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 "JSString.h"
32 #include "Interpreter.h"
33 #include "Parser.h"
34 #include "dtoa.h"
35 #include "Lexer.h"
36 #include "Nodes.h"
37 #include <stdio.h>
38 #include <stdlib.h>
39 #include <string.h>
40 #include <wtf/ASCIICType.h>
41 #include <wtf/Assertions.h>
42 #include <wtf/MathExtras.h>
43 #include <wtf/unicode/UTF8.h>
44
45 using namespace WTF;
46 using namespace Unicode;
47
48 namespace JSC {
49
encode(ExecState * exec,const ArgList & args,const char * doNotEscape)50 static JSValuePtr encode(ExecState* exec, const ArgList& args, const char* doNotEscape)
51 {
52 UString str = args.at(exec, 0).toString(exec);
53 CString cstr = str.UTF8String(true);
54 if (!cstr.c_str())
55 return throwError(exec, URIError, "String contained an illegal UTF-16 sequence.");
56
57 UString result = "";
58 const char* p = cstr.c_str();
59 for (size_t k = 0; k < cstr.size(); k++, p++) {
60 char c = *p;
61 if (c && strchr(doNotEscape, c))
62 result.append(c);
63 else {
64 char tmp[4];
65 sprintf(tmp, "%%%02X", static_cast<unsigned char>(c));
66 result += tmp;
67 }
68 }
69 return jsString(exec, result);
70 }
71
decode(ExecState * exec,const ArgList & args,const char * doNotUnescape,bool strict)72 static JSValuePtr decode(ExecState* exec, const ArgList& args, const char* doNotUnescape, bool strict)
73 {
74 UString result = "";
75 UString str = args.at(exec, 0).toString(exec);
76 int k = 0;
77 int len = str.size();
78 const UChar* d = str.data();
79 UChar u = 0;
80 while (k < len) {
81 const UChar* p = d + k;
82 UChar c = *p;
83 if (c == '%') {
84 int charLen = 0;
85 if (k <= len - 3 && isASCIIHexDigit(p[1]) && isASCIIHexDigit(p[2])) {
86 const char b0 = Lexer::convertHex(p[1], p[2]);
87 const int sequenceLen = UTF8SequenceLength(b0);
88 if (sequenceLen != 0 && k <= len - sequenceLen * 3) {
89 charLen = sequenceLen * 3;
90 char sequence[5];
91 sequence[0] = b0;
92 for (int i = 1; i < sequenceLen; ++i) {
93 const UChar* q = p + i * 3;
94 if (q[0] == '%' && isASCIIHexDigit(q[1]) && isASCIIHexDigit(q[2]))
95 sequence[i] = Lexer::convertHex(q[1], q[2]);
96 else {
97 charLen = 0;
98 break;
99 }
100 }
101 if (charLen != 0) {
102 sequence[sequenceLen] = 0;
103 const int character = decodeUTF8Sequence(sequence);
104 if (character < 0 || character >= 0x110000)
105 charLen = 0;
106 else if (character >= 0x10000) {
107 // Convert to surrogate pair.
108 result.append(static_cast<UChar>(0xD800 | ((character - 0x10000) >> 10)));
109 u = static_cast<UChar>(0xDC00 | ((character - 0x10000) & 0x3FF));
110 } else
111 u = static_cast<UChar>(character);
112 }
113 }
114 }
115 if (charLen == 0) {
116 if (strict)
117 return throwError(exec, URIError);
118 // The only case where we don't use "strict" mode is the "unescape" function.
119 // For that, it's good to support the wonky "%u" syntax for compatibility with WinIE.
120 if (k <= len - 6 && p[1] == 'u'
121 && isASCIIHexDigit(p[2]) && isASCIIHexDigit(p[3])
122 && isASCIIHexDigit(p[4]) && isASCIIHexDigit(p[5])) {
123 charLen = 6;
124 u = Lexer::convertUnicode(p[2], p[3], p[4], p[5]);
125 }
126 }
127 if (charLen && (u == 0 || u >= 128 || !strchr(doNotUnescape, u))) {
128 c = u;
129 k += charLen - 1;
130 }
131 }
132 k++;
133 result.append(c);
134 }
135 return jsString(exec, result);
136 }
137
isStrWhiteSpace(UChar c)138 bool isStrWhiteSpace(UChar c)
139 {
140 switch (c) {
141 case 0x0009:
142 case 0x000A:
143 case 0x000B:
144 case 0x000C:
145 case 0x000D:
146 case 0x0020:
147 case 0x00A0:
148 case 0x2028:
149 case 0x2029:
150 return true;
151 default:
152 return c > 0xff && isSeparatorSpace(c);
153 }
154 }
155
parseDigit(unsigned short c,int radix)156 static int parseDigit(unsigned short c, int radix)
157 {
158 int digit = -1;
159
160 if (c >= '0' && c <= '9')
161 digit = c - '0';
162 else if (c >= 'A' && c <= 'Z')
163 digit = c - 'A' + 10;
164 else if (c >= 'a' && c <= 'z')
165 digit = c - 'a' + 10;
166
167 if (digit >= radix)
168 return -1;
169 return digit;
170 }
171
parseIntOverflow(const char * s,int length,int radix)172 double parseIntOverflow(const char* s, int length, int radix)
173 {
174 double number = 0.0;
175 double radixMultiplier = 1.0;
176
177 for (const char* p = s + length - 1; p >= s; p--) {
178 if (radixMultiplier == Inf) {
179 if (*p != '0') {
180 number = Inf;
181 break;
182 }
183 } else {
184 int digit = parseDigit(*p, radix);
185 number += digit * radixMultiplier;
186 }
187
188 radixMultiplier *= radix;
189 }
190
191 return number;
192 }
193
parseInt(const UString & s,int radix)194 static double parseInt(const UString& s, int radix)
195 {
196 int length = s.size();
197 const UChar* data = s.data();
198 int p = 0;
199
200 while (p < length && isStrWhiteSpace(data[p]))
201 ++p;
202
203 double sign = 1;
204 if (p < length) {
205 if (data[p] == '+')
206 ++p;
207 else if (data[p] == '-') {
208 sign = -1;
209 ++p;
210 }
211 }
212
213 if ((radix == 0 || radix == 16) && length - p >= 2 && data[p] == '0' && (data[p + 1] == 'x' || data[p + 1] == 'X')) {
214 radix = 16;
215 p += 2;
216 } else if (radix == 0) {
217 if (p < length && data[p] == '0')
218 radix = 8;
219 else
220 radix = 10;
221 }
222
223 if (radix < 2 || radix > 36)
224 return NaN;
225
226 int firstDigitPosition = p;
227 bool sawDigit = false;
228 double number = 0;
229 while (p < length) {
230 int digit = parseDigit(data[p], radix);
231 if (digit == -1)
232 break;
233 sawDigit = true;
234 number *= radix;
235 number += digit;
236 ++p;
237 }
238
239 if (number >= mantissaOverflowLowerBound) {
240 if (radix == 10)
241 number = WTF::strtod(s.substr(firstDigitPosition, p - firstDigitPosition).ascii(), 0);
242 else if (radix == 2 || radix == 4 || radix == 8 || radix == 16 || radix == 32)
243 number = parseIntOverflow(s.substr(firstDigitPosition, p - firstDigitPosition).ascii(), p - firstDigitPosition, radix);
244 }
245
246 if (!sawDigit)
247 return NaN;
248
249 return sign * number;
250 }
251
parseFloat(const UString & s)252 static double parseFloat(const UString& s)
253 {
254 // Check for 0x prefix here, because toDouble allows it, but we must treat it as 0.
255 // Need to skip any whitespace and then one + or - sign.
256 int length = s.size();
257 const UChar* data = s.data();
258 int p = 0;
259 while (p < length && isStrWhiteSpace(data[p]))
260 ++p;
261
262 if (p < length && (data[p] == '+' || data[p] == '-'))
263 ++p;
264
265 if (length - p >= 2 && data[p] == '0' && (data[p + 1] == 'x' || data[p + 1] == 'X'))
266 return 0;
267
268 return s.toDouble(true /*tolerant*/, false /* NaN for empty string */);
269 }
270
globalFuncEval(ExecState * exec,JSObject * function,JSValuePtr thisValue,const ArgList & args)271 JSValuePtr globalFuncEval(ExecState* exec, JSObject* function, JSValuePtr thisValue, const ArgList& args)
272 {
273 JSObject* thisObject = thisValue.toThisObject(exec);
274 JSObject* unwrappedObject = thisObject->unwrappedObject();
275 if (!unwrappedObject->isGlobalObject() || static_cast<JSGlobalObject*>(unwrappedObject)->evalFunction() != function)
276 return throwError(exec, EvalError, "The \"this\" value passed to eval must be the global object from which eval originated");
277
278 JSValuePtr x = args.at(exec, 0);
279 if (!x.isString())
280 return x;
281
282 UString s = x.toString(exec);
283
284 int errLine;
285 UString errMsg;
286
287 SourceCode source = makeSource(s);
288 RefPtr<EvalNode> evalNode = exec->globalData().parser->parse<EvalNode>(exec, exec->dynamicGlobalObject()->debugger(), source, &errLine, &errMsg);
289
290 if (!evalNode)
291 return throwError(exec, SyntaxError, errMsg, errLine, source.provider()->asID(), NULL);
292
293 return exec->interpreter()->execute(evalNode.get(), exec, thisObject, static_cast<JSGlobalObject*>(unwrappedObject)->globalScopeChain().node(), exec->exceptionSlot());
294 }
295
globalFuncParseInt(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)296 JSValuePtr globalFuncParseInt(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
297 {
298 JSValuePtr value = args.at(exec, 0);
299 int32_t radix = args.at(exec, 1).toInt32(exec);
300
301 if (value.isNumber() && (radix == 0 || radix == 10)) {
302 if (value.isInt32Fast())
303 return value;
304 double d = value.uncheckedGetNumber();
305 if (isfinite(d))
306 return jsNumber(exec, floor(d));
307 if (isnan(d) || isinf(d))
308 return jsNaN(&exec->globalData());
309 return js0();
310 }
311
312 return jsNumber(exec, parseInt(value.toString(exec), radix));
313 }
314
globalFuncParseFloat(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)315 JSValuePtr globalFuncParseFloat(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
316 {
317 return jsNumber(exec, parseFloat(args.at(exec, 0).toString(exec)));
318 }
319
globalFuncIsNaN(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)320 JSValuePtr globalFuncIsNaN(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
321 {
322 return jsBoolean(isnan(args.at(exec, 0).toNumber(exec)));
323 }
324
globalFuncIsFinite(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)325 JSValuePtr globalFuncIsFinite(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
326 {
327 double n = args.at(exec, 0).toNumber(exec);
328 return jsBoolean(!isnan(n) && !isinf(n));
329 }
330
globalFuncDecodeURI(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)331 JSValuePtr globalFuncDecodeURI(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
332 {
333 static const char do_not_unescape_when_decoding_URI[] =
334 "#$&+,/:;=?@";
335
336 return decode(exec, args, do_not_unescape_when_decoding_URI, true);
337 }
338
globalFuncDecodeURIComponent(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)339 JSValuePtr globalFuncDecodeURIComponent(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
340 {
341 return decode(exec, args, "", true);
342 }
343
globalFuncEncodeURI(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)344 JSValuePtr globalFuncEncodeURI(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
345 {
346 static const char do_not_escape_when_encoding_URI[] =
347 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
348 "abcdefghijklmnopqrstuvwxyz"
349 "0123456789"
350 "!#$&'()*+,-./:;=?@_~";
351
352 return encode(exec, args, do_not_escape_when_encoding_URI);
353 }
354
globalFuncEncodeURIComponent(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)355 JSValuePtr globalFuncEncodeURIComponent(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
356 {
357 static const char do_not_escape_when_encoding_URI_component[] =
358 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
359 "abcdefghijklmnopqrstuvwxyz"
360 "0123456789"
361 "!'()*-._~";
362
363 return encode(exec, args, do_not_escape_when_encoding_URI_component);
364 }
365
globalFuncEscape(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)366 JSValuePtr globalFuncEscape(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
367 {
368 static const char do_not_escape[] =
369 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
370 "abcdefghijklmnopqrstuvwxyz"
371 "0123456789"
372 "*+-./@_";
373
374 UString result = "";
375 UString s;
376 UString str = args.at(exec, 0).toString(exec);
377 const UChar* c = str.data();
378 for (int k = 0; k < str.size(); k++, c++) {
379 int u = c[0];
380 if (u > 255) {
381 char tmp[7];
382 sprintf(tmp, "%%u%04X", u);
383 s = UString(tmp);
384 } else if (u != 0 && strchr(do_not_escape, static_cast<char>(u)))
385 s = UString(c, 1);
386 else {
387 char tmp[4];
388 sprintf(tmp, "%%%02X", u);
389 s = UString(tmp);
390 }
391 result += s;
392 }
393
394 return jsString(exec, result);
395 }
396
globalFuncUnescape(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)397 JSValuePtr globalFuncUnescape(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
398 {
399 UString result = "";
400 UString str = args.at(exec, 0).toString(exec);
401 int k = 0;
402 int len = str.size();
403 while (k < len) {
404 const UChar* c = str.data() + k;
405 UChar u;
406 if (c[0] == '%' && k <= len - 6 && c[1] == 'u') {
407 if (Lexer::isHexDigit(c[2]) && Lexer::isHexDigit(c[3]) && Lexer::isHexDigit(c[4]) && Lexer::isHexDigit(c[5])) {
408 u = Lexer::convertUnicode(c[2], c[3], c[4], c[5]);
409 c = &u;
410 k += 5;
411 }
412 } else if (c[0] == '%' && k <= len - 3 && Lexer::isHexDigit(c[1]) && Lexer::isHexDigit(c[2])) {
413 u = UChar(Lexer::convertHex(c[1], c[2]));
414 c = &u;
415 k += 2;
416 }
417 k++;
418 result.append(*c);
419 }
420
421 return jsString(exec, result);
422 }
423
424 #ifndef NDEBUG
globalFuncJSCPrint(ExecState * exec,JSObject *,JSValuePtr,const ArgList & args)425 JSValuePtr globalFuncJSCPrint(ExecState* exec, JSObject*, JSValuePtr, const ArgList& args)
426 {
427 CStringBuffer string;
428 args.at(exec, 0).toString(exec).getCString(string);
429 puts(string.data());
430 return jsUndefined();
431 }
432 #endif
433
434 } // namespace JSC
435