bpo-31658: Make xml.sax.parse accepting Path objects

BoboTiG · BoboTiG · commit 04498f804cac · 2019-01-02T16:53:30.000+01:00
diff --git a/Doc/library/xml.sax.reader.rst b/Doc/library/xml.sax.reader.rst
@@ -102,13 +102,16 @@ The :class:`XMLReader` interface supports the following methods:
 
    Process an input source, producing SAX events. The *source* object can be a
    system identifier (a string identifying the input source -- typically a file
-   name or a URL), a file-like object, or an :class:`InputSource` object. When
-   :meth:`parse` returns, the input is completely processed, and the parser object
-   can be discarded or reset.
+   name or a URL), a :class:`~pathlib.Path` object, a file-like object or an
+   :class:`InputSource` object. When :meth:`parse` returns, the input is
+   completely processed, and the parser object can be discarded or reset.
 
    .. versionchanged:: 3.5
       Added support of character streams.
 
+   .. versionchanged:: 3.8
+      Added support of :class:`~pathlib.Path` objects.
+
 
 .. method:: XMLReader.getContentHandler()
 
diff --git a/Lib/test/test_sax.py b/Lib/test/test_sax.py
@@ -18,6 +18,7 @@
 from io import BytesIO, StringIO
 import codecs
 import os.path
+import pathlib
 import shutil
 from urllib.error import URLError
 from test import support
@@ -182,6 +183,10 @@ def test_parse_bytes(self):
             with self.assertRaises(SAXException):
                 self.check_parse(f)
 
+    def test_parse_path_object(self):
+        make_xml_file(self.data, 'utf-8', None)
+        self.check_parse(pathlib.Path(TESTFN))
+
     def test_parse_InputSource(self):
         # accept data without declared but with explicitly specified encoding
         make_xml_file(self.data, 'iso-8859-1', None)
@@ -397,6 +402,13 @@ def test_string(self):
         self.checkContent(prep.getByteStream(),
                           b"This was read from a file.")
 
+    def test_path_objects(self):
+        # If the source is a Path object, use it as a system ID and open it.
+        prep = prepare_input_source(pathlib.Path(self.file))
+        self.assertIsNone(prep.getCharacterStream())
+        self.checkContent(prep.getByteStream(),
+                          b"This was read from a file.")
+
     def test_binary_file(self):
         # If the source is a binary file-like object, use it as a byte
         # stream.
diff --git a/Lib/xml/sax/saxutils.py b/Lib/xml/sax/saxutils.py
@@ -340,6 +340,8 @@ def prepare_input_source(source, base=""):
     """This function takes an InputSource and an optional base URL and
     returns a fully resolved InputSource object ready for reading."""
 
+    if isinstance(source, os.PathLike):
+        source = os.fspath(source)
     if isinstance(source, str):
         source = xmlreader.InputSource(source)
     elif hasattr(source, "read"):
diff --git a/Misc/NEWS.d/next/Library/2018-07-30-12-00-15.bpo-31658._bx7a_.rst b/Misc/NEWS.d/next/Library/2018-07-30-12-00-15.bpo-31658._bx7a_.rst
@@ -0,0 +1 @@
+Make xml.sax.parse accepting Path objects. Patch by Mickaël Schoentgen.

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1 @@`
	`1`	`+Make xml.sax.parse accepting Path objects. Patch by Mickaël Schoentgen.`