Thread: [KoCo-CVS] [Commit] iconvcodec _iconv_codec.c
Brought to you by:
perky
Menu
▾
▴
koco-cvs
|
From: Hye-Shik C. <pe...@us...> - 2003-04-20 21:40:08
|
perky 03/04/20 14:40:06
Modified: . _iconv_codec.c
Log:
Python/win32 declares Py_UNICODE_USING as null macro
Revision Changes Path
1.4 +2 -2 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.3
retrieving revision 1.4
diff -u -r1.3 -r1.4
--- _iconv_codec.c 20 Apr 2003 20:45:34 -0000 1.3
+++ _iconv_codec.c 20 Apr 2003 21:40:05 -0000 1.4
@@ -24,14 +24,14 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.3 2003/04/20 20:45:34 perky Exp $
+ * $Id: _iconv_codec.c,v 1.4 2003/04/20 21:40:05 perky Exp $
*/
#include "Python.h"
#include <iconv.h>
#include "_iconv_codec_compat.h"
-#if Py_USING_UNICODE
+#ifdef Py_USING_UNICODE
# if Py_UNICODE_SIZE == 2
# define UCS_N "UCS-2"
# define MBENCODED_LENGTH_MAX 4
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 05:55:30
|
perky 03/06/10 22:55:29
Modified: . _iconv_codec.c
Log:
Sigh, it's not cjkcodecs here! :S
Revision Changes Path
1.6 +2 -2 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.5
retrieving revision 1.6
diff -u -r1.5 -r1.6
--- _iconv_codec.c 11 Jun 2003 05:51:11 -0000 1.5
+++ _iconv_codec.c 11 Jun 2003 05:55:28 -0000 1.6
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.5 2003/06/11 05:51:11 perky Exp $
+ * $Id: _iconv_codec.c,v 1.6 2003/06/11 05:55:28 perky Exp $
*/
#include "Python.h"
@@ -1060,7 +1060,7 @@
| (Py_UNICODE)(ubuf[5] ^ 0x80);
ubuf += 6;
} else
- return 1;
+ goto ilseq;
#endif
}
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 11:06:53
|
perky 03/06/11 04:06:50
Modified: . _iconv_codec.c
Log:
Use 'UTF-16' internal encodings when Py_UNICODE_SIZE == 2.
Revision Changes Path
1.7 +10 -6 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.6
retrieving revision 1.7
diff -u -r1.6 -r1.7
--- _iconv_codec.c 11 Jun 2003 05:55:28 -0000 1.6
+++ _iconv_codec.c 11 Jun 2003 11:06:50 -0000 1.7
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.6 2003/06/11 05:55:28 perky Exp $
+ * $Id: _iconv_codec.c,v 1.7 2003/06/11 11:06:50 perky Exp $
*/
#include "Python.h"
@@ -33,11 +33,9 @@
#ifdef Py_USING_UNICODE
# if Py_UNICODE_SIZE == 2
-# define UCS_N "UCS-2"
# define MBENCODED_LENGTH_MAX 4
# define _Py_UNICODE_SWAP(c) (Py_UNICODE)((c)>>8 | (c)<<8)
# elif Py_UNICODE_SIZE == 4
-# define UCS_N "UCS-4"
# define MBENCODED_LENGTH_MAX 6
# define _Py_UNICODE_SWAP(c) (Py_UNICODE)((c)>>24 | \
((c)&0x00ff0000)>>8 | \
@@ -81,9 +79,15 @@
const char *encoding;
uniinternal_type_t type;
} uniinternal_modes[] = {
- {UCS_N "-INTERNAL", UNIINTERNAL_UCS}, /* GNU libiconv, FreeBSD, APR */
- {UCS_N ENDIANSUFX, UNIINTERNAL_UCS}, /* SunOS */
- {UCS_N, UNIINTERNAL_UCS}, /* GLIBC */
+#if Py_UNICODE_SIZE == 2
+/* Py_UNICODE* may contain surrogate characters */
+ {"UTF-16" ENDIANSUFX, UNIINTERNAL_UCS},
+ {"UTF16" ENDIANSUFX, UNIINTERNAL_UCS},
+#else
+ {"UCS-4-INTERNAL", UNIINTERNAL_UCS}, /* GNU libiconv, FreeBSD, APR */
+ {"UCS-4" ENDIANSUFX, UNIINTERNAL_UCS}, /* SunOS */
+ {"UCS-4", UNIINTERNAL_UCS}, /* GLIBC */
+#endif
{"UTF-8", UNIINTERNAL_UTF_8}, /* SunOS(CJK) */
{"\0", UNIINTERNAL_DONTUSE},
};
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 12:01:59
|
perky 03/06/11 05:01:58
Modified: . _iconv_codec.c
Log:
Utilize UCS-2 Surrogate-Pair to support ISO-10646 extended planes
Revision Changes Path
1.8 +67 -37 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.7
retrieving revision 1.8
diff -u -r1.7 -r1.8
--- _iconv_codec.c 11 Jun 2003 11:06:50 -0000 1.7
+++ _iconv_codec.c 11 Jun 2003 12:01:57 -0000 1.8
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.7 2003/06/11 11:06:50 perky Exp $
+ * $Id: _iconv_codec.c,v 1.8 2003/06/11 12:01:57 perky Exp $
*/
#include "Python.h"
@@ -57,6 +57,14 @@
typedef const char **iconv_arg2_t;
#endif
+#ifndef ucs4_t
+# ifdef uint32_t
+typedef uint32_t ucs4_t;
+# else
+typedef unsigned long ucs4_t;
+# endif
+#endif
+
#define ERROR_STRICT (PyObject *)(1)
#define ERROR_IGNORE (PyObject *)(2)
#define ERROR_REPLACE (PyObject *)(3)
@@ -570,22 +578,30 @@
return -1;
buf->rinbuf_top = buf->rinbuf = rinbuf;
for (; buf->inbuf < buf->inbuf_end; buf->inbuf++) {
- Py_UNICODE code = *buf->inbuf;
+ ucs4_t code = *buf->inbuf;
int size;
if (code < 0x80) size = 1;
else if (code < 0x800) size = 2;
+ else {
#if Py_UNICODE_SIZE == 2
- else size = 3; /* XXX put surrogate characters for EMP! */
-#else
- else if (code < 0x10000) size = 3;
- else if (code < 0x200000) size = 4;
- else if (code < 0x4000000) size = 5;
- else size = 6;
-#endif
+ /* Unfold a Surrogate-Pair */
+ if (code >= 0xd800 && code < 0xdc00 &&
+ buf->inbuf+1 < buf->inbuf_end &&
+ buf->inbuf[1] >= 0xdc00 &&
+ buf->inbuf[1] < 0xe000) {
+ code = 0x10000 + ((code - 0xd800) << 10) +
+ (buf->inbuf[1] - 0xdc00);
+ buf->inbuf++;
+ }
+#endif
+ if (code < 0x10000) size = 3;
+ else if (code < 0x200000) size = 4;
+ else if (code < 0x4000000) size = 5;
+ else size = 6;
+ }
switch (size) {
-#if Py_UNICODE_SIZE == 4
case 6:
rinbuf[5] = 0x80 | (code & 0x3f);
code = code >> 6;
@@ -601,7 +617,6 @@
code = code >> 6;
code |= 0x10000;
/* FALLTHROUGH */
-#endif
case 3:
rinbuf[2] = 0x80 | (code & 0x3f);
code = code >> 6;
@@ -1000,10 +1015,11 @@
if (nch > 0)
RESERVE_DECODEBUFFER(buf, nch)
for (ubuf = ubuf_top; ubuf < ubuf_end;) {
- int uleft = (int)(ubuf_end - ubuf);
+ int uleft = (int)(ubuf_end - ubuf);
+ ucs4_t code;
if (*ubuf < 0x80) {
- *buf->outbuf++ = (unsigned char)*ubuf++;
+ code = (unsigned char)*ubuf++;
} else if (*ubuf < 0xc2) {
ilseq: PyErr_SetString(PyExc_RuntimeError,
"iconv returned illegal utf-8 sequence");
@@ -1011,32 +1027,28 @@
} else if (*ubuf < 0xe0) {
if (uleft < 2 || !((ubuf[1] ^ 0x80) < 0x40))
goto ilseq;
- *buf->outbuf++ = ((Py_UNICODE)(ubuf[0] & 0x1f) << 6)
- | (Py_UNICODE)(ubuf[1] ^ 0x80);
+ code = ((Py_UNICODE)(ubuf[0] & 0x1f) << 6)
+ | (Py_UNICODE)(ubuf[1] ^ 0x80);
ubuf += 2;
} else if (*ubuf < 0xf0) {
if (uleft < 3 || !((ubuf[1] ^ 0x80) < 0x40 &&
(ubuf[2] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xe1 || ubuf[1] >= 0xa0)))
goto ilseq;
- *buf->outbuf++ = ((Py_UNICODE)(ubuf[0] & 0x0f) << 12)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[2] ^ 0x80);
+ code = ((Py_UNICODE)(ubuf[0] & 0x0f) << 12)
+ | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 6)
+ | (Py_UNICODE)(ubuf[2] ^ 0x80);
ubuf += 3;
}
-#if Py_UNICODE_SIZE == 2
- else /* XXX: put surrogate characters here! */
- goto ilseq;
-#else
else if (*ubuf < 0xf8) {
if (uleft < 4 || !((ubuf[1] ^ 0x80) < 0x40 &&
(ubuf[2] ^ 0x80) < 0x40 && (ubuf[3] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xf1 || ubuf[1] >= 0x90)))
goto ilseq;
- *buf->outbuf++ = ((Py_UNICODE)(ubuf[0] & 0x07) << 18)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[3] ^ 0x80);
+ code = ((Py_UNICODE)(ubuf[0] & 0x07) << 18)
+ | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 12)
+ | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 6)
+ | (Py_UNICODE)(ubuf[3] ^ 0x80);
ubuf += 4;
} else if (*ubuf < 0xfc) {
if (uleft < 5 || !((ubuf[1] ^ 0x80) < 0x40 &&
@@ -1044,11 +1056,11 @@
(ubuf[4] ^ 0x80) < 0x40 && (ubuf[0] >= 0xf9 ||
ubuf[1] >= 0x88)))
goto ilseq;
- *buf->outbuf++ = ((Py_UNICODE)(ubuf[0] & 0x03) << 24)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 18)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[4] ^ 0x80);
+ code = ((Py_UNICODE)(ubuf[0] & 0x03) << 24)
+ | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 18)
+ | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 12)
+ | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 6)
+ | (Py_UNICODE)(ubuf[4] ^ 0x80);
ubuf += 5;
} else if (*ubuf < 0xff) {
if (uleft < 6 || !((ubuf[1] ^ 0x80) < 0x40 &&
@@ -1056,16 +1068,34 @@
(ubuf[4] ^ 0x80) < 0x40 && (ubuf[5] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xfd || ubuf[1] >= 0x84)))
goto ilseq;
- *buf->outbuf++ = ((Py_UNICODE)(ubuf[0] & 0x01) << 30)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 24)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 18)
- | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[4] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[5] ^ 0x80);
+ code = ((Py_UNICODE)(ubuf[0] & 0x01) << 30)
+ | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 24)
+ | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 18)
+ | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 12)
+ | ((Py_UNICODE)(ubuf[4] ^ 0x80) << 6)
+ | (Py_UNICODE)(ubuf[5] ^ 0x80);
ubuf += 6;
} else
goto ilseq;
+
+#if Py_UNICODE_SIZE == 2
+ if (code >= 0x10000) {
+ if (code >= 0x110000)
+ goto ilseq;
+
+ if (buf->outbuf_end <= buf->outbuf + 1) {
+ RESERVE_DECODEBUFFER(buf, -1)
+ }
+ *buf->outbuf++ = 0xd800 + ((code - 0x10000) >> 10);
+ *buf->outbuf++ = 0xdc00 + ((code - 0x10000) & 0x3ff);
+ } else
#endif
+ {
+ if (buf->outbuf_end <= buf->outbuf) {
+ RESERVE_DECODEBUFFER(buf, -1)
+ }
+ *buf->outbuf++ = (Py_UNICODE)code;
+ }
}
PyMem_Del(ubuf_top);
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 12:07:02
|
perky 03/06/11 05:07:01
Modified: . _iconv_codec.c
Log:
Use ucs4_t than Py_UNICODE in internal calculations
Revision Changes Path
1.9 +21 -21 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -r1.8 -r1.9
--- _iconv_codec.c 11 Jun 2003 12:01:57 -0000 1.8
+++ _iconv_codec.c 11 Jun 2003 12:07:01 -0000 1.9
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.8 2003/06/11 12:01:57 perky Exp $
+ * $Id: _iconv_codec.c,v 1.9 2003/06/11 12:07:01 perky Exp $
*/
#include "Python.h"
@@ -1027,17 +1027,17 @@
} else if (*ubuf < 0xe0) {
if (uleft < 2 || !((ubuf[1] ^ 0x80) < 0x40))
goto ilseq;
- code = ((Py_UNICODE)(ubuf[0] & 0x1f) << 6)
- | (Py_UNICODE)(ubuf[1] ^ 0x80);
+ code = ((ucs4_t)(ubuf[0] & 0x1f) << 6)
+ | (ucs4_t)(ubuf[1] ^ 0x80);
ubuf += 2;
} else if (*ubuf < 0xf0) {
if (uleft < 3 || !((ubuf[1] ^ 0x80) < 0x40 &&
(ubuf[2] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xe1 || ubuf[1] >= 0xa0)))
goto ilseq;
- code = ((Py_UNICODE)(ubuf[0] & 0x0f) << 12)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[2] ^ 0x80);
+ code = ((ucs4_t)(ubuf[0] & 0x0f) << 12)
+ | ((ucs4_t)(ubuf[1] ^ 0x80) << 6)
+ | (ucs4_t)(ubuf[2] ^ 0x80);
ubuf += 3;
}
else if (*ubuf < 0xf8) {
@@ -1045,10 +1045,10 @@
(ubuf[2] ^ 0x80) < 0x40 && (ubuf[3] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xf1 || ubuf[1] >= 0x90)))
goto ilseq;
- code = ((Py_UNICODE)(ubuf[0] & 0x07) << 18)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[3] ^ 0x80);
+ code = ((ucs4_t)(ubuf[0] & 0x07) << 18)
+ | ((ucs4_t)(ubuf[1] ^ 0x80) << 12)
+ | ((ucs4_t)(ubuf[2] ^ 0x80) << 6)
+ | (ucs4_t)(ubuf[3] ^ 0x80);
ubuf += 4;
} else if (*ubuf < 0xfc) {
if (uleft < 5 || !((ubuf[1] ^ 0x80) < 0x40 &&
@@ -1056,11 +1056,11 @@
(ubuf[4] ^ 0x80) < 0x40 && (ubuf[0] >= 0xf9 ||
ubuf[1] >= 0x88)))
goto ilseq;
- code = ((Py_UNICODE)(ubuf[0] & 0x03) << 24)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 18)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[4] ^ 0x80);
+ code = ((ucs4_t)(ubuf[0] & 0x03) << 24)
+ | ((ucs4_t)(ubuf[1] ^ 0x80) << 18)
+ | ((ucs4_t)(ubuf[2] ^ 0x80) << 12)
+ | ((ucs4_t)(ubuf[3] ^ 0x80) << 6)
+ | (ucs4_t)(ubuf[4] ^ 0x80);
ubuf += 5;
} else if (*ubuf < 0xff) {
if (uleft < 6 || !((ubuf[1] ^ 0x80) < 0x40 &&
@@ -1068,12 +1068,12 @@
(ubuf[4] ^ 0x80) < 0x40 && (ubuf[5] ^ 0x80) < 0x40 &&
(ubuf[0] >= 0xfd || ubuf[1] >= 0x84)))
goto ilseq;
- code = ((Py_UNICODE)(ubuf[0] & 0x01) << 30)
- | ((Py_UNICODE)(ubuf[1] ^ 0x80) << 24)
- | ((Py_UNICODE)(ubuf[2] ^ 0x80) << 18)
- | ((Py_UNICODE)(ubuf[3] ^ 0x80) << 12)
- | ((Py_UNICODE)(ubuf[4] ^ 0x80) << 6)
- | (Py_UNICODE)(ubuf[5] ^ 0x80);
+ code = ((ucs4_t)(ubuf[0] & 0x01) << 30)
+ | ((ucs4_t)(ubuf[1] ^ 0x80) << 24)
+ | ((ucs4_t)(ubuf[2] ^ 0x80) << 18)
+ | ((ucs4_t)(ubuf[3] ^ 0x80) << 12)
+ | ((ucs4_t)(ubuf[4] ^ 0x80) << 6)
+ | (ucs4_t)(ubuf[5] ^ 0x80);
ubuf += 6;
} else
goto ilseq;
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 12:30:43
|
perky 03/06/11 05:30:40
Modified: . _iconv_codec.c
Log:
Use a correct format string on python-ucs4
Revision Changes Path
1.10 +5 -1 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.9
retrieving revision 1.10
diff -u -r1.9 -r1.10
--- _iconv_codec.c 11 Jun 2003 12:07:01 -0000 1.9
+++ _iconv_codec.c 11 Jun 2003 12:30:38 -0000 1.10
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.9 2003/06/11 12:07:01 perky Exp $
+ * $Id: _iconv_codec.c,v 1.10 2003/06/11 12:30:38 perky Exp $
*/
#include "Python.h"
@@ -370,7 +370,11 @@
#ifdef LACKS_ERROR_CALLBACKS
if (esize == 1)
PyErr_Format(PyExc_UnicodeError,
+#if Py_UNICODE_SIZE == 2
"'%s' codec can't encode byte '\\u%04x' in position %d: %s",
+#else
+ "'%s' codec can't encode byte '\\u%08lx' in position %d: %s",
+#endif
self->encoding, *buf->inbuf, start, reason);
else
PyErr_Format(PyExc_UnicodeError,
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 12:35:31
|
perky 03/06/11 05:35:30
Modified: . _iconv_codec.c
Log:
Count 0xfe and 0xff as single-byte sequence.
Revision Changes Path
1.11 +4 -3 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.10
retrieving revision 1.11
diff -u -r1.10 -r1.11
--- _iconv_codec.c 11 Jun 2003 12:30:38 -0000 1.10
+++ _iconv_codec.c 11 Jun 2003 12:35:23 -0000 1.11
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.10 2003/06/11 12:30:38 perky Exp $
+ * $Id: _iconv_codec.c,v 1.11 2003/06/11 12:35:23 perky Exp $
*/
#include "Python.h"
@@ -230,7 +230,8 @@
else if (*(p) < 0xf0) (p) += 3; \
else if (*(p) < 0xf8) (p) += 4; \
else if (*(p) < 0xfc) (p) += 5; \
- else (p) += 6;
+ else if (*(p) < 0xfe) (p) += 6; \
+ else (p)++;
static const unsigned char *
skipchars_utf8(const unsigned char *st, int n)
@@ -1066,7 +1067,7 @@
| ((ucs4_t)(ubuf[3] ^ 0x80) << 6)
| (ucs4_t)(ubuf[4] ^ 0x80);
ubuf += 5;
- } else if (*ubuf < 0xff) {
+ } else if (*ubuf < 0xfe) {
if (uleft < 6 || !((ubuf[1] ^ 0x80) < 0x40 &&
(ubuf[2] ^ 0x80) < 0x40 && (ubuf[3] ^ 0x80) < 0x40 &&
(ubuf[4] ^ 0x80) < 0x40 && (ubuf[5] ^ 0x80) < 0x40 &&
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-11 12:40:19
|
perky 03/06/11 05:40:14
Modified: . _iconv_codec.c
Log:
Remove craps
Revision Changes Path
1.12 +2 -4 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.11
retrieving revision 1.12
diff -u -r1.11 -r1.12
--- _iconv_codec.c 11 Jun 2003 12:35:23 -0000 1.11
+++ _iconv_codec.c 11 Jun 2003 12:40:13 -0000 1.12
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.11 2003/06/11 12:35:23 perky Exp $
+ * $Id: _iconv_codec.c,v 1.12 2003/06/11 12:40:13 perky Exp $
*/
#include "Python.h"
@@ -2073,11 +2073,9 @@
void
init_iconv_codec(void)
{
- PyObject *m;
-
detect_iconv_endian();
- m = Py_InitModule("_iconv_codec", _iconv_codec_methods);
+ Py_InitModule("_iconv_codec", _iconv_codec_methods);
if (PyErr_Occurred())
Py_FatalError("can't initialize the _iconv_codec module");
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-12 05:51:35
|
perky 03/06/11 22:51:33
Modified: . _iconv_codec.c
Log:
Remove unreachable wastes.
Revision Changes Path
1.14 +1 -3 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.13
retrieving revision 1.14
diff -u -r1.13 -r1.14
--- _iconv_codec.c 12 Jun 2003 04:11:19 -0000 1.13
+++ _iconv_codec.c 12 Jun 2003 05:51:33 -0000 1.14
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.13 2003/06/12 04:11:19 perky Exp $
+ * $Id: _iconv_codec.c,v 1.14 2003/06/12 05:51:33 perky Exp $
*/
#include "Python.h"
@@ -775,7 +775,6 @@
"<IconvEncoder from='%s' to='%s' mode='%s'>",
self->unicode_encoding, self->encoding,
uniinternal_type_names[self->unitype]);
- return PyString_FromString("<IconvEncoder>");
}
#endif
@@ -1279,7 +1278,6 @@
"<IconvDecoder from='%s' to='%s' mode='%s'>",
self->encoding, self->unicode_encoding,
uniinternal_type_names[self->unitype]);
- return PyString_FromString("<IconvDecoder>");
}
#endif
|
|
From: Hye-Shik C. <pe...@us...> - 2003-06-16 19:12:46
|
perky 03/06/16 12:12:41
Modified: . _iconv_codec.c
Log:
Minor code clean ups
Revision Changes Path
1.15 +43 -49 iconvcodec/_iconv_codec.c
Index: _iconv_codec.c
===================================================================
RCS file: /cvsroot/koco/iconvcodec/_iconv_codec.c,v
retrieving revision 1.14
retrieving revision 1.15
diff -u -r1.14 -r1.15
--- _iconv_codec.c 12 Jun 2003 05:51:33 -0000 1.14
+++ _iconv_codec.c 16 Jun 2003 19:12:41 -0000 1.15
@@ -24,7 +24,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $Id: _iconv_codec.c,v 1.14 2003/06/12 05:51:33 perky Exp $
+ * $Id: _iconv_codec.c,v 1.15 2003/06/16 19:12:41 perky Exp $
*/
#include "Python.h"
@@ -127,10 +127,10 @@
const char *unicode_encoding;
int unitype;
size_t (*iconvwrap)(
- iconv_t cd, iconv_arg2_t inbuf, size_t * inbytesleft,
- char* *outbuf, size_t * outbytesleft);
+ iconv_t cd, iconv_arg2_t inbuf, size_t *inbytesleft,
+ char **outbuf, size_t *outbytesleft);
size_t (*iconvwrap2)(iconv_t cd, IconvDecoderBuffer *buf,
- int inleft, int outleft);
+ size_t inleft, size_t outleft);
/* an alternative wrapper: for utf8 backend */
} IconvDecoderObject;
@@ -187,9 +187,6 @@
{
PyObject *v, *w;
- if (unicode == NULL)
- return NULL;
-
v = PyTuple_New(2);
if (v == NULL) {
Py_DECREF(unicode);
@@ -225,7 +222,7 @@
}
#define UTF8NEXTCHAR(p) \
- if (*(p) < 128) (p)++; \
+ if (*(p) < 0x80) (p)++; \
else if (*(p) < 0xe0) (p) += 2; \
else if (*(p) < 0xf0) (p) += 3; \
else if (*(p) < 0xf8) (p) += 4; \
@@ -374,7 +371,7 @@
#if Py_UNICODE_SIZE == 2
"'%s' codec can't encode byte '\\u%04x' in position %d: %s",
#else
- "'%s' codec can't encode byte '\\u%08lx' in position %d: %s",
+ "'%s' codec can't encode byte '\\U%08lx' in position %d: %s",
#endif
self->encoding, *buf->inbuf, start, reason);
else
@@ -391,14 +388,11 @@
start, end, reason);
if (buf->excobj == NULL)
goto errorexit;
- } else {
- if (PyUnicodeEncodeError_SetStart(buf->excobj, start) != 0)
- goto errorexit;
- if (PyUnicodeEncodeError_SetEnd(buf->excobj, end) != 0)
- goto errorexit;
- if (PyUnicodeEncodeError_SetReason(buf->excobj, reason) != 0)
- goto errorexit;
- }
+ } else
+ if (PyUnicodeEncodeError_SetStart(buf->excobj, start) ||
+ PyUnicodeEncodeError_SetEnd(buf->excobj, end) ||
+ PyUnicodeEncodeError_SetReason(buf->excobj, reason))
+ goto errorexit;
if (errors == ERROR_STRICT) {
PyCodec_StrictErrors(buf->excobj);
@@ -739,9 +733,9 @@
iconvencoder_makestream(IconvEncoderObject *self,
PyObject *args, PyObject *kwargs)
{
- static char *stream_kwarglist[] = {"stream", "errors", NULL};
- PyObject *stream;
- char *errors = NULL;
+ static char *stream_kwarglist[] = {"stream", "errors", NULL};
+ PyObject *stream;
+ char *errors = NULL;
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O|s:makestream",
stream_kwarglist, &stream, &errors))
@@ -820,7 +814,7 @@
static int
expand_decodebuffer(IconvDecoderBuffer *buf, int esize)
{
- int orgpos, orgsize;
+ int orgpos, orgsize;
orgpos = (int)(buf->outbuf - PyUnicode_AS_UNICODE(buf->outobj));
orgsize = PyUnicode_GET_SIZE(buf->outobj);
@@ -906,14 +900,11 @@
(size_t)(buf->inbuf_end - buf->inbuf_top), start, end, reason);
if (buf->excobj == NULL)
goto errorexit;
- } else {
- if (PyUnicodeDecodeError_SetStart(buf->excobj, start) != 0)
- goto errorexit;
- if (PyUnicodeDecodeError_SetEnd(buf->excobj, end) != 0)
- goto errorexit;
- if (PyUnicodeDecodeError_SetReason(buf->excobj, reason) != 0)
- goto errorexit;
- }
+ } else
+ if (PyUnicodeDecodeError_SetStart(buf->excobj, start) ||
+ PyUnicodeDecodeError_SetEnd(buf->excobj, end) ||
+ PyUnicodeDecodeError_SetReason(buf->excobj, reason))
+ goto errorexit;
if (errors == ERROR_STRICT) {
PyCodec_StrictErrors(buf->excobj);
@@ -983,7 +974,8 @@
}
static size_t
-iconvwrap_utf8(iconv_t ic, IconvDecoderBuffer *buf, int inleft, int outleft)
+iconvwrap_utf8(iconv_t ic, IconvDecoderBuffer *buf,
+ size_t inleft, size_t outleft)
{
unsigned char *ubuf, *ubuf_top, *ubuf_end;
size_t r;
@@ -994,7 +986,7 @@
return -1;
ubuf_top = ubuf;
- if (inleft != -1)
+ if (inleft > 0)
r = iconv(ic, (iconv_arg2_t)&(buf->inbuf), &inleft,
(char **)&ubuf, &outleft);
else
@@ -1100,7 +1092,7 @@
iconvdecoder_conv(IconvDecoderObject *self, iconv_t ic,
IconvDecoderBuffer *buf, PyObject *errors)
{
- size_t r, inleft, outleft;
+ size_t r, inleft, outleft;
for (;;) {
inleft = (size_t)(buf->inbuf_end - buf->inbuf);
@@ -1133,7 +1125,7 @@
iconv_t ic, IconvDecoderBuffer *buf,
PyObject *errors)
{
- size_t r, outleft;
+ size_t r, outleft;
if (buf->inbuf < buf->inbuf_end) /* assumes as left by EINVAL */
if (iconvdecoder_error(self, ic, buf, errors, EINVAL,
@@ -1147,7 +1139,7 @@
r = self->iconvwrap(ic, NULL, NULL,
(char **)&(buf->outbuf), &outleft);
else
- r = self->iconvwrap2(ic, buf, -1, outleft);
+ r = self->iconvwrap2(ic, buf, 0, outleft);
if (r == (size_t)-1) {
if (errno == E2BIG) {
@@ -1242,9 +1234,9 @@
iconvdecoder_makestream(IconvDecoderObject *self,
PyObject *args, PyObject *kwargs)
{
- static char *stream_kwarglist[] = {"stream", "errors", NULL};
- PyObject *stream;
- char *errors = NULL;
+ static char *stream_kwarglist[] = {"stream", "errors", NULL};
+ PyObject *stream;
+ char *errors = NULL;
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O|s:makestream",
stream_kwarglist, &stream, &errors))
@@ -1325,8 +1317,8 @@
const char *method, int sizehint)
{
IconvDecoderBuffer buf;
- PyObject *cres;
- int rsize, r, finalsize = 0;
+ PyObject *cres;
+ int rsize, r, finalsize = 0;
if (sizehint == 0)
return PyUnicode_FromUnicode(NULL, 0);
@@ -1848,9 +1840,9 @@
iconvcodec_makeencoder(PyObject *spam, PyObject *args)
{
IconvEncoderObject *self;
- iconv_t ic;
- char *encoding;
- int i;
+ iconv_t ic;
+ char *encoding;
+ int i;
if (!PyArg_ParseTuple(args, "s:makeencoder", &encoding))
return NULL;
@@ -1898,9 +1890,9 @@
iconvcodec_makedecoder(PyObject *spam, PyObject *args)
{
IconvDecoderObject *self;
- iconv_t ic;
- char *encoding;
- int i;
+ iconv_t ic;
+ char *encoding;
+ int i;
if (!PyArg_ParseTuple(args, "s:makedecoder", &encoding))
return NULL;
@@ -1930,9 +1922,11 @@
self->unitype = um[i].type;
switch (self->unitype) {
case UNIINTERNAL_UCS:
- self->iconvwrap = iconv; break;
+ self->iconvwrap = iconv;
+ break;
case UNIINTERNAL_UCS_SWAPPED:
- self->iconvwrap = iconvwrap_ucsswapped; break;
+ self->iconvwrap = iconvwrap_ucsswapped;
+ break;
case UNIINTERNAL_UTF_8:
self->iconvwrap = NULL;
self->iconvwrap2= iconvwrap_utf8;
@@ -1968,8 +1962,8 @@
static void
detect_iconv_endian(void)
{
- iconv_t ic;
- int i;
+ iconv_t ic;
+ int i;
#define um uniinternal_modes
for (i = 0; um[i].encoding[0]; i++) {
|
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X