Python/codecs.c

/* ------------------------------------------------------------------------

   Python Codec Registry and support functions

Written by Marc-Andre Lemburg (mal@lemburg.com).

Copyright (c) Corporation for National Research Initiatives.

   ------------------------------------------------------------------------ */

#include "Python.h"
#include <ctype.h>

const char *Py_hexdigits = "0123456789abcdef";

/* --- Codec Registry ----------------------------------------------------- */

/* Import the standard encodings package which will register the first
   codec search function.

   This is done in a lazy way so that the Unicode implementation does
   not downgrade startup time of scripts not needing it.

   ImportErrors are silently ignored by this function. Only one try is
   made.

*/

static int _PyCodecRegistry_Init(void); /* Forward */

int PyCodec_Register(PyObject *search_function)
{
    PyInterpreterState *interp = PyThreadState_GET()->interp;
    if (interp->codec_search_path == NULL && _PyCodecRegistry_Init())
        goto onError;
    if (search_function == NULL) {
        PyErr_BadArgument();
        goto onError;
    }
    if (!PyCallable_Check(search_function)) {
        PyErr_SetString(PyExc_TypeError, "argument must be callable");
        goto onError;
    }
    return PyList_Append(interp->codec_search_path, search_function);

 onError:
    return -1;
}

/* Convert a string to a normalized Python string: all characters are
   converted to lower case, spaces are replaced with underscores. */

static
PyObject *normalizestring(const char *string)
{
    size_t i;
    size_t len = strlen(string);
    char *p;
    PyObject *v;

    if (len > PY_SSIZE_T_MAX) {
        PyErr_SetString(PyExc_OverflowError, "string is too large");
        return NULL;
    }

    p = PyMem_Malloc(len + 1);
    if (p == NULL)
        return PyErr_NoMemory();
    for (i = 0; i < len; i++) {
        char ch = string[i];
        if (ch == ' ')
            ch = '-';
        else
            ch = Py_TOLOWER(Py_CHARMASK(ch));
        p[i] = ch;
    }
    p[i] = '\0';
    v = PyUnicode_FromString(p);
    if (v == NULL)
        return NULL;
    PyMem_Free(p);
    return v;
}

/* Lookup the given encoding and return a tuple providing the codec
   facilities.

   The encoding string is looked up converted to all lower-case
   characters. This makes encodings looked up through this mechanism
   effectively case-insensitive.

   If no codec is found, a LookupError is set and NULL returned.

   As side effect, this tries to load the encodings package, if not
   yet done. This is part of the lazy load strategy for the encodings
   package.

*/

PyObject *_PyCodec_Lookup(const char *encoding)
{
    PyInterpreterState *interp;
    PyObject *result, *args = NULL, *v;
    Py_ssize_t i, len;

    if (encoding == NULL) {
        PyErr_BadArgument();
        goto onError;
    }

    interp = PyThreadState_GET()->interp;
    if (interp->codec_search_path == NULL && _PyCodecRegistry_Init())
        goto onError;

    /* Convert the encoding to a normalized Python string: all
       characters are converted to lower case, spaces and hyphens are
       replaced with underscores. */
    v = normalizestring(encoding);
    if (v == NULL)
        goto onError;
    PyUnicode_InternInPlace(&v);

    /* First, try to lookup the name in the registry dictionary */
    result = PyDict_GetItem(interp->codec_search_cache, v);
    if (result != NULL) {
        Py_INCREF(result);
        Py_DECREF(v);
        return result;
    }

    /* Next, scan the search functions in order of registration */
    args = PyTuple_New(1);
    if (args == NULL)
        goto onError;
    PyTuple_SET_ITEM(args,0,v);

    len = PyList_Size(interp->codec_search_path);
    if (len < 0)
        goto onError;
    if (len == 0) {
        PyErr_SetString(PyExc_LookupError,
                        "no codec search functions registered: "
                        "can't find encoding");
        goto onError;
    }

    for (i = 0; i < len; i++) {
        PyObject *func;

        func = PyList_GetItem(interp->codec_search_path, i);
        if (func == NULL)
            goto onError;
        result = PyEval_CallObject(func, args);
        if (result == NULL)
            goto onError;
        if (result == Py_None) {
            Py_DECREF(result);
            continue;
        }
        if (!PyTuple_Check(result) || PyTuple_GET_SIZE(result) != 4) {
            PyErr_SetString(PyExc_TypeError,
                            "codec search functions must return 4-tuples");
            Py_DECREF(result);
            goto onError;
        }
        break;
    }
    if (i == len) {
        /* XXX Perhaps we should cache misses too ? */
        PyErr_Format(PyExc_LookupError,
                     "unknown encoding: %s", encoding);
        goto onError;
    }

    /* Cache and return the result */
    if (PyDict_SetItem(interp->codec_search_cache, v, result) < 0) {
        Py_DECREF(result);
        goto onError;
    }
    Py_DECREF(args);
    return result;

 onError:
    Py_XDECREF(args);
    return NULL;
}

/* Codec registry encoding check API. */

int PyCodec_KnownEncoding(const char *encoding)
{
    PyObject *codecs;

    codecs = _PyCodec_Lookup(encoding);
    if (!codecs) {
        PyErr_Clear();
        return 0;
    }
    else {
        Py_DECREF(codecs);
        return 1;
    }
}

static
PyObject *args_tuple(PyObject *object,
                     const char *errors)
{
    PyObject *args;

    args = PyTuple_New(1 + (errors != NULL));
    if (args == NULL)
        return NULL;
    Py_INCREF(object);
    PyTuple_SET_ITEM(args,0,object);
    if (errors) {
        PyObject *v;

        v = PyUnicode_FromString(errors);
        if (v == NULL) {
            Py_DECREF(args);
            return NULL;
        }
        PyTuple_SET_ITEM(args, 1, v);
    }
    return args;
}

/* Helper function to get a codec item */

static
PyObject *codec_getitem(const char *encoding, int index)
{
    PyObject *codecs;
    PyObject *v;

    codecs = _PyCodec_Lookup(encoding);
    if (codecs == NULL)
        return NULL;
    v = PyTuple_GET_ITEM(codecs, index);
    Py_DECREF(codecs);
    Py_INCREF(v);
    return v;
}

/* Helper function to create an incremental codec. */

static
PyObject *codec_getincrementalcodec(const char *encoding,
                                    const char *errors,
                                    const char *attrname)
{
    PyObject *codecs, *ret, *inccodec;

    codecs = _PyCodec_Lookup(encoding);
    if (codecs == NULL)
        return NULL;
    inccodec = PyObject_GetAttrString(codecs, attrname);
    Py_DECREF(codecs);
    if (inccodec == NULL)
        return NULL;
    if (errors)
        ret = PyObject_CallFunction(inccodec, "s", errors);
    else
        ret = PyObject_CallFunction(inccodec, NULL);
    Py_DECREF(inccodec);
    return ret;
}

/* Helper function to create a stream codec. */

static
PyObject *codec_getstreamcodec(const char *encoding,
                               PyObject *stream,
                               const char *errors,
                               const int index)
{
    PyObject *codecs, *streamcodec, *codeccls;

    codecs = _PyCodec_Lookup(encoding);
    if (codecs == NULL)
        return NULL;

    codeccls = PyTuple_GET_ITEM(codecs, index);
    if (errors != NULL)
        streamcodec = PyObject_CallFunction(codeccls, "Os", stream, errors);
    else
        streamcodec = PyObject_CallFunction(codeccls, "O", stream);
    Py_DECREF(codecs);
    return streamcodec;
}

/* Convenience APIs to query the Codec registry.

   All APIs return a codec object with incremented refcount.

 */

PyObject *PyCodec_Encoder(const char *encoding)
{
    return codec_getitem(encoding, 0);
}

PyObject *PyCodec_Decoder(const char *encoding)
{
    return codec_getitem(encoding, 1);
}

PyObject *PyCodec_IncrementalEncoder(const char *encoding,
                                     const char *errors)
{
    return codec_getincrementalcodec(encoding, errors, "incrementalencoder");
}

PyObject *PyCodec_IncrementalDecoder(const char *encoding,
                                     const char *errors)
{
    return codec_getincrementalcodec(encoding, errors, "incrementaldecoder");
}

PyObject *PyCodec_StreamReader(const char *encoding,
                               PyObject *stream,
                               const char *errors)
{
    return codec_getstreamcodec(encoding, stream, errors, 2);
}

PyObject *PyCodec_StreamWriter(const char *encoding,
                               PyObject *stream,
                               const char *errors)
{
    return codec_getstreamcodec(encoding, stream, errors, 3);
}

/* Helper that tries to ensure the reported exception chain indicates the
 * codec that was invoked to trigger the failure without changing the type
 * of the exception raised.
 */
static void
wrap_codec_error(const char *operation,
                 const char *encoding)
{
    /* TrySetFromCause will replace the active exception with a suitably
     * updated clone if it can, otherwise it will leave the original
     * exception alone.
     */
    _PyErr_TrySetFromCause("%s with '%s' codec failed",
                           operation, encoding);
}

/* Encode an object (e.g. an Unicode object) using the given encoding
   and return the resulting encoded object (usually a Python string).

   errors is passed to the encoder factory as argument if non-NULL. */

PyObject *PyCodec_Encode(PyObject *object,
                         const char *encoding,
                         const char *errors)
{
    PyObject *encoder = NULL;
    PyObject *args = NULL, *result = NULL;
    PyObject *v = NULL;

    encoder = PyCodec_Encoder(encoding);
    if (encoder == NULL)
        goto onError;

    args = args_tuple(object, errors);
    if (args == NULL)
        goto onError;

    result = PyEval_CallObject(encoder, args);
    if (result == NULL) {
        wrap_codec_error("encoding", encoding);
        goto onError;
    }

    if (!PyTuple_Check(result) ||
        PyTuple_GET_SIZE(result) != 2) {
        PyErr_SetString(PyExc_TypeError,
                        "encoder must return a tuple (object, integer)");
        goto onError;
    }
    v = PyTuple_GET_ITEM(result,0);
    Py_INCREF(v);
    /* We don't check or use the second (integer) entry. */

    Py_DECREF(args);
    Py_DECREF(encoder);
    Py_DECREF(result);
    return v;

 onError:
    Py_XDECREF(result);
    Py_XDECREF(args);
    Py_XDECREF(encoder);
    return NULL;
}

/* Decode an object (usually a Python string) using the given encoding
   and return an equivalent object (e.g. an Unicode object).

   errors is passed to the decoder factory as argument if non-NULL. */

PyObject *PyCodec_Decode(PyObject *object,
                         const char *encoding,
                         const char *errors)
{
    PyObject *decoder = NULL;
    PyObject *args = NULL, *result = NULL;
    PyObject *v;

    decoder = PyCodec_Decoder(encoding);
    if (decoder == NULL)
        goto onError;

    args = args_tuple(object, errors);
    if (args == NULL)
        goto onError;

    result = PyEval_CallObject(decoder,args);
    if (result == NULL) {
        wrap_codec_error("decoding", encoding);
        goto onError;
    }
    if (!PyTuple_Check(result) ||
        PyTuple_GET_SIZE(result) != 2) {
        PyErr_SetString(PyExc_TypeError,
                        "decoder must return a tuple (object,integer)");
        goto onError;
    }
    v = PyTuple_GET_ITEM(result,0);
    Py_INCREF(v);
    /* We don't check or use the second (integer) entry. */

    Py_DECREF(args);
    Py_DECREF(decoder);
    Py_DECREF(result);
    return v;

 onError:
    Py_XDECREF(args);
    Py_XDECREF(decoder);
    Py_XDECREF(result);
    return NULL;
}

/* Register the error handling callback function error under the name
   name. This function will be called by the codec when it encounters
   an unencodable characters/undecodable bytes and doesn't know the
   callback name, when name is specified as the error parameter
   in the call to the encode/decode function.
   Return 0 on success, -1 on error */
int PyCodec_RegisterError(const char *name, PyObject *error)
{
    PyInterpreterState *interp = PyThreadState_GET()->interp;
    if (interp->codec_search_path == NULL && _PyCodecRegistry_Init())
        return -1;
    if (!PyCallable_Check(error)) {
        PyErr_SetString(PyExc_TypeError, "handler must be callable");
        return -1;
    }
    return PyDict_SetItemString(interp->codec_error_registry,
                                name, error);
}

/* Lookup the error handling callback function registered under the
   name error. As a special case NULL can be passed, in which case
   the error handling callback for strict encoding will be returned. */
PyObject *PyCodec_LookupError(const char *name)
{
    PyObject *handler = NULL;

    PyInterpreterState *interp = PyThreadState_GET()->interp;
    if (interp->codec_search_path == NULL && _PyCodecRegistry_Init())
        return NULL;

    if (name==NULL)
        name = "strict";
    handler = PyDict_GetItemString(interp->codec_error_registry, name);
    if (!handler)
        PyErr_Format(PyExc_LookupError, "unknown error handler name '%.400s'", name);
    else
        Py_INCREF(handler);
    return handler;
}

static void wrong_exception_type(PyObject *exc)
{
    _Py_IDENTIFIER(__class__);
    _Py_IDENTIFIER(__name__);
    PyObject *type = _PyObject_GetAttrId(exc, &PyId___class__);
    if (type != NULL) {
        PyObject *name = _PyObject_GetAttrId(type, &PyId___name__);
        Py_DECREF(type);
        if (name != NULL) {
            PyErr_Format(PyExc_TypeError,
                         "don't know how to handle %S in error callback", name);
            Py_DECREF(name);
        }
    }
}

PyObject *PyCodec_StrictErrors(PyObject *exc)
{
    if (PyExceptionInstance_Check(exc))
        PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
    else
        PyErr_SetString(PyExc_TypeError, "codec must pass exception instance");
    return NULL;
}


PyObject *PyCodec_IgnoreErrors(PyObject *exc)
{
    Py_ssize_t end;
    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeDecodeError)) {
        if (PyUnicodeDecodeError_GetEnd(exc, &end))
            return NULL;
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeTranslateError)) {
        if (PyUnicodeTranslateError_GetEnd(exc, &end))
            return NULL;
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
    return Py_BuildValue("(Nn)", PyUnicode_New(0, 0), end);
}


PyObject *PyCodec_ReplaceErrors(PyObject *exc)
{
    Py_ssize_t start, end, i, len;

    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        PyObject *res;
        int kind;
        void *data;
        if (PyUnicodeEncodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
        len = end - start;
        res = PyUnicode_New(len, '?');
        if (res == NULL)
            return NULL;
        kind = PyUnicode_KIND(res);
        data = PyUnicode_DATA(res);
        for (i = 0; i < len; ++i)
            PyUnicode_WRITE(kind, data, i, '?');
        assert(_PyUnicode_CheckConsistency(res, 1));
        return Py_BuildValue("(Nn)", res, end);
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeDecodeError)) {
        if (PyUnicodeDecodeError_GetEnd(exc, &end))
            return NULL;
        return Py_BuildValue("(Cn)",
                             (int)Py_UNICODE_REPLACEMENT_CHARACTER,
                             end);
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeTranslateError)) {
        PyObject *res;
        int kind;
        void *data;
        if (PyUnicodeTranslateError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeTranslateError_GetEnd(exc, &end))
            return NULL;
        len = end - start;
        res = PyUnicode_New(len, Py_UNICODE_REPLACEMENT_CHARACTER);
        if (res == NULL)
            return NULL;
        kind = PyUnicode_KIND(res);
        data = PyUnicode_DATA(res);
        for (i=0; i < len; i++)
            PyUnicode_WRITE(kind, data, i, Py_UNICODE_REPLACEMENT_CHARACTER);
        assert(_PyUnicode_CheckConsistency(res, 1));
        return Py_BuildValue("(Nn)", res, end);
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
}

PyObject *PyCodec_XMLCharRefReplaceErrors(PyObject *exc)
{
    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        PyObject *restuple;
        PyObject *object;
        Py_ssize_t i;
        Py_ssize_t start;
        Py_ssize_t end;
        PyObject *res;
        unsigned char *outp;
        int ressize;
        Py_UCS4 ch;
        if (PyUnicodeEncodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
        if (!(object = PyUnicodeEncodeError_GetObject(exc)))
            return NULL;
        for (i = start, ressize = 0; i < end; ++i) {
            /* object is guaranteed to be "ready" */
            ch = PyUnicode_READ_CHAR(object, i);
            if (ch<10)
                ressize += 2+1+1;
            else if (ch<100)
                ressize += 2+2+1;
            else if (ch<1000)
                ressize += 2+3+1;
            else if (ch<10000)
                ressize += 2+4+1;
            else if (ch<100000)
                ressize += 2+5+1;
            else if (ch<1000000)
                ressize += 2+6+1;
            else
                ressize += 2+7+1;
        }
        /* allocate replacement */
        res = PyUnicode_New(ressize, 127);
        if (res == NULL) {
            Py_DECREF(object);
            return NULL;
        }
        outp = PyUnicode_1BYTE_DATA(res);
        /* generate replacement */
        for (i = start; i < end; ++i) {
            int digits;
            int base;
            ch = PyUnicode_READ_CHAR(object, i);
            *outp++ = '&';
            *outp++ = '#';
            if (ch<10) {
                digits = 1;
                base = 1;
            }
            else if (ch<100) {
                digits = 2;
                base = 10;
            }
            else if (ch<1000) {
                digits = 3;
                base = 100;
            }
            else if (ch<10000) {
                digits = 4;
                base = 1000;
            }
            else if (ch<100000) {
                digits = 5;
                base = 10000;
            }
            else if (ch<1000000) {
                digits = 6;
                base = 100000;
            }
            else {
                digits = 7;
                base = 1000000;
            }
            while (digits-->0) {
                *outp++ = '0' + ch/base;
                ch %= base;
                base /= 10;
            }
            *outp++ = ';';
        }
        assert(_PyUnicode_CheckConsistency(res, 1));
        restuple = Py_BuildValue("(Nn)", res, end);
        Py_DECREF(object);
        return restuple;
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
}

PyObject *PyCodec_BackslashReplaceErrors(PyObject *exc)
{
    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        PyObject *restuple;
        PyObject *object;
        Py_ssize_t i;
        Py_ssize_t start;
        Py_ssize_t end;
        PyObject *res;
        unsigned char *outp;
        int ressize;
        Py_UCS4 c;
        if (PyUnicodeEncodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
        if (!(object = PyUnicodeEncodeError_GetObject(exc)))
            return NULL;
        for (i = start, ressize = 0; i < end; ++i) {
            /* object is guaranteed to be "ready" */
            c = PyUnicode_READ_CHAR(object, i);
            if (c >= 0x10000) {
                ressize += 1+1+8;
            }
            else if (c >= 0x100) {
                ressize += 1+1+4;
            }
            else
                ressize += 1+1+2;
        }
        res = PyUnicode_New(ressize, 127);
        if (res==NULL)
            return NULL;
        for (i = start, outp = PyUnicode_1BYTE_DATA(res);
            i < end; ++i) {
            c = PyUnicode_READ_CHAR(object, i);
            *outp++ = '\\';
            if (c >= 0x00010000) {
                *outp++ = 'U';
                *outp++ = Py_hexdigits[(c>>28)&0xf];
                *outp++ = Py_hexdigits[(c>>24)&0xf];
                *outp++ = Py_hexdigits[(c>>20)&0xf];
                *outp++ = Py_hexdigits[(c>>16)&0xf];
                *outp++ = Py_hexdigits[(c>>12)&0xf];
                *outp++ = Py_hexdigits[(c>>8)&0xf];
            }
            else if (c >= 0x100) {
                *outp++ = 'u';
                *outp++ = Py_hexdigits[(c>>12)&0xf];
                *outp++ = Py_hexdigits[(c>>8)&0xf];
            }
            else
                *outp++ = 'x';
            *outp++ = Py_hexdigits[(c>>4)&0xf];
            *outp++ = Py_hexdigits[c&0xf];
        }

        assert(_PyUnicode_CheckConsistency(res, 1));
        restuple = Py_BuildValue("(Nn)", res, end);
        Py_DECREF(object);
        return restuple;
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
}

/* This handler is declared static until someone demonstrates
   a need to call it directly. */
static PyObject *
PyCodec_SurrogatePassErrors(PyObject *exc)
{
    PyObject *restuple;
    PyObject *object;
    Py_ssize_t i;
    Py_ssize_t start;
    Py_ssize_t end;
    PyObject *res;
    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        char *outp;
        if (PyUnicodeEncodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
        if (!(object = PyUnicodeEncodeError_GetObject(exc)))
            return NULL;
        res = PyBytes_FromStringAndSize(NULL, 3*(end-start));
        if (!res) {
            Py_DECREF(object);
            return NULL;
        }
        outp = PyBytes_AsString(res);
        for (i = start; i < end; i++) {
            /* object is guaranteed to be "ready" */
            Py_UCS4 ch = PyUnicode_READ_CHAR(object, i);
            if (!Py_UNICODE_IS_SURROGATE(ch)) {
                /* Not a surrogate, fail with original exception */
                PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
                Py_DECREF(res);
                Py_DECREF(object);
                return NULL;
            }
            *outp++ = (char)(0xe0 | (ch >> 12));
            *outp++ = (char)(0x80 | ((ch >> 6) & 0x3f));
            *outp++ = (char)(0x80 | (ch & 0x3f));
        }
        restuple = Py_BuildValue("(On)", res, end);
        Py_DECREF(res);
        Py_DECREF(object);
        return restuple;
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeDecodeError)) {
        unsigned char *p;
        Py_UCS4 ch = 0;
        if (PyUnicodeDecodeError_GetStart(exc, &start))
            return NULL;
        if (!(object = PyUnicodeDecodeError_GetObject(exc)))
            return NULL;
        if (!(p = (unsigned char*)PyBytes_AsString(object))) {
            Py_DECREF(object);
            return NULL;
        }
        /* Try decoding a single surrogate character. If
           there are more, let the codec call us again. */
        p += start;
        if (PyBytes_GET_SIZE(object) - start >= 3 &&
            (p[0] & 0xf0) == 0xe0 &&
            (p[1] & 0xc0) == 0x80 &&
            (p[2] & 0xc0) == 0x80) {
            /* it's a three-byte code */
            ch = ((p[0] & 0x0f) << 12) + ((p[1] & 0x3f) << 6) + (p[2] & 0x3f);
            if (!Py_UNICODE_IS_SURROGATE(ch))
                /* it's not a surrogate - fail */
                ch = 0;
        }
        Py_DECREF(object);
        if (ch == 0) {
            PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
            return NULL;
        }
        res = PyUnicode_FromOrdinal(ch);
        if (res == NULL)
            return NULL;
        return Py_BuildValue("(Nn)", res, start+3);
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
}

static PyObject *
PyCodec_SurrogateEscapeErrors(PyObject *exc)
{
    PyObject *restuple;
    PyObject *object;
    Py_ssize_t i;
    Py_ssize_t start;
    Py_ssize_t end;
    PyObject *res;
    if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
        char *outp;
        if (PyUnicodeEncodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeEncodeError_GetEnd(exc, &end))
            return NULL;
        if (!(object = PyUnicodeEncodeError_GetObject(exc)))
            return NULL;
        res = PyBytes_FromStringAndSize(NULL, end-start);
        if (!res) {
            Py_DECREF(object);
            return NULL;
        }
        outp = PyBytes_AsString(res);
        for (i = start; i < end; i++) {
            /* object is guaranteed to be "ready" */
            Py_UCS4 ch = PyUnicode_READ_CHAR(object, i);
            if (ch < 0xdc80 || ch > 0xdcff) {
                /* Not a UTF-8b surrogate, fail with original exception */
                PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
                Py_DECREF(res);
                Py_DECREF(object);
                return NULL;
            }
            *outp++ = ch - 0xdc00;
        }
        restuple = Py_BuildValue("(On)", res, end);
        Py_DECREF(res);
        Py_DECREF(object);
        return restuple;
    }
    else if (PyObject_IsInstance(exc, PyExc_UnicodeDecodeError)) {
        PyObject *str;
        unsigned char *p;
        Py_UCS2 ch[4]; /* decode up to 4 bad bytes. */
        int consumed = 0;
        if (PyUnicodeDecodeError_GetStart(exc, &start))
            return NULL;
        if (PyUnicodeDecodeError_GetEnd(exc, &end))
            return NULL;
        if (!(object = PyUnicodeDecodeError_GetObject(exc)))
            return NULL;
        if (!(p = (unsigned char*)PyBytes_AsString(object))) {
            Py_DECREF(object);
            return NULL;
        }
        while (consumed < 4 && consumed < end-start) {
            /* Refuse to escape ASCII bytes. */
            if (p[start+consumed] < 128)
                break;
            ch[consumed] = 0xdc00 + p[start+consumed];
            consumed++;
        }
        Py_DECREF(object);
        if (!consumed) {
            /* codec complained about ASCII byte. */
            PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
            return NULL;
        }
        str = PyUnicode_FromKindAndData(PyUnicode_2BYTE_KIND, ch, consumed);
        if (str == NULL)
            return NULL;
        return Py_BuildValue("(Nn)", str, start+consumed);
    }
    else {
        wrong_exception_type(exc);
        return NULL;
    }
}


static PyObject *strict_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_StrictErrors(exc);
}


static PyObject *ignore_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_IgnoreErrors(exc);
}


static PyObject *replace_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_ReplaceErrors(exc);
}


static PyObject *xmlcharrefreplace_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_XMLCharRefReplaceErrors(exc);
}


static PyObject *backslashreplace_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_BackslashReplaceErrors(exc);
}

static PyObject *surrogatepass_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_SurrogatePassErrors(exc);
}

static PyObject *surrogateescape_errors(PyObject *self, PyObject *exc)
{
    return PyCodec_SurrogateEscapeErrors(exc);
}

static int _PyCodecRegistry_Init(void)
{
    static struct {
        char *name;
        PyMethodDef def;
    } methods[] =
    {
        {
            "strict",
            {
                "strict_errors",
                strict_errors,
                METH_O,
                PyDoc_STR("Implements the 'strict' error handling, which "
                          "raises a UnicodeError on coding errors.")
            }
        },
        {
            "ignore",
            {
                "ignore_errors",
                ignore_errors,
                METH_O,
                PyDoc_STR("Implements the 'ignore' error handling, which "
                          "ignores malformed data and continues.")
            }
        },
        {
            "replace",
            {
                "replace_errors",
                replace_errors,
                METH_O,
                PyDoc_STR("Implements the 'replace' error handling, which "
                          "replaces malformed data with a replacement marker.")
            }
        },
        {
            "xmlcharrefreplace",
            {
                "xmlcharrefreplace_errors",
                xmlcharrefreplace_errors,
                METH_O,
                PyDoc_STR("Implements the 'xmlcharrefreplace' error handling, "
                          "which replaces an unencodable character with the "
                          "appropriate XML character reference.")
            }
        },
        {
            "backslashreplace",
            {
                "backslashreplace_errors",
                backslashreplace_errors,
                METH_O,
                PyDoc_STR("Implements the 'backslashreplace' error handling, "
                          "which replaces an unencodable character with a "
                          "backslashed escape sequence.")
            }
        },
        {
            "surrogatepass",
            {
                "surrogatepass",
                surrogatepass_errors,
                METH_O
            }
        },
        {
            "surrogateescape",
            {
                "surrogateescape",
                surrogateescape_errors,
                METH_O
            }
        }
    };

    PyInterpreterState *interp = PyThreadState_GET()->interp;
    PyObject *mod;
    unsigned i;

    if (interp->codec_search_path != NULL)
        return 0;

    interp->codec_search_path = PyList_New(0);
    interp->codec_search_cache = PyDict_New();
    interp->codec_error_registry = PyDict_New();

    if (interp->codec_error_registry) {
        for (i = 0; i < Py_ARRAY_LENGTH(methods); ++i) {
            PyObject *func = PyCFunction_NewEx(&methods[i].def, NULL, NULL);
            int res;
            if (!func)
                Py_FatalError("can't initialize codec error registry");
            res = PyCodec_RegisterError(methods[i].name, func);
            Py_DECREF(func);
            if (res)
                Py_FatalError("can't initialize codec error registry");
        }
    }

    if (interp->codec_search_path == NULL ||
        interp->codec_search_cache == NULL ||
        interp->codec_error_registry == NULL)
        Py_FatalError("can't initialize codec registry");

    mod = PyImport_ImportModuleNoBlock("encodings");
    if (mod == NULL) {
        return -1;
    }
    Py_DECREF(mod);
    interp->codecs_initialized = 1;
    return 0;
}