Move the codec decode type checks to bytes/bytearray.decode().
Use faster PyUnicode_FromEncodedObject() for bytes/bytearray.decode().
Add new PyCodec_KnownEncoding() API.
Add new PyUnicode_AsDecodedUnicode() and PyUnicode_AsEncodedUnicode() APIs.
Add missing PyUnicode_AsDecodedObject() to unicodeobject.h
Fix punicode codec to also work on memoryviews.
diff --git a/Include/codecs.h b/Include/codecs.h
index 0d76241..c979e86 100644
--- a/Include/codecs.h
+++ b/Include/codecs.h
@@ -27,7 +27,7 @@
PyObject *search_function
);
-/* Codec register lookup API.
+/* Codec registry lookup API.
Looks up the given encoding and returns a CodecInfo object with
function attributes which implement the different aspects of
@@ -49,6 +49,17 @@
const char *encoding
);
+/* Codec registry encoding check API.
+
+ Returns 1/0 depending on whether there is a registered codec for
+ the given encoding.
+
+*/
+
+PyAPI_FUNC(int) PyCodec_KnownEncoding(
+ const char *encoding
+ );
+
/* Generic codec based encoding API.
object is passed through the encoder function found for the given
diff --git a/Include/unicodeobject.h b/Include/unicodeobject.h
index 384cd55..7af2eba 100644
--- a/Include/unicodeobject.h
+++ b/Include/unicodeobject.h
@@ -139,8 +139,11 @@
# define PyUnicode_AsASCIIString PyUnicodeUCS2_AsASCIIString
# define PyUnicode_AsCharmapString PyUnicodeUCS2_AsCharmapString
+# define PyUnicode_AsDecodedObject PyUnicodeUCS2_AsDecodedObject
+# define PyUnicode_AsDecodedUnicode PyUnicodeUCS2_AsDecodedUnicode
# define PyUnicode_AsEncodedObject PyUnicodeUCS2_AsEncodedObject
# define PyUnicode_AsEncodedString PyUnicodeUCS2_AsEncodedString
+# define PyUnicode_AsEncodedUnicode PyUnicodeUCS2_AsEncodedUnicode
# define PyUnicode_AsLatin1String PyUnicodeUCS2_AsLatin1String
# define PyUnicode_AsRawUnicodeEscapeString PyUnicodeUCS2_AsRawUnicodeEscapeString
# define PyUnicode_AsUTF32String PyUnicodeUCS2_AsUTF32String
@@ -233,8 +236,11 @@
# define PyUnicode_AsASCIIString PyUnicodeUCS4_AsASCIIString
# define PyUnicode_AsCharmapString PyUnicodeUCS4_AsCharmapString
+# define PyUnicode_AsDecodedObject PyUnicodeUCS4_AsDecodedObject
+# define PyUnicode_AsDecodedUnicode PyUnicodeUCS4_AsDecodedUnicode
# define PyUnicode_AsEncodedObject PyUnicodeUCS4_AsEncodedObject
# define PyUnicode_AsEncodedString PyUnicodeUCS4_AsEncodedString
+# define PyUnicode_AsEncodedUnicode PyUnicodeUCS4_AsEncodedUnicode
# define PyUnicode_AsLatin1String PyUnicodeUCS4_AsLatin1String
# define PyUnicode_AsRawUnicodeEscapeString PyUnicodeUCS4_AsRawUnicodeEscapeString
# define PyUnicode_AsUTF32String PyUnicodeUCS4_AsUTF32String
@@ -744,6 +750,24 @@
const char *errors /* error handling */
);
+/* Decode a Unicode object unicode and return the result as Python
+ object. */
+
+PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedObject(
+ PyObject *unicode, /* Unicode object */
+ const char *encoding, /* encoding */
+ const char *errors /* error handling */
+ );
+
+/* Decode a Unicode object unicode and return the result as Unicode
+ object. */
+
+PyAPI_FUNC(PyObject*) PyUnicode_AsDecodedUnicode(
+ PyObject *unicode, /* Unicode object */
+ const char *encoding, /* encoding */
+ const char *errors /* error handling */
+ );
+
/* Encodes a Py_UNICODE buffer of the given size and returns a
Python string object. */
@@ -772,11 +796,21 @@
const char *errors /* error handling */
);
+/* Encodes a Unicode object and returns the result as Unicode
+ object. */
+
+PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedUnicode(
+ PyObject *unicode, /* Unicode object */
+ const char *encoding, /* encoding */
+ const char *errors /* error handling */
+ );
+
+/* Build an encoding map. */
+
PyAPI_FUNC(PyObject*) PyUnicode_BuildEncodingMap(
PyObject* string /* 256 character map */
);
-
/* --- UTF-7 Codecs ------------------------------------------------------- */
PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7(