ag/cpython-source-deps
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Import Tcl 8.6.12

Browse Source
This commit is contained in:
Steve Dower
2021-11-08 17:30:58 +00:00
parent 1aadb2455c
commit 674867e7e6
608 changed files with 78089 additions and 60360 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
generic/tclEncoding.c
View File

@@ -196,13 +196,13 @@ static unsigned short emptyPage[256];
*/
static Tcl_EncodingConvertProc BinaryProc;
static Tcl_DupInternalRepProc DupEncodingIntRep;
static Tcl_DupInternalRepProc DupEncodingInternalRep;
static Tcl_EncodingFreeProc EscapeFreeProc;
static Tcl_EncodingConvertProc EscapeFromUtfProc;
static Tcl_EncodingConvertProc EscapeToUtfProc;
static void FillEncodingFileMap(void);
static void FreeEncoding(Tcl_Encoding encoding);
static Tcl_FreeInternalRepProc FreeEncodingIntRep;
static Tcl_FreeInternalRepProc FreeEncodingInternalRep;
static Encoding * GetTableEncoding(EscapeEncodingData *dataPtr,
int state);
static Tcl_Encoding LoadEncodingFile(Tcl_Interp *interp,
@@ -262,12 +262,12 @@ static int Iso88591ToUtfProc(ClientData clientData,
/*
* A Tcl_ObjType for holding a cached Tcl_Encoding in the twoPtrValue.ptr1 field
* of the intrep. This should help the lifetime of encodings be more useful.
* of the internalrep. This should help the lifetime of encodings be more useful.
* See concerns raised in [Bug 1077262].
*/
static const Tcl_ObjType encodingType = {
"encoding", FreeEncodingIntRep, DupEncodingIntRep, NULL, NULL
"encoding", FreeEncodingInternalRep, DupEncodingInternalRep, NULL, NULL
};
/*
@@ -314,7 +314,7 @@ Tcl_GetEncodingFromObj(
/*
*----------------------------------------------------------------------
*
* FreeEncodingIntRep --
* FreeEncodingInternalRep --
*
* The Tcl_FreeInternalRepProc for the "encoding" Tcl_ObjType.
*
@@ -322,7 +322,7 @@ Tcl_GetEncodingFromObj(
*/
static void
FreeEncodingIntRep(
FreeEncodingInternalRep(
Tcl_Obj *objPtr)
{
Tcl_FreeEncoding((Tcl_Encoding)objPtr->internalRep.twoPtrValue.ptr1);
@@ -332,7 +332,7 @@ FreeEncodingIntRep(
/*
*----------------------------------------------------------------------
*
* DupEncodingIntRep --
* DupEncodingInternalRep --
*
* The Tcl_DupInternalRepProc for the "encoding" Tcl_ObjType.
*
@@ -340,7 +340,7 @@ FreeEncodingIntRep(
*/
static void
DupEncodingIntRep(
DupEncodingInternalRep(
Tcl_Obj *srcPtr,
Tcl_Obj *dupPtr)
{
@@ -1241,7 +1241,6 @@ Tcl_ExternalToUtf(
dstLen--;
}
do {
int savedFlags = flags;
Tcl_EncodingState savedState = *statePtr;
result = encodingPtr->toUtfProc(encodingPtr->clientData, src, srcLen,
@@ -1251,7 +1250,6 @@ Tcl_ExternalToUtf(
break;
}
dstLen = Tcl_UtfAtIndex(dst, maxChars) - dst + (TCL_UTF_MAX - 1);
flags = savedFlags;
*statePtr = savedState;
} while (1);
if (!noTerminate) {
@@ -2324,7 +2322,7 @@ UtfToUtfProc(
result = TCL_CONVERT_NOSPACE;
break;
}
if (UCHAR(*src) < 0x80 && !(UCHAR(*src) == 0 && pureNullMode == 0)) {
if (UCHAR(*src) < 0x80 && !((UCHAR(*src) == 0) && (pureNullMode == 0))) {
/*
* Copy 7bit characters, but skip null-bytes when we are in input
* mode, so that they get converted to 0xC080.
@@ -2332,8 +2330,8 @@ UtfToUtfProc(
*dst++ = *src++;
*chPtr = 0; /* reset surrogate handling */
} else if (pureNullMode == 1 && UCHAR(*src) == 0xC0 &&
(src + 1 < srcEnd) && UCHAR(*(src+1)) == 0x80) {
} else if ((UCHAR(*src) == 0xC0) && (src + 1 < srcEnd)
&& (UCHAR(src[1]) == 0x80) && (pureNullMode == 1)) {
/*
* Convert 0xC080 to real nulls when we are in output mode.
*/
@@ -2345,15 +2343,24 @@ UtfToUtfProc(
/*
* Always check before using TclUtfToUniChar. Not doing can so
* cause it run beyond the end of the buffer! If we happen such an
* incomplete char its bytes are made to represent themselves.
* incomplete char its bytes are made to represent themselves
* unless the user has explicitly asked to be told.
*/
if ((flags & TCL_ENCODING_STOPONERROR) && (pureNullMode == 0)) {
result = TCL_CONVERT_MULTIBYTE;
break;
}
*chPtr = UCHAR(*src);
src += 1;
dst += Tcl_UniCharToUtf(*chPtr, dst);
} else {
size_t len = TclUtfToUniChar(src, chPtr);
if ((len < 2) && (*chPtr != 0) && (flags & TCL_ENCODING_STOPONERROR)
&& ((*chPtr & ~0x7FF) != 0xD800) && (pureNullMode == 0)) {
result = TCL_CONVERT_SYNTAX;
break;
}
src += len;
if ((*chPtr & ~0x7FF) == 0xD800) {
Tcl_UniChar low;
@@ -2453,8 +2460,13 @@ UnicodeToUtfProc(
result = TCL_CONVERT_MULTIBYTE;
srcLen--;
}
/* If last code point is a high surrogate, we cannot handle that yet */
if ((srcLen >= 2) && ((src[srcLen - (clientData?1:2)] & 0xFC) == 0xD8)) {
/*
* If last code point is a high surrogate, we cannot handle that yet.
*/
if ((srcLen >= 2) &&
((src[srcLen - (clientData?1:2)] & 0xFC) == 0xD8)) {
result = TCL_CONVERT_MULTIBYTE;
srcLen-= 2;
}
@@ -2476,10 +2488,12 @@ UnicodeToUtfProc(
} else {
ch = (src[0] & 0xFF) << 8 | (src[1] & 0xFF);
}
/*
* Special case for 1-byte utf chars for speed. Make sure we work with
* unsigned short-size data.
*/
if (ch && ch < 0x80) {
*dst++ = (ch & 0xFF);
} else {
@@ -2580,7 +2594,7 @@ UtfToUnicodeProc(
*dst++ = (((*chPtr - 0x10000) >> 10) & 0xFF);
*dst++ = (((*chPtr - 0x10000) >> 18) & 0x3) | 0xD8;
*dst++ = (*chPtr & 0xFF);
*dst++ = ((*chPtr & 0x3) >> 8) | 0xDC;
*dst++ = ((*chPtr >> 8) & 0x3) | 0xDC;
}
#else
*dst++ = (*chPtr & 0xFF);
@@ -2592,10 +2606,10 @@ UtfToUnicodeProc(
*dst++ = (*chPtr >> 8);
*dst++ = (*chPtr & 0xFF);
} else {
*dst++ = ((*chPtr & 0x3) >> 8) | 0xDC;
*dst++ = (*chPtr & 0xFF);
*dst++ = (((*chPtr - 0x10000) >> 18) & 0x3) | 0xD8;
*dst++ = (((*chPtr - 0x10000) >> 10) & 0xFF);
*dst++ = ((*chPtr >> 8) & 0x3) | 0xDC;
*dst++ = (*chPtr & 0xFF);
}
#else
*dst++ = (*chPtr >> 8);
@@ -3015,7 +3029,9 @@ Iso88591FromUtfProc(
break;
}
#if TCL_UTF_MAX == 4
if ((ch >= 0xD800) && (len < 3)) len = 4;
if ((ch >= 0xD800) && (len < 3)) {
len = 4;
}
#endif
/*
* Plunge on, using '?' as a fallback character.
@@ -3060,7 +3076,7 @@ TableFreeProc(
ClientData clientData) /* TableEncodingData that specifies
* encoding. */
{
TableEncodingData *dataPtr = (TableEncodingData *)clientData;
TableEncodingData *dataPtr = (TableEncodingData *) clientData;
/*
* Make sure we aren't freeing twice on shutdown. [Bug 219314]
@@ -3118,7 +3134,7 @@ EscapeToUtfProc(
* correspond to the bytes stored in the
* output buffer. */
{
EscapeEncodingData *dataPtr = (EscapeEncodingData *)clientData;
EscapeEncodingData *dataPtr = (EscapeEncodingData *) clientData;
const char *prefixBytes, *tablePrefixBytes, *srcStart, *srcEnd;
const unsigned short *const *tableToUnicode;
const Encoding *encodingPtr;