From 729fcd5f0987531367d7572998e8be23743febd1 Mon Sep 17 00:00:00 2001 From: "jan.nijtmans" Date: Thu, 9 Nov 2017 12:48:54 +0000 Subject: Make "scan %c" and the internal function ExtendUnicodeRepWithString() work as expected for TCL_UTF_MAX == 4 when handling characters > U+ffff. No effect when TCL_UTF_MAX == 3 or TCL_UTF_MAX == 6 --- generic/tclScan.c | 12 ++++++++++-- generic/tclStringObj.c | 6 +++--- 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/generic/tclScan.c b/generic/tclScan.c index e1fcad4..7f71262 100644 --- a/generic/tclScan.c +++ b/generic/tclScan.c @@ -885,9 +885,17 @@ Tcl_ScanObjCmd( * Scan a single Unicode character. */ - string += TclUtfToUniChar(string, &sch); + offset = TclUtfToUniChar(string, &sch); + i = (int)sch; +#if TCL_UTF_MAX == 4 + if (!offset) { + offset = Tcl_UtfToUniChar(string, &sch); + i = (((i<<10) & 0x0FFC00) + 0x10000) + (sch & 0x3FF); + } +#endif + string += offset; if (!(flags & SCAN_SUPPRESS)) { - objPtr = Tcl_NewIntObj((int)sch); + objPtr = Tcl_NewIntObj(i); Tcl_IncrRefCount(objPtr); CLANG_ASSERT(objs); objs[objIndex++] = objPtr; diff --git a/generic/tclStringObj.c b/generic/tclStringObj.c index 01b044e..d43cc45 100644 --- a/generic/tclStringObj.c +++ b/generic/tclStringObj.c @@ -2720,7 +2720,6 @@ TclStringObjReverse( * Tcl_SetObjLength into growing the unicode rep buffer. */ - ch = 0; objPtr = Tcl_NewUnicodeObj(&ch, 1); Tcl_SetObjLength(objPtr, stringPtr->numChars); to = Tcl_GetUnicode(objPtr); @@ -2823,7 +2822,7 @@ ExtendUnicodeRepWithString( { String *stringPtr = GET_STRING(objPtr); int needed, numOrigChars = 0; - Tcl_UniChar *dst; + Tcl_UniChar *dst, unichar = 0; if (stringPtr->hasUnicode) { numOrigChars = stringPtr->numChars; @@ -2846,7 +2845,8 @@ ExtendUnicodeRepWithString( numAppendChars = 0; } for (dst=stringPtr->unicode + numOrigChars; numAppendChars-- > 0; dst++) { - bytes += TclUtfToUniChar(bytes, dst); + bytes += TclUtfToUniChar(bytes, &unichar); + *dst = unichar; } *dst = 0; } -- cgit v0.12