with open(self.html_path, "w", encoding="utf-8") as file:
file.write(self.dedent(in_html))
expect_html = self.dedent(expect_html)
expect_assets = {
key: os.path.join(self.top_dir, val) for key, val in expect_assets.items()
}
parser = ManualParser(
locale=locale,
html_path=self.html_path,
top_dir=self.top_dir,
all_locales=self.all_locales,
)
out_html, out_assets = parser.process()
self.assertEqual(out_html, expect_html, "HTML should match")
self.assertEqual(out_assets, expect_assets)
def assert_body_out(
self,
in_body: str,
expect_body: str,
expect_assets: dict[str, str],
locale: str = "ar",
) -> None:
indent = " "
in_body = textwrap.indent(self.dedent(in_body), prefix=indent)
expect_body = textwrap.indent(self.dedent(expect_body), prefix=indent)
in_html = f"""
Test
\n{in_body}{indent}
"""
expect_html = f"""
{EXPECT_CSP_META}
Test
{EXPECT_SCRIPT}
\n{expect_body}{indent}
"""
self.assert_html_out(in_html, locale, expect_html, expect_assets)
def assert_html_raises(
self, in_html: str, locale: str, expect_err_str: str
) -> None:
assert self.html_path is not None
assert self.top_dir is not None
with open(self.html_path, "w", encoding="utf-8") as file:
file.write(self.dedent(in_html))
parser = ManualParser(
locale=locale,
html_path=self.html_path,
top_dir=self.top_dir,
all_locales=self.all_locales,
)
with self.assertRaisesRegex(
ValueError,
r"^[^:]+.html: \([0-9]+, [0-9]+\): " + re.escape(expect_err_str) + "$",
):
parser.process()
def assert_body_raises(self, in_body: str, expect_err_str: str) -> None:
indent = " "
in_body = textwrap.indent(self.dedent(in_body), prefix=indent)
in_html = f"""
Test
\n{in_body}{indent}
"""
self.assert_html_raises(in_html, "ar", expect_err_str)
def test_empty_body(self) -> None:
self.assert_body_out("\n", "\n", {})
def test_html_tag(self) -> None:
self.assert_html_out(
"""
Test
""",
"en",
f"""
{EXPECT_CSP_META}
Test
{EXPECT_SCRIPT}
""",
{},
)
# Invalid.
self.assert_html_raises(
"""
""",
"ar",
"html has a parent",
)
self.assert_html_raises(
"""
""",
"ar",
"More than one html tag",
)
self.assert_html_raises(
"""
""",
"ar",
"html is missing a lang attribute",
)
self.assert_html_raises(
"""
""",
"ar",
"html is missing a dir attribute",
)
self.assert_html_raises(
"""
""",
"ar",
"Unexpected lang attribute: en",
)
self.assert_html_raises(
"""
""",
"en",
"Unexpected dir attribute: LTR",
)
self.assert_html_raises(
"""
""",
"en",
"Unexpected dir attribute: auto",
)
def test_head_ignored(self) -> None:
# Most of the tags and attributes in the head are ignored.
self.assert_html_out(
"""
دليل استخدام متصفح تور
oops
oops
""",
"ar",
f"""
{EXPECT_CSP_META}
دليل استخدام متصفح تور
{EXPECT_SCRIPT}
""",
{"html__assets1__style1.css": "html/assets1/style1.css"},
)
def test_title(self) -> None:
# NOTE: Expected to fail in python 3.13.5, 3.12.11, 3.11.12, 3.10.18,
# 3.9.22 and earlier, due to a bug in html.parser, which fails to treat
# the "
" as text content. This should not cause any issues in
# practice, since the title is not expected include a raw <.
# It is tested here for robustness.
self.assert_html_out(
"""
title content <div> 'more"
""",
"ar",
f"""
{EXPECT_CSP_META}
title content <div> 'more"
{EXPECT_SCRIPT}
""",
{},
)
# elements outside the <head> are ignored.
self.assert_html_out(
"""
<html lang="ar" dir="rtl">
<head>
<title>TestSome other titleBody title