https://github.com/python/cpython/commit/7ebbd271444d89218870169624921b795a717470
commit: 7ebbd271444d89218870169624921b795a717470
branch: main
author: Serhiy Storchaka <storch...@gmail.com>
committer: serhiy-storchaka <storch...@gmail.com>
date: 2025-04-09T11:08:04+03:00
summary:
gh-130631: Make join_header_words() more similar to the original Perl version
(GH-130632)
* Always quote strings with non-ASCII characters.
* Allow some non-separator and non-control characters (like "." or "-")
be unquoted.
* Always quote strings that end with "\n".
* Use the fullmatch() method for clarity and optimization.
files:
A Misc/NEWS.d/next/Library/2025-02-27-14-25-01.gh-issue-130631.dmZcZM.rst
M Lib/http/cookiejar.py
M Lib/test/test_http_cookiejar.py
diff --git a/Lib/http/cookiejar.py b/Lib/http/cookiejar.py
index fb0fd2e97999af..68cf16c93cc1c8 100644
--- a/Lib/http/cookiejar.py
+++ b/Lib/http/cookiejar.py
@@ -430,6 +430,7 @@ def split_header_words(header_values):
if pairs: result.append(pairs)
return result
+HEADER_JOIN_TOKEN_RE = re.compile(r"[!#$%&'*+\-.^_`|~0-9A-Za-z]+")
HEADER_JOIN_ESCAPE_RE = re.compile(r"([\"\\])")
def join_header_words(lists):
"""Do the inverse (almost) of the conversion done by split_header_words.
@@ -437,10 +438,10 @@ def join_header_words(lists):
Takes a list of lists of (key, value) pairs and produces a single header
value. Attribute values are quoted if needed.
- >>> join_header_words([[("text/plain", None), ("charset", "iso-8859-1")]])
- 'text/plain; charset="iso-8859-1"'
- >>> join_header_words([[("text/plain", None)], [("charset",
"iso-8859-1")]])
- 'text/plain, charset="iso-8859-1"'
+ >>> join_header_words([[("text/plain", None), ("charset", "iso-8859/1")]])
+ 'text/plain; charset="iso-8859/1"'
+ >>> join_header_words([[("text/plain", None)], [("charset",
"iso-8859/1")]])
+ 'text/plain, charset="iso-8859/1"'
"""
headers = []
@@ -448,7 +449,7 @@ def join_header_words(lists):
attr = []
for k, v in pairs:
if v is not None:
- if not re.search(r"^\w+$", v):
+ if not HEADER_JOIN_TOKEN_RE.fullmatch(v):
v = HEADER_JOIN_ESCAPE_RE.sub(r"\\\1", v) # escape " and \
v = '"%s"' % v
k = "%s=%s" % (k, v)
diff --git a/Lib/test/test_http_cookiejar.py b/Lib/test/test_http_cookiejar.py
index 25a671809d4499..cf02c5b43a2e43 100644
--- a/Lib/test/test_http_cookiejar.py
+++ b/Lib/test/test_http_cookiejar.py
@@ -285,11 +285,21 @@ def test_roundtrip(self):
("foo=bar;bar=baz", "foo=bar; bar=baz"),
('foo bar baz', "foo; bar; baz"),
(r'foo="\"" bar="\\"', r'foo="\""; bar="\\"'),
+ ("föo=bär", 'föo="bär"'),
('foo,,,bar', 'foo, bar'),
('foo=bar,bar=baz', 'foo=bar, bar=baz'),
+ ("foo=\n", 'foo=""'),
+ ('foo="\n"', 'foo="\n"'),
+ ('foo=bar\n', 'foo=bar'),
+ ('foo="bar\n"', 'foo="bar\n"'),
+ ('foo=bar\nbaz', 'foo=bar; baz'),
+ ('foo="bar\nbaz"', 'foo="bar\nbaz"'),
('text/html; charset=iso-8859-1',
- 'text/html; charset="iso-8859-1"'),
+ 'text/html; charset=iso-8859-1'),
+
+ ('text/html; charset="iso-8859/1"',
+ 'text/html; charset="iso-8859/1"'),
('foo="bar"; port="80,81"; discard, bar=baz',
'foo=bar; port="80,81"; discard, bar=baz'),
@@ -297,8 +307,8 @@ def test_roundtrip(self):
(r'Basic realm="\"foo\\\\bar\""',
r'Basic; realm="\"foo\\\\bar\""'),
- ('n; foo="foo;_", bar=foo!_',
- 'n; foo="foo;_", bar="foo!_"'),
+ ('n; foo="foo;_", bar="foo,_"',
+ 'n; foo="foo;_", bar="foo,_"'),
]
for arg, expect in tests:
@@ -553,7 +563,7 @@ def test_missing_value(self):
self.assertIsNone(cookie.value)
self.assertEqual(cookie.name, '"spam"')
self.assertEqual(lwp_cookie_str(cookie), (
- r'"spam"; path="/foo/"; domain="www.acme.com"; '
+ r'"spam"; path="/foo/"; domain=www.acme.com; '
'path_spec; discard; version=0'))
old_str = repr(c)
c.save(ignore_expires=True, ignore_discard=True)
diff --git
a/Misc/NEWS.d/next/Library/2025-02-27-14-25-01.gh-issue-130631.dmZcZM.rst
b/Misc/NEWS.d/next/Library/2025-02-27-14-25-01.gh-issue-130631.dmZcZM.rst
new file mode 100644
index 00000000000000..c9dc9ba87878c8
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2025-02-27-14-25-01.gh-issue-130631.dmZcZM.rst
@@ -0,0 +1,3 @@
+:func:`!http.cookiejar.join_header_words` is now more similar to the original
+Perl version. It now quotes the same set of characters and always quote
+values that end with ``"\n"``.
_______________________________________________
Python-checkins mailing list -- python-checkins@python.org
To unsubscribe send an email to python-checkins-le...@python.org
https://mail.python.org/mailman3/lists/python-checkins.python.org/
Member address: arch...@mail-archive.com
