diff options
| author | dgp <dgp@users.sourceforge.net> | 2018-03-15 13:40:40 (GMT) |
|---|---|---|
| committer | dgp <dgp@users.sourceforge.net> | 2018-03-15 13:40:40 (GMT) |
| commit | 26e714137a987c67af5a932fdaf7bd1138d97a2d (patch) | |
| tree | 9358f84805ded45680d28bba41db129dfafb91e2 /generic/tclScan.c | |
| parent | a457b16dfc3bd4a4db9171364cd2a5ab04392bb8 (diff) | |
| parent | aa199edba612a516e6309290fb6dc4442a49a5ee (diff) | |
| download | tcl-26e714137a987c67af5a932fdaf7bd1138d97a2d.zip tcl-26e714137a987c67af5a932fdaf7bd1138d97a2d.tar.gz tcl-26e714137a987c67af5a932fdaf7bd1138d97a2d.tar.bz2 | |
merge 8.7
Diffstat (limited to 'generic/tclScan.c')
| -rw-r--r-- | generic/tclScan.c | 111 |
1 files changed, 62 insertions, 49 deletions
diff --git a/generic/tclScan.c b/generic/tclScan.c index 3edb8be..113b4c6 100644 --- a/generic/tclScan.c +++ b/generic/tclScan.c @@ -10,6 +10,7 @@ */ #include "tclInt.h" +#include "tommath.h" /* * Flag values used by Tcl_ScanObjCmd. @@ -72,17 +73,17 @@ BuildCharSet( CharSet *cset, const char *format) /* Points to first char of set. */ { - Tcl_UniChar ch, start; + Tcl_UniChar ch = 0, start; int offset, nranges; const char *end; memset(cset, 0, sizeof(CharSet)); - offset = Tcl_UtfToUniChar(format, &ch); + offset = TclUtfToUniChar(format, &ch); if (ch == '^') { cset->exclude = 1; format += offset; - offset = Tcl_UtfToUniChar(format, &ch); + offset = TclUtfToUniChar(format, &ch); } end = format + offset; @@ -91,14 +92,14 @@ BuildCharSet( */ if (ch == ']') { - end += Tcl_UtfToUniChar(end, &ch); + end += TclUtfToUniChar(end, &ch); } nranges = 0; while (ch != ']') { if (ch == '-') { nranges++; } - end += Tcl_UtfToUniChar(end, &ch); + end += TclUtfToUniChar(end, &ch); } cset->chars = ckalloc(sizeof(Tcl_UniChar) * (end - format - 1)); @@ -113,11 +114,11 @@ BuildCharSet( */ cset->nchars = cset->nranges = 0; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); start = ch; if (ch == ']' || ch == '-') { cset->chars[cset->nchars++] = ch; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } while (ch != ']') { if (*format == '-') { @@ -138,7 +139,7 @@ BuildCharSet( cset->chars[cset->nchars++] = start; cset->chars[cset->nchars++] = ch; } else { - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); /* * Check to see if the range is in reverse order. @@ -156,7 +157,7 @@ BuildCharSet( } else { cset->chars[cset->nchars++] = ch; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } return format; } @@ -257,7 +258,7 @@ ValidateFormat( { int gotXpg, gotSequential, value, i, flags; char *end; - Tcl_UniChar ch; + Tcl_UniChar ch = 0; int objIndex, xpgSize, nspace = numVars; int *nassign = TclStackAlloc(interp, nspace * sizeof(int)); char buf[TCL_UTF_MAX+1]; @@ -279,20 +280,20 @@ ValidateFormat( xpgSize = objIndex = gotXpg = gotSequential = 0; while (*format != '\0') { - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); flags = 0; if (ch != '%') { continue; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); if (ch == '%') { continue; } if (ch == '*') { flags |= SCAN_SUPPRESS; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); goto xpgCheckDone; } @@ -308,7 +309,7 @@ ValidateFormat( goto notXpg; } format = end+1; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); gotXpg = 1; if (gotSequential) { goto mixedXPG; @@ -347,7 +348,7 @@ ValidateFormat( if ((ch < 0x80) && isdigit(UCHAR(ch))) { /* INTL: "C" locale. */ value = strtoul(format-1, (char **) &format, 10); /* INTL: "C" locale. */ flags |= SCAN_WIDTH; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } /* @@ -359,13 +360,13 @@ ValidateFormat( if (*format == 'l') { flags |= SCAN_BIG; format += 1; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); break; } case 'L': flags |= SCAN_LONGER; case 'h': - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } if (!(flags & SCAN_SUPPRESS) && numVars && (objIndex >= numVars)) { @@ -415,14 +416,7 @@ ValidateFormat( case 'x': case 'X': case 'b': - break; case 'u': - if (flags & SCAN_BIG) { - Tcl_SetObjResult(interp, Tcl_NewStringObj( - "unsigned bignum scans are invalid", -1)); - Tcl_SetErrorCode(interp, "TCL", "FORMAT", "BADUNSIGNED",NULL); - goto error; - } break; /* * Bracket terms need special checking @@ -434,24 +428,24 @@ ValidateFormat( if (*format == '\0') { goto badSet; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); if (ch == '^') { if (*format == '\0') { goto badSet; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } if (ch == ']') { if (*format == '\0') { goto badSet; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } while (ch != ']') { if (*format == '\0') { goto badSet; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } break; badSet: @@ -582,7 +576,7 @@ Tcl_ScanObjCmd( char op = 0; int width, underflow = 0; Tcl_WideInt wideValue; - Tcl_UniChar ch, sch; + Tcl_UniChar ch = 0, sch = 0; Tcl_Obj **objs = NULL, *objPtr = NULL; int flags; char buf[513]; /* Temporary buffer to hold scanned number @@ -630,7 +624,7 @@ Tcl_ScanObjCmd( nconversions = 0; while (*format != '\0') { int parseFlag = TCL_PARSE_NO_WHITESPACE; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); flags = 0; @@ -639,13 +633,13 @@ Tcl_ScanObjCmd( */ if (Tcl_UniCharIsSpace(ch)) { - offset = Tcl_UtfToUniChar(string, &sch); + offset = TclUtfToUniChar(string, &sch); while (Tcl_UniCharIsSpace(sch)) { if (*string == '\0') { goto done; } string += offset; - offset = Tcl_UtfToUniChar(string, &sch); + offset = TclUtfToUniChar(string, &sch); } continue; } @@ -656,14 +650,14 @@ Tcl_ScanObjCmd( underflow = 1; goto done; } - string += Tcl_UtfToUniChar(string, &sch); + string += TclUtfToUniChar(string, &sch); if (ch != sch) { goto done; } continue; } - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); if (ch == '%') { goto literal; } @@ -675,13 +669,13 @@ Tcl_ScanObjCmd( if (ch == '*') { flags |= SCAN_SUPPRESS; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } else if ((ch < 0x80) && isdigit(UCHAR(ch))) { /* INTL: "C" locale. */ char *formatEnd; value = strtoul(format-1, &formatEnd, 10);/* INTL: "C" locale. */ if (*formatEnd == '$') { format = formatEnd+1; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); objIndex = (int) value - 1; } } @@ -692,7 +686,7 @@ Tcl_ScanObjCmd( if ((ch < 0x80) && isdigit(UCHAR(ch))) { /* INTL: "C" locale. */ width = (int) strtoul(format-1, (char **) &format, 10);/* INTL: "C" locale. */ - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } else { width = 0; } @@ -706,7 +700,7 @@ Tcl_ScanObjCmd( if (*format == 'l') { flags |= SCAN_BIG; format += 1; - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); break; } case 'L': @@ -715,7 +709,7 @@ Tcl_ScanObjCmd( * Fall through so we skip to the next character. */ case 'h': - format += Tcl_UtfToUniChar(format, &ch); + format += TclUtfToUniChar(format, &ch); } /* @@ -799,7 +793,7 @@ Tcl_ScanObjCmd( if (!(flags & SCAN_NOSKIP)) { while (*string != '\0') { - offset = Tcl_UtfToUniChar(string, &sch); + offset = TclUtfToUniChar(string, &sch); if (!Tcl_UniCharIsSpace(sch)) { break; } @@ -826,7 +820,7 @@ Tcl_ScanObjCmd( } end = string; while (*end != '\0') { - offset = Tcl_UtfToUniChar(end, &sch); + offset = TclUtfToUniChar(end, &sch); if (Tcl_UniCharIsSpace(sch)) { break; } @@ -854,7 +848,7 @@ Tcl_ScanObjCmd( format = BuildCharSet(&cset, format); while (*end != '\0') { - offset = Tcl_UtfToUniChar(end, &sch); + offset = TclUtfToUniChar(end, &sch); if (!CharInSet(&cset, (int)sch)) { break; } @@ -885,9 +879,17 @@ Tcl_ScanObjCmd( * Scan a single Unicode character. */ - string += Tcl_UtfToUniChar(string, &sch); + offset = TclUtfToUniChar(string, &sch); + i = (int)sch; +#if TCL_UTF_MAX == 4 + if (!offset) { + offset = TclUtfToUniChar(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; @@ -924,9 +926,9 @@ Tcl_ScanObjCmd( } if (flags & SCAN_LONGER) { if (Tcl_GetWideIntFromObj(NULL, objPtr, &wideValue) != TCL_OK) { - wideValue = ~(Tcl_WideUInt)0 >> 1; /* WIDE_MAX */ + wideValue = LLONG_MAX; if (TclGetString(objPtr)[0] == '-') { - wideValue++; /* WIDE_MAX + 1 = WIDE_MIN */ + wideValue = LLONG_MIN; } } if ((flags & SCAN_UNSIGNED) && (wideValue < 0)) { @@ -934,9 +936,20 @@ Tcl_ScanObjCmd( (Tcl_WideUInt)wideValue); Tcl_SetStringObj(objPtr, buf, -1); } else { - Tcl_SetWideIntObj(objPtr, wideValue); + TclSetIntObj(objPtr, wideValue); + } + } else if (flags & SCAN_BIG) { + if (flags & SCAN_UNSIGNED) { + mp_int big; + if ((Tcl_GetBignumFromObj(interp, objPtr, &big) != TCL_OK) + || mp_isneg(&big)) { + Tcl_SetObjResult(interp, Tcl_NewStringObj( + "unsigned bignum scans are invalid", -1)); + Tcl_SetErrorCode(interp, "TCL", "FORMAT", "BADUNSIGNED",NULL); + return TCL_ERROR; + } } - } else if (!(flags & SCAN_BIG)) { + } else { if (TclGetLongFromObj(NULL, objPtr, &value) != TCL_OK) { if (TclGetString(objPtr)[0] == '-') { value = LONG_MIN; @@ -948,7 +961,7 @@ Tcl_ScanObjCmd( sprintf(buf, "%lu", value); /* INTL: ISO digit */ Tcl_SetStringObj(objPtr, buf, -1); } else { - Tcl_SetLongObj(objPtr, value); + TclSetIntObj(objPtr, value); } } objs[objIndex++] = objPtr; |
