Issue 19548: update codecs module documentation

ncoghlan · ncoghlan · commit b9fdb7a452c2 · 2015-01-07T00:22:00.000+10:00
- clarified the distinction between text encodings and other codecs
- clarified relationship with builtin open and the io module
- consolidated documentation of error handlers into one section
- clarified type constraints of some behaviours
- added tests for some of the new statements in the docs
diff --git a/Doc/glossary.rst b/Doc/glossary.rst
@@ -820,10 +820,13 @@ Glossary
       :meth:`~collections.somenamedtuple._asdict`. Examples of struct sequences
       include :data:`sys.float_info` and the return value of :func:`os.stat`.
 
+   text encoding
+      A codec which encodes Unicode strings to bytes.
+
    text file
       A :term:`file object` able to read and write :class:`str` objects.
       Often, a text file actually accesses a byte-oriented datastream
-      and handles the text encoding automatically.
+      and handles the :term:`text encoding` automatically.
 
       .. seealso::
          A :term:`binary file` reads and write :class:`bytes` objects.
diff --git a/Doc/library/codecs.rst b/Doc/library/codecs.rst
diff --git a/Doc/library/functions.rst b/Doc/library/functions.rst
@@ -939,15 +939,17 @@ are always available.  They are listed here in alphabetical order.
    *encoding* is the name of the encoding used to decode or encode the file.
    This should only be used in text mode.  The default encoding is platform
    dependent (whatever :func:`locale.getpreferredencoding` returns), but any
-   encoding supported by Python can be used.  See the :mod:`codecs` module for
+   :term:`text encoding` supported by Python
+   can be used.  See the :mod:`codecs` module for
    the list of supported encodings.
 
    *errors* is an optional string that specifies how encoding and decoding
    errors are to be handled--this cannot be used in binary mode.
-   A variety of standard error handlers are available, though any
+   A variety of standard error handlers are available
+   (listed under :ref:`error-handlers`), though any
    error handling name that has been registered with
    :func:`codecs.register_error` is also valid.  The standard names
-   are:
+   include:
 
    * ``'strict'`` to raise a :exc:`ValueError` exception if there is
      an encoding error.  The default value of ``None`` has the same
diff --git a/Doc/library/stdtypes.rst b/Doc/library/stdtypes.rst
@@ -1512,7 +1512,7 @@ expression support in the :mod:`re` module).
    a :exc:`UnicodeError`. Other possible
    values are ``'ignore'``, ``'replace'``, ``'xmlcharrefreplace'``,
    ``'backslashreplace'`` and any other name registered via
-   :func:`codecs.register_error`, see section :ref:`codec-base-classes`. For a
+   :func:`codecs.register_error`, see section :ref:`error-handlers`. For a
    list of possible encodings, see section :ref:`standard-encodings`.
 
    .. versionchanged:: 3.1
@@ -2384,7 +2384,7 @@ arbitrary binary data.
    error handling scheme.  The default for *errors* is ``'strict'``, meaning
    that encoding errors raise a :exc:`UnicodeError`.  Other possible values are
    ``'ignore'``, ``'replace'`` and any other name registered via
-   :func:`codecs.register_error`, see section :ref:`codec-base-classes`. For a
+   :func:`codecs.register_error`, see section :ref:`error-handlers`. For a
    list of possible encodings, see section :ref:`standard-encodings`.
 
    .. note::
diff --git a/Doc/library/tarfile.rst b/Doc/library/tarfile.rst
@@ -794,7 +794,7 @@ metadata must be either decoded or encoded. If *encoding* is not set
 appropriately, this conversion may fail.
 
 The *errors* argument defines how characters are treated that cannot be
-converted. Possible values are listed in section :ref:`codec-base-classes`.
+converted. Possible values are listed in section :ref:`error-handlers`.
 The default scheme is ``'surrogateescape'`` which Python also uses for its
 file system calls, see :ref:`os-filenames`.
 
diff --git a/Lib/codecs.py b/Lib/codecs.py
@@ -346,8 +346,7 @@ def __init__(self, stream, errors='strict'):
 
         """ Creates a StreamWriter instance.
 
-            stream must be a file-like object open for writing
-            (binary) data.
+            stream must be a file-like object open for writing.
 
             The StreamWriter may use different error handling
             schemes by providing the errors keyword argument. These
@@ -421,8 +420,7 @@ def __init__(self, stream, errors='strict'):
 
         """ Creates a StreamReader instance.
 
-            stream must be a file-like object open for reading
-            (binary) data.
+            stream must be a file-like object open for reading.
 
             The StreamReader may use different error handling
             schemes by providing the errors keyword argument. These
@@ -450,13 +448,12 @@ def read(self, size=-1, chars=-1, firstline=False):
         """ Decodes data from the stream self.stream and returns the
             resulting object.
 
-            chars indicates the number of characters to read from the
-            stream. read() will never return more than chars
-            characters, but it might return less, if there are not enough
-            characters available.
+            chars indicates the number of decoded code points or bytes to
+            return. read() will never return more data than requested,
+            but it might return less, if there is not enough available.
 
-            size indicates the approximate maximum number of bytes to
-            read from the stream for decoding purposes. The decoder
+            size indicates the approximate maximum number of decoded
+            bytes or code points to read for decoding. The decoder
             can modify this setting as appropriate. The default value
             -1 indicates to read and decode as much as possible.  size
             is intended to prevent having to decode huge files in one
@@ -467,7 +464,7 @@ def read(self, size=-1, chars=-1, firstline=False):
             will be returned, the rest of the input will be kept until the
             next call to read().
 
-            The method should use a greedy read strategy meaning that
+            The method should use a greedy read strategy, meaning that
             it should read as much data as is allowed within the
             definition of the encoding and the given size, e.g.  if
             optional encoding endings or state markers are available
@@ -602,7 +599,7 @@ def readline(self, size=None, keepends=True):
     def readlines(self, sizehint=None, keepends=True):
 
         """ Read all lines available on the input stream
-            and return them as list of lines.
+            and return them as a list.
 
             Line breaks are implemented using the codec's decoder
             method and are included in the list entries.
@@ -750,19 +747,18 @@ def __exit__(self, type, value, tb):
 
 class StreamRecoder:
 
-    """ StreamRecoder instances provide a frontend - backend
-        view of encoding data.
+    """ StreamRecoder instances translate data from one encoding to another.
 
         They use the complete set of APIs returned by the
         codecs.lookup() function to implement their task.
 
-        Data written to the stream is first decoded into an
-        intermediate format (which is dependent on the given codec
-        combination) and then written to the stream using an instance
-        of the provided Writer class.
+        Data written to the StreamRecoder is first decoded into an
+        intermediate format (depending on the "decode" codec) and then
+        written to the underlying stream using an instance of the provided
+        Writer class.
 
-        In the other direction, data is read from the stream using a
-        Reader instance and then return encoded data to the caller.
+        In the other direction, data is read from the underlying stream using
+        a Reader instance and then encoded and returned to the caller.
 
     """
     # Optional attributes set by the file wrappers below
@@ -774,22 +770,17 @@ def __init__(self, stream, encode, decode, Reader, Writer,
 
         """ Creates a StreamRecoder instance which implements a two-way
             conversion: encode and decode work on the frontend (the
-            input to .read() and output of .write()) while
-            Reader and Writer work on the backend (reading and
-            writing to the stream).
+            data visible to .read() and .write()) while Reader and Writer
+            work on the backend (the data in stream).
 
-            You can use these objects to do transparent direct
-            recodings from e.g. latin-1 to utf-8 and back.
+            You can use these objects to do transparent
+            transcodings from e.g. latin-1 to utf-8 and back.
 
             stream must be a file-like object.
 
-            encode, decode must adhere to the Codec interface, Reader,
+            encode and decode must adhere to the Codec interface; Reader and
             Writer must be factory functions or classes providing the
-            StreamReader, StreamWriter interface resp.
-
-            encode and decode are needed for the frontend translation,
-            Reader and Writer for the backend translation. Unicode is
-            used as intermediate encoding.
+            StreamReader and StreamWriter interfaces resp.
 
             Error handling is done in the same way as defined for the
             StreamWriter/Readers.
@@ -864,7 +855,7 @@ def __exit__(self, type, value, tb):
 
 ### Shortcuts
 
-def open(filename, mode='rb', encoding=None, errors='strict', buffering=1):
+def open(filename, mode='r', encoding=None, errors='strict', buffering=1):
 
     """ Open an encoded file using the given mode and return
         a wrapped version providing transparent encoding/decoding.
@@ -874,10 +865,8 @@ def open(filename, mode='rb', encoding=None, errors='strict', buffering=1):
         codecs. Output is also codec dependent and will usually be
         Unicode as well.
 
-        Files are always opened in binary mode, even if no binary mode
-        was specified. This is done to avoid data loss due to encodings
-        using 8-bit values. The default file mode is 'rb' meaning to
-        open the file in binary read mode.
+        Underlying encoded files are always opened in binary mode.
+        The default file mode is 'r', meaning to open the file in read mode.
 
         encoding specifies the encoding which is to be used for the
         file.
@@ -913,13 +902,13 @@ def EncodedFile(file, data_encoding, file_encoding=None, errors='strict'):
     """ Return a wrapped version of file which provides transparent
         encoding translation.
 
-        Strings written to the wrapped file are interpreted according
-        to the given data_encoding and then written to the original
-        file as string using file_encoding. The intermediate encoding
+        Data written to the wrapped file is decoded according
+        to the given data_encoding and then encoded to the underlying
+        file using file_encoding. The intermediate data type
         will usually be Unicode but depends on the specified codecs.
 
-        Strings are read from the file using file_encoding and then
-        passed back to the caller as string using data_encoding.
+        Bytes read from the file are decoded using file_encoding and then
+        passed back to the caller encoded using data_encoding.
 
         If file_encoding is not given, it defaults to data_encoding.
 
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
@@ -1139,6 +1139,8 @@ def test_recoding(self):
         # Python used to crash on this at exit because of a refcount
         # bug in _codecsmodule.c
 
+        self.assertTrue(f.closed)
+
 # From RFC 3492
 punycode_testcases = [
     # A Arabic (Egyptian):
@@ -1591,6 +1593,16 @@ def test_incremental_encode(self):
         self.assertEqual(encoder.encode("ample.org."), b"xn--xample-9ta.org.")
         self.assertEqual(encoder.encode("", True), b"")
 
+    def test_errors(self):
+        """Only supports "strict" error handler"""
+        "python.org".encode("idna", "strict")
+        b"python.org".decode("idna", "strict")
+        for errors in ("ignore", "replace", "backslashreplace",
+                "surrogateescape"):
+            self.assertRaises(Exception, "python.org".encode, "idna", errors)
+            self.assertRaises(Exception,
+                b"python.org".decode, "idna", errors)
+
 class CodecsModuleTest(unittest.TestCase):
 
     def test_decode(self):
@@ -1668,6 +1680,24 @@ def test_all(self):
         for api in codecs.__all__:
             getattr(codecs, api)
 
+    def test_open(self):
+        self.addCleanup(support.unlink, support.TESTFN)
+        for mode in ('w', 'r', 'r+', 'w+', 'a', 'a+'):
+            with self.subTest(mode), \
+                    codecs.open(support.TESTFN, mode, 'ascii') as file:
+                self.assertIsInstance(file, codecs.StreamReaderWriter)
+
+    def test_undefined(self):
+        self.assertRaises(UnicodeError, codecs.encode, 'abc', 'undefined')
+        self.assertRaises(UnicodeError, codecs.decode, b'abc', 'undefined')
+        self.assertRaises(UnicodeError, codecs.encode, '', 'undefined')
+        self.assertRaises(UnicodeError, codecs.decode, b'', 'undefined')
+        for errors in ('strict', 'ignore', 'replace', 'backslashreplace'):
+            self.assertRaises(UnicodeError,
+                codecs.encode, 'abc', 'undefined', errors)
+            self.assertRaises(UnicodeError,
+                codecs.decode, b'abc', 'undefined', errors)
+
 class StreamReaderTest(unittest.TestCase):
 
     def setUp(self):
@@ -1801,13 +1831,10 @@ def test_basic(self):
 #    "undefined"
 
 # The following encodings don't work in stateful mode
-broken_unicode_with_streams = [
+broken_unicode_with_stateful = [
     "punycode",
     "unicode_internal"
 ]
-broken_incremental_coders = broken_unicode_with_streams + [
-    "idna",
-]
 
 class BasicUnicodeTest(unittest.TestCase, MixInCheckStateHandling):
     def test_basics(self):
@@ -1827,7 +1854,7 @@ def test_basics(self):
                 (chars, size) = codecs.getdecoder(encoding)(b)
                 self.assertEqual(chars, s, "encoding=%r" % encoding)
 
-            if encoding not in broken_unicode_with_streams:
+            if encoding not in broken_unicode_with_stateful:
                 # check stream reader/writer
                 q = Queue(b"")
                 writer = codecs.getwriter(encoding)(q)
@@ -1845,7 +1872,7 @@ def test_basics(self):
                     decodedresult += reader.read()
                 self.assertEqual(decodedresult, s, "encoding=%r" % encoding)
 
-            if encoding not in broken_incremental_coders:
+            if encoding not in broken_unicode_with_stateful:
                 # check incremental decoder/encoder and iterencode()/iterdecode()
                 try:
                     encoder = codecs.getincrementalencoder(encoding)()
@@ -1894,7 +1921,7 @@ def test_basics_capi(self):
         from _testcapi import codec_incrementalencoder, codec_incrementaldecoder
         s = "abc123"  # all codecs should be able to encode these
         for encoding in all_unicode_encodings:
-            if encoding not in broken_incremental_coders:
+            if encoding not in broken_unicode_with_stateful:
                 # check incremental decoder/encoder (fetched via the C API)
                 try:
                     cencoder = codec_incrementalencoder(encoding)
@@ -1934,7 +1961,7 @@ def test_seek(self):
         for encoding in all_unicode_encodings:
             if encoding == "idna": # FIXME: See SF bug #1163178
                 continue
-            if encoding in broken_unicode_with_streams:
+            if encoding in broken_unicode_with_stateful:
                 continue
             reader = codecs.getreader(encoding)(io.BytesIO(s.encode(encoding)))
             for t in range(5):
@@ -1967,7 +1994,7 @@ def test_decoder_state(self):
         # Check that getstate() and setstate() handle the state properly
         u = "abc123"
         for encoding in all_unicode_encodings:
-            if encoding not in broken_incremental_coders:
+            if encoding not in broken_unicode_with_stateful:
                 self.check_state_handling_decode(encoding, u, u.encode(encoding))
                 self.check_state_handling_encode(encoding, u, u.encode(encoding))
 
@@ -2171,6 +2198,7 @@ def test_encodedfile(self):
         f = io.BytesIO(b"\xc3\xbc")
         with codecs.EncodedFile(f, "latin-1", "utf-8") as ef:
             self.assertEqual(ef.read(), b"\xfc")
+        self.assertTrue(f.closed)
 
     def test_streamreaderwriter(self):
         f = io.BytesIO(b"\xc3\xbc")
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -265,6 +265,10 @@ IDLE
 Tests
 -----
 
+- Issue #19548: Added some additional checks to test_codecs to ensure that
+  statements in the updated documentation remain accurate. Patch by Martin
+  Panter.
+
 - Issue #22838: All test_re tests now work with unittest test discovery.
 
 - Issue #22173: Update lib2to3 tests to use unittest test discovery.
@@ -297,6 +301,10 @@ Build
 Documentation
 -------------
 
+- Issue #19548: Update the codecs module documentation to better cover the
+  distinction between text encodings and other codecs, together with other
+  clarifications. Patch by Martin Panter.
+
 - Issue #22914: Update the Python 2/3 porting HOWTO to describe a more automated
   approach.
 
diff --git a/Modules/_codecsmodule.c b/Modules/_codecsmodule.c
@@ -54,9 +54,9 @@ PyDoc_STRVAR(register__doc__,
 "register(search_function)\n\
 \n\
 Register a codec search function. Search functions are expected to take\n\
-one argument, the encoding name in all lower case letters, and return\n\
-a tuple of functions (encoder, decoder, stream_reader, stream_writer)\n\
-(or a CodecInfo object).");
+one argument, the encoding name in all lower case letters, and either\n\
+return None, or a tuple of functions (encoder, decoder, stream_reader,\n\
+stream_writer) (or a CodecInfo object).");
 
 static
 PyObject *codec_register(PyObject *self, PyObject *search_function)
