[Python-checkins] bpo-22347: Update mimetypes.guess_type to allow proper parsing of URLs (GH-15522)

[Miss Islington (bot)]

https://github.com/python/cpython/commit/87bd2071c756188b6cd577889fb1682831142ceb
commit: 87bd2071c756188b6cd577889fb1682831142ceb
branch: master
author: Dong-hee Na <donghee.na92 at gmail.com>
committer: Miss Islington (bot) <31488909+miss-islington at users.noreply.github.com>
date: 2019-09-04T17:34:35-07:00
summary:

bpo-22347: Update mimetypes.guess_type to allow proper parsing of URLs (GH-15522)



https://bugs.python.org/issue22347

files:
A Misc/NEWS.d/next/Library/2019-08-27-01-03-26.bpo-22347._TRpYr.rst
M Lib/mimetypes.py
M Lib/test/test_mimetypes.py
M Lib/test/test_urllib2.py

diff --git a/Lib/mimetypes.py b/Lib/mimetypes.py
index 01a16fdf9aa1..f38005c9d295 100644
--- a/Lib/mimetypes.py
+++ b/Lib/mimetypes.py
@@ -114,7 +114,8 @@ def guess_type(self, url, strict=True):
         but non-standard types.
         """
         url = os.fspath(url)
-        scheme, url = urllib.parse._splittype(url)
+        p = urllib.parse.urlparse(url)
+        scheme, url = p.scheme, p.path
         if scheme == 'data':
             # syntax of data URLs:
             # dataurl   := "data:" [ mediatype ] [ ";base64" ] "," data
diff --git a/Lib/test/test_mimetypes.py b/Lib/test/test_mimetypes.py
index bfd5eeedaa77..7761c3fe867a 100644
--- a/Lib/test/test_mimetypes.py
+++ b/Lib/test/test_mimetypes.py
@@ -51,6 +51,14 @@ def test_non_standard_types(self):
         eq(self.db.guess_type('foo.xul', strict=False), ('text/xul', None))
         eq(self.db.guess_extension('image/jpg', strict=False), '.jpg')
 
+    def test_url(self):
+        result = self.db.guess_type('http://host.html')
+        msg = 'URL only has a host name, not a file'
+        self.assertSequenceEqual(result, (None, None), msg)
+        result = self.db.guess_type('http://example.com/host.html')
+        msg = 'Should be text/html'
+        self.assertSequenceEqual(result, ('text/html', None), msg)
+
     def test_guess_all_types(self):
         eq = self.assertEqual
         unless = self.assertTrue
diff --git a/Lib/test/test_urllib2.py b/Lib/test/test_urllib2.py
index 7b576db4e3aa..c228fa7d1dc1 100644
--- a/Lib/test/test_urllib2.py
+++ b/Lib/test/test_urllib2.py
@@ -742,7 +742,7 @@ def connect_ftp(self, user, passwd, host, port, dirs,
              ["foo", "bar"], "", None),
             ("ftp://localhost/baz.gif;type=a",
              "localhost", ftplib.FTP_PORT, "", "", "A",
-             [], "baz.gif", None),  # XXX really this should guess image/gif
+             [], "baz.gif", "image/gif"),
             ]:
             req = Request(url)
             req.timeout = None
diff --git a/Misc/NEWS.d/next/Library/2019-08-27-01-03-26.bpo-22347._TRpYr.rst b/Misc/NEWS.d/next/Library/2019-08-27-01-03-26.bpo-22347._TRpYr.rst
new file mode 100644
index 000000000000..1a3c19938217
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2019-08-27-01-03-26.bpo-22347._TRpYr.rst
@@ -0,0 +1,2 @@
+Update mimetypes.guess_type to allow proper parsing of URLs with only a host name.
+Patch by Dong-hee Na.