cpython/Python/codecs.c

436 lines
10 KiB
C
Raw Normal View History

/* ------------------------------------------------------------------------
Python Codec Registry and support functions
Written by Marc-Andre Lemburg (mal@lemburg.com).
(c) Copyright CNRI, All Rights Reserved. NO WARRANTY.
------------------------------------------------------------------------ */
#include "Python.h"
#include <ctype.h>
/* --- Globals ------------------------------------------------------------ */
static PyObject *_PyCodec_SearchPath;
static PyObject *_PyCodec_SearchCache;
/* Flag used for lazy import of the standard encodings package */
static int import_encodings_called = 0;
/* --- 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 import_encodings()
{
PyObject *mod;
import_encodings_called = 1;
mod = PyImport_ImportModule("encodings");
if (mod == NULL) {
if (PyErr_ExceptionMatches(PyExc_ImportError)) {
/* Ignore ImportErrors... this is done so that
distributions can disable the encodings package. Note
that other errors are not masked, e.g. SystemErrors
raised to inform the user of an error in the Python
configuration are still reported back to the user. */
PyErr_Clear();
return 0;
}
return -1;
}
Py_DECREF(mod);
return 0;
}
/* Register a new codec search function.
As side effect, this tries to load the encodings package, if not
yet done, to make sure that it is always first in the list of
search functions.
The search_function's refcount is incremented by this function. */
int PyCodec_Register(PyObject *search_function)
{
if (!import_encodings_called) {
if (import_encodings())
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(_PyCodec_SearchPath, search_function);
onError:
return -1;
}
Marc-Andre's third try at this bulk patch seems to work (except that his copy of test_contains.py seems to be broken -- the lines he deleted were already absent). Checkin messages: New Unicode support for int(), float(), complex() and long(). - new APIs PyInt_FromUnicode() and PyLong_FromUnicode() - added support for Unicode to PyFloat_FromString() - new encoding API PyUnicode_EncodeDecimal() which converts Unicode to a decimal char* string (used in the above new APIs) - shortcuts for calls like int(<int object>) and float(<float obj>) - tests for all of the above Unicode compares and contains checks: - comparing Unicode and non-string types now works; TypeErrors are masked, all other errors such as ValueError during Unicode coercion are passed through (note that PyUnicode_Compare does not implement the masking -- PyObject_Compare does this) - contains now works for non-string types too; TypeErrors are masked and 0 returned; all other errors are passed through Better testing support for the standard codecs. Misc minor enhancements, such as an alias dbcs for the mbcs codec. Changes: - PyLong_FromString() now applies the same error checks as does PyInt_FromString(): trailing garbage is reported as error and not longer silently ignored. The only characters which may be trailing the digits are 'L' and 'l' -- these are still silently ignored. - string.ato?() now directly interface to int(), long() and float(). The error strings are now a little different, but the type still remains the same. These functions are now ready to get declared obsolete ;-) - PyNumber_Int() now also does a check for embedded NULL chars in the input string; PyNumber_Long() already did this (and still does) Followed by: Looks like I've gone a step too far there... (and test_contains.py seem to have a bug too). I've changed back to reporting all errors in PyUnicode_Contains() and added a few more test cases to test_contains.py (plus corrected the join() NameError).
2000-04-05 17:11:21 -03:00
/* Convert a string to a normalized Python string: all characters are
converted to lower case, spaces are replaced with underscores. */
static
Marc-Andre's third try at this bulk patch seems to work (except that his copy of test_contains.py seems to be broken -- the lines he deleted were already absent). Checkin messages: New Unicode support for int(), float(), complex() and long(). - new APIs PyInt_FromUnicode() and PyLong_FromUnicode() - added support for Unicode to PyFloat_FromString() - new encoding API PyUnicode_EncodeDecimal() which converts Unicode to a decimal char* string (used in the above new APIs) - shortcuts for calls like int(<int object>) and float(<float obj>) - tests for all of the above Unicode compares and contains checks: - comparing Unicode and non-string types now works; TypeErrors are masked, all other errors such as ValueError during Unicode coercion are passed through (note that PyUnicode_Compare does not implement the masking -- PyObject_Compare does this) - contains now works for non-string types too; TypeErrors are masked and 0 returned; all other errors are passed through Better testing support for the standard codecs. Misc minor enhancements, such as an alias dbcs for the mbcs codec. Changes: - PyLong_FromString() now applies the same error checks as does PyInt_FromString(): trailing garbage is reported as error and not longer silently ignored. The only characters which may be trailing the digits are 'L' and 'l' -- these are still silently ignored. - string.ato?() now directly interface to int(), long() and float(). The error strings are now a little different, but the type still remains the same. These functions are now ready to get declared obsolete ;-) - PyNumber_Int() now also does a check for embedded NULL chars in the input string; PyNumber_Long() already did this (and still does) Followed by: Looks like I've gone a step too far there... (and test_contains.py seem to have a bug too). I've changed back to reporting all errors in PyUnicode_Contains() and added a few more test cases to test_contains.py (plus corrected the join() NameError).
2000-04-05 17:11:21 -03:00
PyObject *normalizestring(const char *string)
{
register int i;
int len = strlen(string);
char *p;
PyObject *v;
v = PyString_FromStringAndSize(NULL, len);
if (v == NULL)
return NULL;
p = PyString_AS_STRING(v);
Marc-Andre's third try at this bulk patch seems to work (except that his copy of test_contains.py seems to be broken -- the lines he deleted were already absent). Checkin messages: New Unicode support for int(), float(), complex() and long(). - new APIs PyInt_FromUnicode() and PyLong_FromUnicode() - added support for Unicode to PyFloat_FromString() - new encoding API PyUnicode_EncodeDecimal() which converts Unicode to a decimal char* string (used in the above new APIs) - shortcuts for calls like int(<int object>) and float(<float obj>) - tests for all of the above Unicode compares and contains checks: - comparing Unicode and non-string types now works; TypeErrors are masked, all other errors such as ValueError during Unicode coercion are passed through (note that PyUnicode_Compare does not implement the masking -- PyObject_Compare does this) - contains now works for non-string types too; TypeErrors are masked and 0 returned; all other errors are passed through Better testing support for the standard codecs. Misc minor enhancements, such as an alias dbcs for the mbcs codec. Changes: - PyLong_FromString() now applies the same error checks as does PyInt_FromString(): trailing garbage is reported as error and not longer silently ignored. The only characters which may be trailing the digits are 'L' and 'l' -- these are still silently ignored. - string.ato?() now directly interface to int(), long() and float(). The error strings are now a little different, but the type still remains the same. These functions are now ready to get declared obsolete ;-) - PyNumber_Int() now also does a check for embedded NULL chars in the input string; PyNumber_Long() already did this (and still does) Followed by: Looks like I've gone a step too far there... (and test_contains.py seem to have a bug too). I've changed back to reporting all errors in PyUnicode_Contains() and added a few more test cases to test_contains.py (plus corrected the join() NameError).
2000-04-05 17:11:21 -03:00
for (i = 0; i < len; i++) {
register char ch = string[i];
if (ch == ' ')
ch = '-';
else
ch = tolower(ch);
p[i] = ch;
}
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 KeyError 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)
{
PyObject *result, *args = NULL, *v;
int i, len;
if (_PyCodec_SearchCache == NULL ||
_PyCodec_SearchPath == NULL) {
PyErr_SetString(PyExc_SystemError,
"codec module not properly initialized");
goto onError;
}
if (!import_encodings_called) {
if (import_encodings())
goto onError;
}
Marc-Andre's third try at this bulk patch seems to work (except that his copy of test_contains.py seems to be broken -- the lines he deleted were already absent). Checkin messages: New Unicode support for int(), float(), complex() and long(). - new APIs PyInt_FromUnicode() and PyLong_FromUnicode() - added support for Unicode to PyFloat_FromString() - new encoding API PyUnicode_EncodeDecimal() which converts Unicode to a decimal char* string (used in the above new APIs) - shortcuts for calls like int(<int object>) and float(<float obj>) - tests for all of the above Unicode compares and contains checks: - comparing Unicode and non-string types now works; TypeErrors are masked, all other errors such as ValueError during Unicode coercion are passed through (note that PyUnicode_Compare does not implement the masking -- PyObject_Compare does this) - contains now works for non-string types too; TypeErrors are masked and 0 returned; all other errors are passed through Better testing support for the standard codecs. Misc minor enhancements, such as an alias dbcs for the mbcs codec. Changes: - PyLong_FromString() now applies the same error checks as does PyInt_FromString(): trailing garbage is reported as error and not longer silently ignored. The only characters which may be trailing the digits are 'L' and 'l' -- these are still silently ignored. - string.ato?() now directly interface to int(), long() and float(). The error strings are now a little different, but the type still remains the same. These functions are now ready to get declared obsolete ;-) - PyNumber_Int() now also does a check for embedded NULL chars in the input string; PyNumber_Long() already did this (and still does) Followed by: Looks like I've gone a step too far there... (and test_contains.py seem to have a bug too). I've changed back to reporting all errors in PyUnicode_Contains() and added a few more test cases to test_contains.py (plus corrected the join() NameError).
2000-04-05 17:11:21 -03:00
/* Convert the encoding to a normalized Python string: all
characters are converted to lower case, spaces and hypens are
replaced with underscores. */
v = normalizestring(encoding);
if (v == NULL)
goto onError;
PyString_InternInPlace(&v);
/* First, try to lookup the name in the registry dictionary */
result = PyDict_GetItem(_PyCodec_SearchCache, 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(_PyCodec_SearchPath);
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(_PyCodec_SearchPath, 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_SetString(PyExc_LookupError,
"unknown encoding");
goto onError;
}
/* Cache and return the result */
PyDict_SetItem(_PyCodec_SearchCache, v, result);
Py_DECREF(args);
return result;
onError:
Py_XDECREF(args);
return NULL;
}
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 = PyString_FromString(errors);
if (v == NULL) {
Py_DECREF(args);
return NULL;
}
PyTuple_SET_ITEM(args, 1, v);
}
return args;
}
/* Build a codec by calling factory(stream[,errors]) or just
factory(errors) depending on whether the given parameters are
non-NULL. */
static
PyObject *build_stream_codec(PyObject *factory,
PyObject *stream,
const char *errors)
{
PyObject *args, *codec;
args = args_tuple(stream, errors);
if (args == NULL)
return NULL;
codec = PyEval_CallObject(factory, args);
Py_DECREF(args);
return codec;
}
/* Convenience APIs to query the Codec registry.
All APIs return a codec object with incremented refcount.
*/
PyObject *PyCodec_Encoder(const char *encoding)
{
PyObject *codecs;
PyObject *v;
codecs = _PyCodec_Lookup(encoding);
if (codecs == NULL)
goto onError;
v = PyTuple_GET_ITEM(codecs,0);
Py_INCREF(v);
return v;
onError:
return NULL;
}
PyObject *PyCodec_Decoder(const char *encoding)
{
PyObject *codecs;
PyObject *v;
codecs = _PyCodec_Lookup(encoding);
if (codecs == NULL)
goto onError;
v = PyTuple_GET_ITEM(codecs,1);
Py_INCREF(v);
return v;
onError:
return NULL;
}
PyObject *PyCodec_StreamReader(const char *encoding,
PyObject *stream,
const char *errors)
{
PyObject *codecs;
codecs = _PyCodec_Lookup(encoding);
if (codecs == NULL)
goto onError;
return build_stream_codec(PyTuple_GET_ITEM(codecs,2),stream,errors);
onError:
return NULL;
}
PyObject *PyCodec_StreamWriter(const char *encoding,
PyObject *stream,
const char *errors)
{
PyObject *codecs;
codecs = _PyCodec_Lookup(encoding);
if (codecs == NULL)
goto onError;
return build_stream_codec(PyTuple_GET_ITEM(codecs,3),stream,errors);
onError:
return NULL;
}
/* 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;
PyObject *v;
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)
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(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)
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;
}
void _PyCodecRegistry_Init()
{
if (_PyCodec_SearchPath == NULL)
_PyCodec_SearchPath = PyList_New(0);
if (_PyCodec_SearchCache == NULL)
_PyCodec_SearchCache = PyDict_New();
if (_PyCodec_SearchPath == NULL ||
_PyCodec_SearchCache == NULL)
Py_FatalError("can't intialize codec registry");
}
void _PyCodecRegistry_Fini()
{
Py_XDECREF(_PyCodec_SearchPath);
_PyCodec_SearchPath = NULL;
Py_XDECREF(_PyCodec_SearchCache);
_PyCodec_SearchCache = NULL;
}