gh-63293: Deprecate providing false values besides strings or bytes to urllib.parse functions

Doc/library/urllib.parse.rst

@@ -193,6 +193,11 @@ or on combining URL components into a URL string.
       Characters that affect netloc parsing under NFKC normalization will
       now raise :exc:`ValueError`.
 
+   .. versionchanged:: 3.14
+      Values for ``url`` and ``scheme`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 
 .. function:: parse_qs(qs, keep_blank_values=False, strict_parsing=False, encoding='utf-8', errors='replace', max_num_fields=None, separator='&')
 
@@ -282,6 +287,11 @@ or on combining URL components into a URL string.
       query parameter separator. This has been changed to allow only a single
       separator key, with ``&`` as the default separator.
 
+   .. versionchanged:: 3.14
+      Values for ``qs`` and ``separator`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 
 .. function:: urlunparse(parts)
 
@@ -291,6 +301,11 @@ or on combining URL components into a URL string.
    unnecessary delimiters (for example, a ``?`` with an empty query; the RFC
    states that these are equivalent).
 
+   .. versionchanged:: 3.14
+      Items in ``parts`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 
 .. function:: urlsplit(urlstring, scheme='', allow_fragments=True)
 
@@ -364,6 +379,11 @@ or on combining URL components into a URL string.
    .. versionchanged:: 3.12
       Leading WHATWG C0 control and space characters are stripped from the URL.
 
+   .. versionchanged:: 3.14
+      Values for the url argument and ``scheme`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 .. _WHATWG spec: https://url.spec.whatwg.org/#concept-basic-url-parser
 
 .. function:: urlunsplit(parts)
@@ -374,6 +394,11 @@ or on combining URL components into a URL string.
    URL that was parsed originally had unnecessary delimiters (for example, a ?
    with an empty query; the RFC states that these are equivalent).
 
+   .. versionchanged:: 3.14
+      Items in ``parts`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 
 .. function:: urljoin(base, url, allow_fragments=True)
 
@@ -408,6 +433,11 @@ or on combining URL components into a URL string.
 
       Behavior updated to match the semantics defined in :rfc:`3986`.
 
+   .. versionchanged:: 3.14
+      Values for ``base`` and ``url`` other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 
 .. function:: urldefrag(url)
 
@@ -433,6 +463,11 @@ or on combining URL components into a URL string.
    .. versionchanged:: 3.2
       Result is a structured object rather than a simple 2-tuple.
 
+   .. versionchanged:: 3.14
+      Values other than strings or bytes raise
+      :exc:`TypeError` if true or :exc:`DeprecationWarning` if false (to be
+      changed to :exc:`TypeError` in future versions of Python).
+
 .. function:: unwrap(url)
 
    Extract the url from a wrapped URL (that is, a string formatted as

Doc/whatsnew/3.14.rst

@@ -552,6 +552,11 @@ Deprecated
   use :func:`inspect.iscoroutinefunction` instead.
   (Contributed by Jiahao Li and Kumar Aditya in :gh:`122875`.)
 
+* Providing anything but a string or bytes object to :mod:`urllib.parse`
+  functions expecting strings or bytes now raises :exc:`DeprecationWarning`
+  if the value tests false, or :exc:`TypeError` if it tests true.
+  (Contributed by Jacob Walls in :issue:`19094`.)
+
 .. Add deprecations above alphabetically, not here at the end.
 
 .. include:: ../deprecations/c-api-pending-removal-in-3.15.rst

Lib/test/test_urlparse.py

@@ -1128,6 +1128,12 @@ def test_mixed_types_rejected(self):
         with self.assertRaisesRegex(TypeError, "Cannot mix str"):
             urllib.parse.urljoin(b"http://python.org", "http://python.org")
 
+    def test_non_string_true_values_rejected(self):
+        # True values raise informative TypeErrors
+        msg = "Expected a string or bytes object: got <class "
+        with self.assertRaisesRegex(TypeError, msg):
+            urllib.parse.urlsplit(1, b'http')
+
     def _check_result_type(self, str_type):
         num_args = len(str_type._fields)
         bytes_type = str_type._encoded_counterpart
@@ -1665,6 +1671,27 @@ def test_to_bytes_deprecation(self):
         self.assertEqual(str(cm.warning),
                          'urllib.parse.to_bytes() is deprecated as of 3.8')
 
+    def test_falsey_deprecation(self):
+        pattern = (
+            "Providing false values other than strings or bytes to urllib.parse "
+            "is deprecated: got <class "
+        )
+        cases = [
+            (urllib.parse.urljoin, ['http://www.python.org', []]),
+            (urllib.parse.urljoin, [[], b'docs']),
+            (urllib.parse.urlparse, [b'www.python.org', None]),
+            (urllib.parse.urlparse, [{}, '']),
+            (urllib.parse.urlsplit, [0, b'http']),
+            (urllib.parse.urlsplit, [b'http://www.python.org', None]),
+            (urllib.parse.urldefrag, [{}]),
+            (urllib.parse.urlunparse, [[None, b'www.python.org', None, None, None, None]]),
+            (urllib.parse.urlunsplit, [['http', 0, '', '', '']]),
+        ]
+        for callable, args in cases:
+            with self.subTest(callable=callable):
+                with self.assertWarnsRegex(DeprecationWarning, pattern):
+                    callable(*args)
+
 
 if __name__ == "__main__":
     unittest.main()

Lib/urllib/parse.py

@@ -122,6 +122,15 @@ def _coerce_args(*args):
     # an appropriate result coercion function
     #   - noop for str inputs
     #   - encoding function otherwise
+    for arg in args:
+        if not (isinstance(arg, str) or hasattr(arg, 'decode')):
+            if arg:
+                raise TypeError(f"Expected a string or bytes object: got {type(arg)}")
+            else:
+                warnings.warn(
+                    f"Providing false values other than strings or bytes "
+                    f"to urllib.parse is deprecated: got {type(arg)}",
+                    DeprecationWarning, stacklevel=3)
     str_input = isinstance(args[0], str)
     for arg in args[1:]:
         # We special-case the empty string to support the
@@ -564,12 +573,12 @@ def _urlunsplit(scheme, netloc, url, query, fragment):
 def urljoin(base, url, allow_fragments=True):
     """Join a base URL and a possibly relative URL to form an absolute
     interpretation of the latter."""
+    base, url, _coerce_result = _coerce_args(base, url)
     if not base:
-        return url
+        return _coerce_result(url)
     if not url:
-        return base
+        return _coerce_result(base)
 
-    base, url, _coerce_result = _coerce_args(base, url)
     bscheme, bnetloc, bpath, bquery, bfragment = \
             _urlsplit(base, None, allow_fragments)
     scheme, netloc, path, query, fragment = \

Misc/NEWS.d/next/Library/2021-06-11-20-00-16.bpo-19094.rMRoIL.rst

@@ -0,0 +1,3 @@
+Providing anything but a string or bytes object to :mod:`urllib.parse`
+functions expecting strings or bytes now raises :exc:`DeprecationWarning`
+if the value tests false, or :exc:`TypeError` if it tests true.