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

Import Tcl 8.6.10

Browse Source
This commit is contained in:
Steve Dower
2020-09-24 22:53:56 +01:00
parent 0343d03b22
commit 3bb8e3e086
1005 changed files with 593700 additions and 41637 deletions
Download Patch File Download Diff File Expand all files Collapse all files

128
generic/tclEncoding.c
View File

@@ -305,7 +305,7 @@ Tcl_GetEncodingFromObj(
Tcl_Obj *objPtr,
Tcl_Encoding *encodingPtr)
{
const char *name = Tcl_GetString(objPtr);
const char *name = TclGetString(objPtr);
if (objPtr->typePtr != &encodingType) {
Tcl_Encoding encoding = Tcl_GetEncoding(interp, name);
@@ -704,7 +704,7 @@ Tcl_GetDefaultEncodingDir(void)
}
Tcl_ListObjIndex(NULL, searchPath, 0, &first);
return Tcl_GetString(first);
return TclGetString(first);
}
/*
@@ -1260,7 +1260,7 @@ Tcl_ExternalToUtf(
if (*dstCharsPtr <= maxChars) {
break;
}
dstLen = Tcl_UtfAtIndex(dst, maxChars) - 1 - dst + TCL_UTF_MAX;
dstLen = Tcl_UtfAtIndex(dst, maxChars) - dst + (TCL_UTF_MAX - 1);
flags = savedFlags;
*statePtr = savedState;
} while (1);
@@ -1518,10 +1518,10 @@ OpenEncodingFileChannel(
}
}
if (!verified) {
const char *dirString = Tcl_GetString(directory);
const char *dirString = TclGetString(directory);
for (i=0; i<numDirs && !verified; i++) {
if (strcmp(dirString, Tcl_GetString(dir[i])) == 0) {
if (strcmp(dirString, TclGetString(dir[i])) == 0) {
verified = 1;
}
}
@@ -1720,7 +1720,9 @@ LoadTableEncoding(
};
Tcl_DStringInit(&lineString);
Tcl_Gets(chan, &lineString);
if (Tcl_Gets(chan, &lineString) == -1) {
return NULL;
}
line = Tcl_DStringValue(&lineString);
fallback = (int) strtol(line, &line, 16);
@@ -1760,9 +1762,12 @@ LoadTableEncoding(
for (i = 0; i < numPages; i++) {
int ch;
const char *p;
int expected = 3 + 16 * (16 * 4 + 1);
Tcl_ReadChars(chan, objPtr, 3 + 16 * (16 * 4 + 1), 0);
p = Tcl_GetString(objPtr);
if (Tcl_ReadChars(chan, objPtr, expected, 0) != expected) {
return NULL;
}
p = TclGetString(objPtr);
hi = (staticHex[UCHAR(p[0])] << 4) + staticHex[UCHAR(p[1])];
dataPtr->toUnicode[hi] = pageMemPtr;
p += 2;
@@ -2049,13 +2054,13 @@ LoadEscapeEncoding(
+ Tcl_DStringLength(&escapeData);
dataPtr = ckalloc(size);
dataPtr->initLen = strlen(init);
memcpy(dataPtr->init, init, (unsigned) dataPtr->initLen + 1);
memcpy(dataPtr->init, init, dataPtr->initLen + 1);
dataPtr->finalLen = strlen(final);
memcpy(dataPtr->final, final, (unsigned) dataPtr->finalLen + 1);
memcpy(dataPtr->final, final, dataPtr->finalLen + 1);
dataPtr->numSubTables =
Tcl_DStringLength(&escapeData) / sizeof(EscapeSubTable);
memcpy(dataPtr->subTables, Tcl_DStringValue(&escapeData),
(size_t) Tcl_DStringLength(&escapeData));
Tcl_DStringLength(&escapeData));
Tcl_DStringFree(&escapeData);
memset(dataPtr->prefixBytes, 0, sizeof(dataPtr->prefixBytes));
@@ -2143,14 +2148,14 @@ BinaryProc(
*srcReadPtr = srcLen;
*dstWrotePtr = srcLen;
*dstCharsPtr = srcLen;
memcpy(dst, src, (size_t) srcLen);
memcpy(dst, src, srcLen);
return result;
}
/*
*-------------------------------------------------------------------------
*
* UtfExtToUtfIntProc --
* UtfIntToUtfExtProc --
*
* Convert from UTF-8 to UTF-8. While converting null-bytes from the
* Tcl's internal representation (0xc0, 0x80) to the official
@@ -2291,27 +2296,30 @@ UtfToUtfProc(
* output buffer. */
int pureNullMode) /* Convert embedded nulls from internal
* representation to real null-bytes or vice
* versa. */
* versa. Also combine or separate surrogate pairs */
{
const char *srcStart, *srcEnd, *srcClose;
const char *dstStart, *dstEnd;
int result, numChars, charLimit = INT_MAX;
Tcl_UniChar ch = 0;
Tcl_UniChar *chPtr = (Tcl_UniChar *) statePtr;
if (flags & TCL_ENCODING_START) {
*statePtr = 0;
}
result = TCL_OK;
srcStart = src;
srcEnd = src + srcLen;
srcClose = srcEnd;
if ((flags & TCL_ENCODING_END) == 0) {
srcClose -= TCL_UTF_MAX;
srcClose -= 6;
}
if (flags & TCL_ENCODING_CHAR_LIMIT) {
charLimit = *dstCharsPtr;
}
dstStart = dst;
dstEnd = dst + dstLen - TCL_UTF_MAX;
dstEnd = dst + dstLen - ((pureNullMode == 1) ? 4 : TCL_UTF_MAX);
for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {
if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
@@ -2349,12 +2357,32 @@ UtfToUtfProc(
* incomplete char its bytes are made to represent themselves.
*/
ch = (unsigned char) *src;
*chPtr = (unsigned char) *src;
src += 1;
dst += Tcl_UniCharToUtf(ch, dst);
dst += Tcl_UniCharToUtf(*chPtr, dst);
} else {
src += TclUtfToUniChar(src, &ch);
dst += Tcl_UniCharToUtf(ch, dst);
src += TclUtfToUniChar(src, chPtr);
if ((*chPtr & 0xFC00) == 0xD800) {
/* A high surrogate character is detected, handle especially */
Tcl_UniChar low = *chPtr;
size_t len = (src <= srcEnd-3) ? Tcl_UtfToUniChar(src, &low) : 0;
if ((low & 0xFC00) != 0xDC00) {
*dst++ = (char) (((*chPtr >> 12) | 0xE0) & 0xEF);
*dst++ = (char) (((*chPtr >> 6) | 0x80) & 0xBF);
*dst++ = (char) ((*chPtr | 0x80) & 0xBF);
continue;
} else if (pureNullMode == 1) {
int full = (((*chPtr & 0x3FF) << 10) | (low & 0x3FF)) + 0x10000;
*dst++ = (char) (((full >> 18) | 0xF0) & 0xF7);
*dst++ = (char) (((full >> 12) | 0x80) & 0xBF);
*dst++ = (char) (((full >> 6) | 0x80) & 0xBF);
*dst++ = (char) ((full | 0x80) & 0xBF);
*chPtr = 0;
src += len;
continue;
}
}
dst += Tcl_UniCharToUtf(*chPtr, dst);
}
}
@@ -2410,7 +2438,7 @@ UnicodeToUtfProc(
const char *srcStart, *srcEnd;
const char *dstEnd, *dstStart;
int result, numChars, charLimit = INT_MAX;
Tcl_UniChar ch = 0;
Tcl_UniChar ch;
if (flags & TCL_ENCODING_CHAR_LIMIT) {
charLimit = *dstCharsPtr;
@@ -2500,8 +2528,11 @@ UtfToUnicodeProc(
{
const char *srcStart, *srcEnd, *srcClose, *dstStart, *dstEnd;
int result, numChars;
Tcl_UniChar ch = 0;
Tcl_UniChar *chPtr = (Tcl_UniChar *) statePtr;
if (flags & TCL_ENCODING_START) {
*statePtr = 0;
}
srcStart = src;
srcEnd = src + srcLen;
srcClose = srcEnd;
@@ -2527,7 +2558,7 @@ UtfToUnicodeProc(
result = TCL_CONVERT_NOSPACE;
break;
}
src += TclUtfToUniChar(src, &ch);
src += TclUtfToUniChar(src, chPtr);
/*
* Need to handle this in a way that won't cause misalignment by
@@ -2536,23 +2567,23 @@ UtfToUnicodeProc(
#ifdef WORDS_BIGENDIAN
#if TCL_UTF_MAX > 4
*dst++ = (ch >> 24);
*dst++ = ((ch >> 16) & 0xFF);
*dst++ = ((ch >> 8) & 0xFF);
*dst++ = (ch & 0xFF);
*dst++ = (*chPtr >> 24);
*dst++ = ((*chPtr >> 16) & 0xFF);
*dst++ = ((*chPtr >> 8) & 0xFF);
*dst++ = (*chPtr & 0xFF);
#else
*dst++ = (ch >> 8);
*dst++ = (ch & 0xFF);
*dst++ = (*chPtr >> 8);
*dst++ = (*chPtr & 0xFF);
#endif
#else
#if TCL_UTF_MAX > 4
*dst++ = (ch & 0xFF);
*dst++ = ((ch >> 8) & 0xFF);
*dst++ = ((ch >> 16) & 0xFF);
*dst++ = (ch >> 24);
*dst++ = (*chPtr & 0xFF);
*dst++ = ((*chPtr >> 8) & 0xFF);
*dst++ = ((*chPtr >> 16) & 0xFF);
*dst++ = (*chPtr >> 24);
#else
*dst++ = (ch & 0xFF);
*dst++ = (ch >> 8);
*dst++ = (*chPtr & 0xFF);
*dst++ = (*chPtr >> 8);
#endif
#endif
}
@@ -2754,7 +2785,7 @@ TableFromUtfProc(
}
len = TclUtfToUniChar(src, &ch);
#if TCL_UTF_MAX > 3
#if TCL_UTF_MAX > 4
/*
* This prevents a crash condition. More evaluation is required for
* full support of int Tcl_UniChar. [Bug 1004065]
@@ -2763,6 +2794,10 @@ TableFromUtfProc(
if (ch & 0xffff0000) {
word = 0;
} else
#elif TCL_UTF_MAX == 4
if (!len) {
word = 0;
} else
#endif
word = fromUnicode[(ch >> 8)][ch & 0xff];
@@ -2928,6 +2963,7 @@ Iso88591FromUtfProc(
const char *srcStart, *srcEnd, *srcClose;
const char *dstStart, *dstEnd;
int result, numChars;
Tcl_UniChar ch = 0;
result = TCL_OK;
@@ -2942,7 +2978,6 @@ Iso88591FromUtfProc(
dstEnd = dst + dstLen - 1;
for (numChars = 0; src < srcEnd; numChars++) {
Tcl_UniChar ch = 0;
int len;
if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
@@ -2960,11 +2995,18 @@ Iso88591FromUtfProc(
* Check for illegal characters.
*/
if (ch > 0xff) {
if (ch > 0xff
#if TCL_UTF_MAX == 4
|| ((ch >= 0xD800) && (len < 3))
#endif
) {
if (flags & TCL_ENCODING_STOPONERROR) {
result = TCL_CONVERT_UNKNOWN;
break;
}
#if TCL_UTF_MAX == 4
if ((ch >= 0xD800) && (len < 3)) len = 4;
#endif
/*
* Plunge on, using '?' as a fallback character.
@@ -3289,6 +3331,7 @@ EscapeFromUtfProc(
const TableEncodingData *tableDataPtr;
const char *tablePrefixBytes;
const unsigned short *const *tableFromUnicode;
Tcl_UniChar ch = 0;
result = TCL_OK;
@@ -3314,7 +3357,7 @@ EscapeFromUtfProc(
*dstWrotePtr = 0;
return TCL_CONVERT_NOSPACE;
}
memcpy(dst, dataPtr->init, (size_t)dataPtr->initLen);
memcpy(dst, dataPtr->init, dataPtr->initLen);
dst += dataPtr->initLen;
} else {
state = PTR2INT(*statePtr);
@@ -3329,7 +3372,6 @@ EscapeFromUtfProc(
for (numChars = 0; src < srcEnd; numChars++) {
unsigned len;
int word;
Tcl_UniChar ch = 0;
if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
/*
@@ -3393,7 +3435,7 @@ EscapeFromUtfProc(
break;
}
memcpy(dst, subTablePtr->sequence,
(size_t) subTablePtr->sequenceLen);
subTablePtr->sequenceLen);
dst += subTablePtr->sequenceLen;
}
}
@@ -3436,7 +3478,7 @@ EscapeFromUtfProc(
memcpy(dst, dataPtr->subTables[0].sequence, len);
dst += len;
}
memcpy(dst, dataPtr->final, (size_t) dataPtr->finalLen);
memcpy(dst, dataPtr->final, dataPtr->finalLen);
dst += dataPtr->finalLen;
state &= ~TCL_ENCODING_END;
}