1 /*
2 * Copyright (C) 1999-2002 Harri Porten (porten@kde.org)
3 * Copyright (C) 2001 Peter Kelly (pmk@post.com)
4 * Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009 Apple Inc. All rights reserved.
5 * Copyright (C) 2007 Cameron Zwarich (cwzwarich@uwaterloo.ca)
6 * Copyright (C) 2007 Maks Orlovich
7 *
8 * This library is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Library General Public
10 * License as published by the Free Software Foundation; either
11 * version 2 of the License, or (at your option) any later version.
12 *
13 * This library is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 * Library General Public License for more details.
17 *
18 * You should have received a copy of the GNU Library General Public License
19 * along with this library; see the file COPYING.LIB. If not, write to
20 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
21 * Boston, MA 02110-1301, USA.
22 *
23 */
24
25 #include "config.h"
26 #include "JSGlobalObjectFunctions.h"
27
28 #include "CallFrame.h"
29 #include "GlobalEvalFunction.h"
30 #include "JSGlobalObject.h"
31 #include "LiteralParser.h"
32 #include "JSString.h"
33 #include "Interpreter.h"
34 #include "Parser.h"
35 #include "dtoa.h"
36 #include "Lexer.h"
37 #include "Nodes.h"
38 #include <stdio.h>
39 #include <stdlib.h>
40 #include <string.h>
41 #include <wtf/ASCIICType.h>
42 #include <wtf/Assertions.h>
43 #include <wtf/MathExtras.h>
44 #include <wtf/unicode/UTF8.h>
45
46 using namespace WTF;
47 using namespace Unicode;
48
49 namespace JSC {
50
encode(ExecState * exec,const ArgList & args,const char * doNotEscape)51 static JSValue encode(ExecState* exec, const ArgList& args, const char* doNotEscape)
52 {
53 UString str = args.at(0).toString(exec);
54 CString cstr = str.UTF8String(true);
55 if (!cstr.c_str())
56 return throwError(exec, URIError, "String contained an illegal UTF-16 sequence.");
57
58 UString result = "";
59 const char* p = cstr.c_str();
60 for (size_t k = 0; k < cstr.size(); k++, p++) {
61 char c = *p;
62 if (c && strchr(doNotEscape, c))
63 result.append(c);
64 else {
65 char tmp[4];
66 sprintf(tmp, "%%%02X", static_cast<unsigned char>(c));
67 result += tmp;
68 }
69 }
70 return jsString(exec, result);
71 }
72
decode(ExecState * exec,const ArgList & args,const char * doNotUnescape,bool strict)73 static JSValue decode(ExecState* exec, const ArgList& args, const char* doNotUnescape, bool strict)
74 {
75 UString result = "";
76 UString str = args.at(0).toString(exec);
77 int k = 0;
78 int len = str.size();
79 const UChar* d = str.data();
80 UChar u = 0;
81 while (k < len) {
82 const UChar* p = d + k;
83 UChar c = *p;
84 if (c == '%') {
85 int charLen = 0;
86 if (k <= len - 3 && isASCIIHexDigit(p[1]) && isASCIIHexDigit(p[2])) {
87 const char b0 = Lexer::convertHex(p[1], p[2]);
88 const int sequenceLen = UTF8SequenceLength(b0);
89 if (sequenceLen != 0 && k <= len - sequenceLen * 3) {
90 charLen = sequenceLen * 3;
91 char sequence[5];
92 sequence[0] = b0;
93 for (int i = 1; i < sequenceLen; ++i) {
94 const UChar* q = p + i * 3;
95 if (q[0] == '%' && isASCIIHexDigit(q[1]) && isASCIIHexDigit(q[2]))
96 sequence[i] = Lexer::convertHex(q[1], q[2]);
97 else {
98 charLen = 0;
99 break;
100 }
101 }
102 if (charLen != 0) {
103 sequence[sequenceLen] = 0;
104 const int character = decodeUTF8Sequence(sequence);
105 if (character < 0 || character >= 0x110000)
106 charLen = 0;
107 else if (character >= 0x10000) {
108 // Convert to surrogate pair.
109 result.append(static_cast<UChar>(0xD800 | ((character - 0x10000) >> 10)));
110 u = static_cast<UChar>(0xDC00 | ((character - 0x10000) & 0x3FF));
111 } else
112 u = static_cast<UChar>(character);
113 }
114 }
115 }
116 if (charLen == 0) {
117 if (strict)
118 return throwError(exec, URIError);
119 // The only case where we don't use "strict" mode is the "unescape" function.
120 // For that, it's good to support the wonky "%u" syntax for compatibility with WinIE.
121 if (k <= len - 6 && p[1] == 'u'
122 && isASCIIHexDigit(p[2]) && isASCIIHexDigit(p[3])
123 && isASCIIHexDigit(p[4]) && isASCIIHexDigit(p[5])) {
124 charLen = 6;
125 u = Lexer::convertUnicode(p[2], p[3], p[4], p[5]);
126 }
127 }
128 if (charLen && (u == 0 || u >= 128 || !strchr(doNotUnescape, u))) {
129 c = u;
130 k += charLen - 1;
131 }
132 }
133 k++;
134 result.append(c);
135 }
136 return jsString(exec, result);
137 }
138
isStrWhiteSpace(UChar c)139 bool isStrWhiteSpace(UChar c)
140 {
141 switch (c) {
142 case 0x0009:
143 case 0x000A:
144 case 0x000B:
145 case 0x000C:
146 case 0x000D:
147 case 0x0020:
148 case 0x00A0:
149 case 0x2028:
150 case 0x2029:
151 return true;
152 default:
153 return c > 0xff && isSeparatorSpace(c);
154 }
155 }
156
parseDigit(unsigned short c,int radix)157 static int parseDigit(unsigned short c, int radix)
158 {
159 int digit = -1;
160
161 if (c >= '0' && c <= '9')
162 digit = c - '0';
163 else if (c >= 'A' && c <= 'Z')
164 digit = c - 'A' + 10;
165 else if (c >= 'a' && c <= 'z')
166 digit = c - 'a' + 10;
167
168 if (digit >= radix)
169 return -1;
170 return digit;
171 }
172
parseIntOverflow(const char * s,int length,int radix)173 double parseIntOverflow(const char* s, int length, int radix)
174 {
175 double number = 0.0;
176 double radixMultiplier = 1.0;
177
178 for (const char* p = s + length - 1; p >= s; p--) {
179 if (radixMultiplier == Inf) {
180 if (*p != '0') {
181 number = Inf;
182 break;
183 }
184 } else {
185 int digit = parseDigit(*p, radix);
186 number += digit * radixMultiplier;
187 }
188
189 radixMultiplier *= radix;
190 }
191
192 return number;
193 }
194
parseInt(const UString & s,int radix)195 static double parseInt(const UString& s, int radix)
196 {
197 int length = s.size();
198 const UChar* data = s.data();
199 int p = 0;
200
201 while (p < length && isStrWhiteSpace(data[p]))
202 ++p;
203
204 double sign = 1;
205 if (p < length) {
206 if (data[p] == '+')
207 ++p;
208 else if (data[p] == '-') {
209 sign = -1;
210 ++p;
211 }
212 }
213
214 if ((radix == 0 || radix == 16) && length - p >= 2 && data[p] == '0' && (data[p + 1] == 'x' || data[p + 1] == 'X')) {
215 radix = 16;
216 p += 2;
217 } else if (radix == 0) {
218 if (p < length && data[p] == '0')
219 radix = 8;
220 else
221 radix = 10;
222 }
223
224 if (radix < 2 || radix > 36)
225 return NaN;
226
227 int firstDigitPosition = p;
228 bool sawDigit = false;
229 double number = 0;
230 while (p < length) {
231 int digit = parseDigit(data[p], radix);
232 if (digit == -1)
233 break;
234 sawDigit = true;
235 number *= radix;
236 number += digit;
237 ++p;
238 }
239
240 if (number >= mantissaOverflowLowerBound) {
241 if (radix == 10)
242 number = WTF::strtod(s.substr(firstDigitPosition, p - firstDigitPosition).ascii(), 0);
243 else if (radix == 2 || radix == 4 || radix == 8 || radix == 16 || radix == 32)
244 number = parseIntOverflow(s.substr(firstDigitPosition, p - firstDigitPosition).ascii(), p - firstDigitPosition, radix);
245 }
246
247 if (!sawDigit)
248 return NaN;
249
250 return sign * number;
251 }
252
parseFloat(const UString & s)253 static double parseFloat(const UString& s)
254 {
255 // Check for 0x prefix here, because toDouble allows it, but we must treat it as 0.
256 // Need to skip any whitespace and then one + or - sign.
257 int length = s.size();
258 const UChar* data = s.data();
259 int p = 0;
260 while (p < length && isStrWhiteSpace(data[p]))
261 ++p;
262
263 if (p < length && (data[p] == '+' || data[p] == '-'))
264 ++p;
265
266 if (length - p >= 2 && data[p] == '0' && (data[p + 1] == 'x' || data[p + 1] == 'X'))
267 return 0;
268
269 return s.toDouble(true /*tolerant*/, false /* NaN for empty string */);
270 }
271
globalFuncEval(ExecState * exec,JSObject * function,JSValue thisValue,const ArgList & args)272 JSValue JSC_HOST_CALL globalFuncEval(ExecState* exec, JSObject* function, JSValue thisValue, const ArgList& args)
273 {
274 JSObject* thisObject = thisValue.toThisObject(exec);
275 JSObject* unwrappedObject = thisObject->unwrappedObject();
276 if (!unwrappedObject->isGlobalObject() || static_cast<JSGlobalObject*>(unwrappedObject)->evalFunction() != function)
277 return throwError(exec, EvalError, "The \"this\" value passed to eval must be the global object from which eval originated");
278
279 JSValue x = args.at(0);
280 if (!x.isString())
281 return x;
282
283 UString s = x.toString(exec);
284
285 LiteralParser preparser(exec, s, LiteralParser::NonStrictJSON);
286 if (JSValue parsedObject = preparser.tryLiteralParse())
287 return parsedObject;
288
289 int errLine;
290 UString errMsg;
291
292 SourceCode source = makeSource(s);
293 RefPtr<EvalNode> evalNode = exec->globalData().parser->parse<EvalNode>(exec, exec->dynamicGlobalObject()->debugger(), source, &errLine, &errMsg);
294
295 if (!evalNode)
296 return throwError(exec, SyntaxError, errMsg, errLine, source.provider()->asID(), NULL);
297
298 return exec->interpreter()->execute(evalNode.get(), exec, thisObject, static_cast<JSGlobalObject*>(unwrappedObject)->globalScopeChain().node(), exec->exceptionSlot());
299 }
300
globalFuncParseInt(ExecState * exec,JSObject *,JSValue,const ArgList & args)301 JSValue JSC_HOST_CALL globalFuncParseInt(ExecState* exec, JSObject*, JSValue, const ArgList& args)
302 {
303 JSValue value = args.at(0);
304 int32_t radix = args.at(1).toInt32(exec);
305
306 if (radix != 0 && radix != 10)
307 return jsNumber(exec, parseInt(value.toString(exec), radix));
308
309 if (value.isInt32())
310 return value;
311
312 if (value.isDouble()) {
313 double d = value.asDouble();
314 if (isfinite(d))
315 return jsNumber(exec, (d > 0) ? floor(d) : ceil(d));
316 if (isnan(d) || isinf(d))
317 return jsNaN(exec);
318 return jsNumber(exec, 0);
319 }
320
321 return jsNumber(exec, parseInt(value.toString(exec), radix));
322 }
323
globalFuncParseFloat(ExecState * exec,JSObject *,JSValue,const ArgList & args)324 JSValue JSC_HOST_CALL globalFuncParseFloat(ExecState* exec, JSObject*, JSValue, const ArgList& args)
325 {
326 return jsNumber(exec, parseFloat(args.at(0).toString(exec)));
327 }
328
globalFuncIsNaN(ExecState * exec,JSObject *,JSValue,const ArgList & args)329 JSValue JSC_HOST_CALL globalFuncIsNaN(ExecState* exec, JSObject*, JSValue, const ArgList& args)
330 {
331 return jsBoolean(isnan(args.at(0).toNumber(exec)));
332 }
333
globalFuncIsFinite(ExecState * exec,JSObject *,JSValue,const ArgList & args)334 JSValue JSC_HOST_CALL globalFuncIsFinite(ExecState* exec, JSObject*, JSValue, const ArgList& args)
335 {
336 double n = args.at(0).toNumber(exec);
337 return jsBoolean(!isnan(n) && !isinf(n));
338 }
339
globalFuncDecodeURI(ExecState * exec,JSObject *,JSValue,const ArgList & args)340 JSValue JSC_HOST_CALL globalFuncDecodeURI(ExecState* exec, JSObject*, JSValue, const ArgList& args)
341 {
342 static const char do_not_unescape_when_decoding_URI[] =
343 "#$&+,/:;=?@";
344
345 return decode(exec, args, do_not_unescape_when_decoding_URI, true);
346 }
347
globalFuncDecodeURIComponent(ExecState * exec,JSObject *,JSValue,const ArgList & args)348 JSValue JSC_HOST_CALL globalFuncDecodeURIComponent(ExecState* exec, JSObject*, JSValue, const ArgList& args)
349 {
350 return decode(exec, args, "", true);
351 }
352
globalFuncEncodeURI(ExecState * exec,JSObject *,JSValue,const ArgList & args)353 JSValue JSC_HOST_CALL globalFuncEncodeURI(ExecState* exec, JSObject*, JSValue, const ArgList& args)
354 {
355 static const char do_not_escape_when_encoding_URI[] =
356 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
357 "abcdefghijklmnopqrstuvwxyz"
358 "0123456789"
359 "!#$&'()*+,-./:;=?@_~";
360
361 return encode(exec, args, do_not_escape_when_encoding_URI);
362 }
363
globalFuncEncodeURIComponent(ExecState * exec,JSObject *,JSValue,const ArgList & args)364 JSValue JSC_HOST_CALL globalFuncEncodeURIComponent(ExecState* exec, JSObject*, JSValue, const ArgList& args)
365 {
366 static const char do_not_escape_when_encoding_URI_component[] =
367 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
368 "abcdefghijklmnopqrstuvwxyz"
369 "0123456789"
370 "!'()*-._~";
371
372 return encode(exec, args, do_not_escape_when_encoding_URI_component);
373 }
374
globalFuncEscape(ExecState * exec,JSObject *,JSValue,const ArgList & args)375 JSValue JSC_HOST_CALL globalFuncEscape(ExecState* exec, JSObject*, JSValue, const ArgList& args)
376 {
377 static const char do_not_escape[] =
378 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
379 "abcdefghijklmnopqrstuvwxyz"
380 "0123456789"
381 "*+-./@_";
382
383 UString result = "";
384 UString s;
385 UString str = args.at(0).toString(exec);
386 const UChar* c = str.data();
387 for (int k = 0; k < str.size(); k++, c++) {
388 int u = c[0];
389 if (u > 255) {
390 char tmp[7];
391 sprintf(tmp, "%%u%04X", u);
392 s = UString(tmp);
393 } else if (u != 0 && strchr(do_not_escape, static_cast<char>(u)))
394 s = UString(c, 1);
395 else {
396 char tmp[4];
397 sprintf(tmp, "%%%02X", u);
398 s = UString(tmp);
399 }
400 result += s;
401 }
402
403 return jsString(exec, result);
404 }
405
globalFuncUnescape(ExecState * exec,JSObject *,JSValue,const ArgList & args)406 JSValue JSC_HOST_CALL globalFuncUnescape(ExecState* exec, JSObject*, JSValue, const ArgList& args)
407 {
408 UString result = "";
409 UString str = args.at(0).toString(exec);
410 int k = 0;
411 int len = str.size();
412 while (k < len) {
413 const UChar* c = str.data() + k;
414 UChar u;
415 if (c[0] == '%' && k <= len - 6 && c[1] == 'u') {
416 if (isASCIIHexDigit(c[2]) && isASCIIHexDigit(c[3]) && isASCIIHexDigit(c[4]) && isASCIIHexDigit(c[5])) {
417 u = Lexer::convertUnicode(c[2], c[3], c[4], c[5]);
418 c = &u;
419 k += 5;
420 }
421 } else if (c[0] == '%' && k <= len - 3 && isASCIIHexDigit(c[1]) && isASCIIHexDigit(c[2])) {
422 u = UChar(Lexer::convertHex(c[1], c[2]));
423 c = &u;
424 k += 2;
425 }
426 k++;
427 result.append(*c);
428 }
429
430 return jsString(exec, result);
431 }
432
433 #ifndef NDEBUG
globalFuncJSCPrint(ExecState * exec,JSObject *,JSValue,const ArgList & args)434 JSValue JSC_HOST_CALL globalFuncJSCPrint(ExecState* exec, JSObject*, JSValue, const ArgList& args)
435 {
436 CStringBuffer string;
437 args.at(0).toString(exec).getCString(string);
438 puts(string.data());
439 return jsUndefined();
440 }
441 #endif
442
443 } // namespace JSC
444