summaryrefslogtreecommitdiffstats
path: root/Python/codecs.c
diff options
context:
space:
mode:
Diffstat (limited to 'Python/codecs.c')
-rw-r--r--Python/codecs.c233
1 files changed, 206 insertions, 27 deletions
diff --git a/Python/codecs.c b/Python/codecs.c
index ea33c49..6849c0f 100644
--- a/Python/codecs.c
+++ b/Python/codecs.c
@@ -53,7 +53,7 @@ int PyCodec_Register(PyObject *search_function)
static
PyObject *normalizestring(const char *string)
{
- register size_t i;
+ size_t i;
size_t len = strlen(string);
char *p;
PyObject *v;
@@ -65,9 +65,9 @@ PyObject *normalizestring(const char *string)
p = PyMem_Malloc(len + 1);
if (p == NULL)
- return NULL;
+ return PyErr_NoMemory();
for (i = 0; i < len; i++) {
- register char ch = string[i];
+ char ch = string[i];
if (ch == ' ')
ch = '-';
else
@@ -185,6 +185,32 @@ PyObject *_PyCodec_Lookup(const char *encoding)
return NULL;
}
+int _PyCodec_Forget(const char *encoding)
+{
+ PyInterpreterState *interp;
+ PyObject *v;
+ int result;
+
+ interp = PyThreadState_GET()->interp;
+ if (interp->codec_search_path == NULL) {
+ return -1;
+ }
+
+ /* 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) {
+ return -1;
+ }
+
+ /* Drop the named codec from the internal cache */
+ result = PyDict_DelItem(interp->codec_search_cache, v);
+ Py_DECREF(v);
+
+ return result;
+}
+
/* Codec registry encoding check API. */
int PyCodec_KnownEncoding(const char *encoding)
@@ -360,6 +386,22 @@ PyObject *PyCodec_StreamWriter(const char *encoding,
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).
@@ -379,8 +421,10 @@ _PyCodec_EncodeInternal(PyObject *object,
goto onError;
result = PyEval_CallObject(encoder, args);
- if (result == NULL)
+ if (result == NULL) {
+ wrap_codec_error("encoding", encoding);
goto onError;
+ }
if (!PyTuple_Check(result) ||
PyTuple_GET_SIZE(result) != 2) {
@@ -423,8 +467,10 @@ _PyCodec_DecodeInternal(PyObject *object,
goto onError;
result = PyEval_CallObject(decoder,args);
- if (result == NULL)
+ if (result == NULL) {
+ wrap_codec_error("decoding", encoding);
goto onError;
+ }
if (!PyTuple_Check(result) ||
PyTuple_GET_SIZE(result) != 2) {
PyErr_SetString(PyExc_TypeError,
@@ -589,7 +635,7 @@ int PyCodec_RegisterError(const char *name, PyObject *error)
return -1;
}
return PyDict_SetItemString(interp->codec_error_registry,
- (char *)name, error);
+ name, error);
}
/* Lookup the error handling callback function registered under the
@@ -605,7 +651,7 @@ PyObject *PyCodec_LookupError(const char *name)
if (name==NULL)
name = "strict";
- handler = PyDict_GetItemString(interp->codec_error_registry, (char *)name);
+ handler = PyDict_GetItemString(interp->codec_error_registry, name);
if (!handler)
PyErr_Format(PyExc_LookupError, "unknown error handler name '%.400s'", name);
else
@@ -848,8 +894,10 @@ PyObject *PyCodec_BackslashReplaceErrors(PyObject *exc)
ressize += 1+1+2;
}
res = PyUnicode_New(ressize, 127);
- if (res==NULL)
+ if (res == NULL) {
+ Py_DECREF(object);
return NULL;
+ }
for (i = start, outp = PyUnicode_1BYTE_DATA(res);
i < end; ++i) {
c = PyUnicode_READ_CHAR(object, i);
@@ -885,6 +933,65 @@ PyObject *PyCodec_BackslashReplaceErrors(PyObject *exc)
}
}
+#define ENC_UTF8 0
+#define ENC_UTF16BE 1
+#define ENC_UTF16LE 2
+#define ENC_UTF32BE 3
+#define ENC_UTF32LE 4
+
+static int
+get_standard_encoding(const char *encoding, int *bytelength)
+{
+ if (Py_TOLOWER(encoding[0]) == 'u' &&
+ Py_TOLOWER(encoding[1]) == 't' &&
+ Py_TOLOWER(encoding[2]) == 'f') {
+ encoding += 3;
+ if (*encoding == '-' || *encoding == '_' )
+ encoding++;
+ if (encoding[0] == '1' && encoding[1] == '6') {
+ encoding += 2;
+ *bytelength = 2;
+ if (*encoding == '\0') {
+#ifdef WORDS_BIGENDIAN
+ return ENC_UTF16BE;
+#else
+ return ENC_UTF16LE;
+#endif
+ }
+ if (*encoding == '-' || *encoding == '_' )
+ encoding++;
+ if (Py_TOLOWER(encoding[1]) == 'e' && encoding[2] == '\0') {
+ if (Py_TOLOWER(encoding[0]) == 'b')
+ return ENC_UTF16BE;
+ if (Py_TOLOWER(encoding[0]) == 'l')
+ return ENC_UTF16LE;
+ }
+ }
+ else if (encoding[0] == '3' && encoding[1] == '2') {
+ encoding += 2;
+ *bytelength = 4;
+ if (*encoding == '\0') {
+#ifdef WORDS_BIGENDIAN
+ return ENC_UTF32BE;
+#else
+ return ENC_UTF32LE;
+#endif
+ }
+ if (*encoding == '-' || *encoding == '_' )
+ encoding++;
+ if (Py_TOLOWER(encoding[1]) == 'e' && encoding[2] == '\0') {
+ if (Py_TOLOWER(encoding[0]) == 'b')
+ return ENC_UTF32BE;
+ if (Py_TOLOWER(encoding[0]) == 'l')
+ return ENC_UTF32LE;
+ }
+ }
+ }
+ /* utf-8 */
+ *bytelength = 3;
+ return ENC_UTF8;
+}
+
/* This handler is declared static until someone demonstrates
a need to call it directly. */
static PyObject *
@@ -892,37 +999,79 @@ PyCodec_SurrogatePassErrors(PyObject *exc)
{
PyObject *restuple;
PyObject *object;
+ PyObject *encode;
+ char *encoding;
+ int code;
+ int bytelength;
Py_ssize_t i;
Py_ssize_t start;
Py_ssize_t end;
PyObject *res;
if (PyObject_IsInstance(exc, PyExc_UnicodeEncodeError)) {
- char *outp;
+ unsigned 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 (!(encode = PyUnicodeEncodeError_GetEncoding(exc))) {
+ Py_DECREF(object);
+ return NULL;
+ }
+ if (!(encoding = PyUnicode_AsUTF8(encode))) {
+ Py_DECREF(object);
+ Py_DECREF(encode);
+ return NULL;
+ }
+ code = get_standard_encoding(encoding, &bytelength);
+ Py_DECREF(encode);
+
+ if (end - start > PY_SSIZE_T_MAX / bytelength)
+ end = start + PY_SSIZE_T_MAX / bytelength;
+ res = PyBytes_FromStringAndSize(NULL, bytelength*(end-start));
if (!res) {
Py_DECREF(object);
return NULL;
}
- outp = PyBytes_AsString(res);
+ outp = (unsigned char*)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 < 0xd800 || ch > 0xdfff) {
+ 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));
+ switch (code) {
+ case ENC_UTF8:
+ *outp++ = (unsigned char)(0xe0 | (ch >> 12));
+ *outp++ = (unsigned char)(0x80 | ((ch >> 6) & 0x3f));
+ *outp++ = (unsigned char)(0x80 | (ch & 0x3f));
+ break;
+ case ENC_UTF16LE:
+ *outp++ = (unsigned char) ch;
+ *outp++ = (unsigned char)(ch >> 8);
+ break;
+ case ENC_UTF16BE:
+ *outp++ = (unsigned char)(ch >> 8);
+ *outp++ = (unsigned char) ch;
+ break;
+ case ENC_UTF32LE:
+ *outp++ = (unsigned char) ch;
+ *outp++ = (unsigned char)(ch >> 8);
+ *outp++ = (unsigned char)(ch >> 16);
+ *outp++ = (unsigned char)(ch >> 24);
+ break;
+ case ENC_UTF32BE:
+ *outp++ = (unsigned char)(ch >> 24);
+ *outp++ = (unsigned char)(ch >> 16);
+ *outp++ = (unsigned char)(ch >> 8);
+ *outp++ = (unsigned char) ch;
+ break;
+ }
}
restuple = Py_BuildValue("(On)", res, end);
Py_DECREF(res);
@@ -934,34 +1083,64 @@ PyCodec_SurrogatePassErrors(PyObject *exc)
Py_UCS4 ch = 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;
}
+ if (!(encode = PyUnicodeDecodeError_GetEncoding(exc))) {
+ Py_DECREF(object);
+ return NULL;
+ }
+ if (!(encoding = PyUnicode_AsUTF8(encode))) {
+ Py_DECREF(object);
+ Py_DECREF(encode);
+ return NULL;
+ }
+ code = get_standard_encoding(encoding, &bytelength);
+ Py_DECREF(encode);
+
/* 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 (ch < 0xd800 || ch > 0xdfff)
- /* it's not a surrogate - fail */
- ch = 0;
+ if (PyBytes_GET_SIZE(object) - start >= bytelength) {
+ switch (code) {
+ case ENC_UTF8:
+ if ((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);
+ }
+ break;
+ case ENC_UTF16LE:
+ ch = p[1] << 8 | p[0];
+ break;
+ case ENC_UTF16BE:
+ ch = p[0] << 8 | p[1];
+ break;
+ case ENC_UTF32LE:
+ ch = (p[3] << 24) | (p[2] << 16) | (p[1] << 8) | p[0];
+ break;
+ case ENC_UTF32BE:
+ ch = (p[0] << 24) | (p[1] << 16) | (p[2] << 8) | p[3];
+ break;
+ }
}
+
Py_DECREF(object);
- if (ch == 0) {
+ if (!Py_UNICODE_IS_SURROGATE(ch)) {
+ /* it's not a surrogate - fail */
PyErr_SetObject(PyExceptionInstance_Class(exc), exc);
return NULL;
}
res = PyUnicode_FromOrdinal(ch);
if (res == NULL)
return NULL;
- return Py_BuildValue("(Nn)", res, start+3);
+ return Py_BuildValue("(Nn)", res, start + bytelength);
}
else {
wrong_exception_type(exc);
@@ -1178,7 +1357,7 @@ static int _PyCodecRegistry_Init(void)
if (interp->codec_error_registry) {
for (i = 0; i < Py_ARRAY_LENGTH(methods); ++i) {
- PyObject *func = PyCFunction_New(&methods[i].def, NULL);
+ PyObject *func = PyCFunction_NewEx(&methods[i].def, NULL, NULL);
int res;
if (!func)
Py_FatalError("can't initialize codec error registry");
generated by cgit v0.12 at 2025-10-31 16:13:43 (GMT)