diff options
| -rw-r--r-- | generic/tclCompExpr.c | 32 | ||||
| -rw-r--r-- | generic/tclParse.c | 18 | ||||
| -rw-r--r-- | tests/parse.test | 3 | ||||
| -rw-r--r-- | tests/parseExpr.test | 6 |
4 files changed, 25 insertions, 34 deletions
diff --git a/generic/tclCompExpr.c b/generic/tclCompExpr.c index 9142e2b..dde4e56 100644 --- a/generic/tclCompExpr.c +++ b/generic/tclCompExpr.c @@ -1969,31 +1969,23 @@ ParseLexeme( } } - if (Tcl_UtfCharComplete(start, numBytes)) { - scanned = Tcl_UtfToUniChar(start, &ch); - } else { - char utfBytes[TCL_UTF_MAX]; - memcpy(utfBytes, start, (size_t) numBytes); - utfBytes[numBytes] = '\0'; - scanned = Tcl_UtfToUniChar(utfBytes, &ch); - } - if (!isalnum(UCHAR(ch))) { - *lexemePtr = INVALID; - Tcl_DecrRefCount(literal); - return scanned; - } - end = start; - while (isalnum(UCHAR(ch)) || (UCHAR(ch) == '_')) { - end += scanned; - numBytes -= scanned; - if (Tcl_UtfCharComplete(end, numBytes)) { - scanned = Tcl_UtfToUniChar(end, &ch); + if (!isalnum(UCHAR(*start))) { + if (Tcl_UtfCharComplete(start, numBytes)) { + scanned = Tcl_UtfToUniChar(start, &ch); } else { char utfBytes[TCL_UTF_MAX]; - memcpy(utfBytes, end, (size_t) numBytes); + memcpy(utfBytes, start, (size_t) numBytes); utfBytes[numBytes] = '\0'; scanned = Tcl_UtfToUniChar(utfBytes, &ch); } + *lexemePtr = INVALID; + Tcl_DecrRefCount(literal); + return scanned; + } + end = start; + while (numBytes && (isalnum(UCHAR(*end)) || (UCHAR(*end) == '_'))) { + end += 1; + numBytes -= 1; } *lexemePtr = BAREWORD; if (literalPtr) { diff --git a/generic/tclParse.c b/generic/tclParse.c index e475fb8..90ec43d 100644 --- a/generic/tclParse.c +++ b/generic/tclParse.c @@ -1344,8 +1344,7 @@ Tcl_ParseVarName( Tcl_Token *tokenPtr; register const char *src; unsigned char c; - int varIndex, offset; - Tcl_UniChar ch; + int varIndex; unsigned array; if ((numBytes == 0) || (start == NULL)) { @@ -1428,19 +1427,10 @@ Tcl_ParseVarName( tokenPtr->numComponents = 0; while (numBytes) { - if (Tcl_UtfCharComplete(src, numBytes)) { - offset = Tcl_UtfToUniChar(src, &ch); - } else { - char utfBytes[TCL_UTF_MAX]; - - memcpy(utfBytes, src, (size_t) numBytes); - utfBytes[numBytes] = '\0'; - offset = Tcl_UtfToUniChar(utfBytes, &ch); - } - c = UCHAR(ch); + c = UCHAR(*src); if (isalnum(c) || (c == '_')) { /* INTL: ISO only, UCHAR. */ - src += offset; - numBytes -= offset; + src += 1; + numBytes -= 1; continue; } if ((c == ':') && (numBytes != 1) && (src[1] == ':')) { diff --git a/tests/parse.test b/tests/parse.test index d7de5ff..cd02386 100644 --- a/tests/parse.test +++ b/tests/parse.test @@ -656,6 +656,9 @@ test parse-12.24 {Tcl_ParseVarName procedure, missing close paren in array refer test parse-12.25 {Tcl_ParseVarName procedure, nested array reference} testparser { testparser {$x(a$y(b$z))} 0 } {- {$x(a$y(b$z))} 1 word {$x(a$y(b$z))} 8 variable {$x(a$y(b$z))} 7 text x 0 text a 0 variable {$y(b$z)} 4 text y 0 text b 0 variable {$z} 1 text z 0 {}} +test parse-12.26 {Tcl_ParseVarName [d2ffcca163] non-ascii} testparser { + testparser "$\u0433" -1 +} "- {$\u0433} 1 word {$\u0433} 2 text {$} 0 text \u0433 0 {}" test parse-13.1 {Tcl_ParseVar procedure} testparsevar { set abc 24 diff --git a/tests/parseExpr.test b/tests/parseExpr.test index c1c489b..3e0df29 100644 --- a/tests/parseExpr.test +++ b/tests/parseExpr.test @@ -1051,6 +1051,12 @@ test parseExpr-22.18 {Bug 3401704} -constraints testexprparser -body { testexprparser 0b02 -1 } -returnCodes error -match glob -result {*invalid binary number*} +test parseExpr-22.19 {Bug d2ffcca163} -constraints testexprparser -body { + testexprparser \u0433 -1 +} -returnCodes error -match glob -result {*invalid character*} +test parseExpr-22.20 {Bug d2ffcca163} -constraints testexprparser -body { + testexprparser \u043f -1 +} -returnCodes error -match glob -result {*invalid character*} # cleanup cleanupTests |