kurl.cpp

linux下开源浏览器WebKit的源码,市面上的很多商用浏览器都是移植自WebKit

/*
 * Copyright (C) 2004, 2007, 2008 Apple Inc. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE COMPUTER, INC. OR
 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 
 */

#include "config.h"

#if !USE(GOOGLEURL)

#include "KURL.h"

#include "CString.h"
#include "PlatformString.h"
#include "TextEncoding.h"
#include <wtf/StdLibExtras.h>

#if USE(ICU_UNICODE)
#include <unicode/uidna.h>
#elif USE(QT4_UNICODE)
#include <QUrl>
#endif

#include <stdio.h>

using namespace std;
using namespace WTF;

namespace WebCore {

typedef Vector<char, 512> CharBuffer;
typedef Vector<UChar, 512> UCharBuffer;

// FIXME: This file makes too much use of the + operator on String.
// We either have to optimize that operator so it doesn't involve
// so many allocations, or change this to use Vector<UChar> instead.

enum URLCharacterClasses {
    // alpha 
    SchemeFirstChar = 1 << 0,

    // ( alpha | digit | "+" | "-" | "." )
    SchemeChar = 1 << 1,

    // mark        = "-" | "_" | "." | "!" | "~" | "*" | "'" | "(" | ")"
    // unreserved  = alphanum | mark
    // ( unreserved | escaped | ";" | ":" | "&" | "=" | "+" | "$" | "," )
    UserInfoChar = 1 << 2,

    // alnum | "." | "-" | "%"
    // The above is what the specification says, but we are lenient to
    // match existing practice and also allow:
    // "_"
    HostnameChar = 1 << 3,

    // hexdigit | ":" | "%"
    IPv6Char = 1 << 4,

    // "#" | "?" | "/" | nul
    PathSegmentEndChar = 1 << 5,

    // not allowed in path
    BadChar = 1 << 6
};

static const char hexDigits[17] = "0123456789ABCDEF";

static const unsigned char characterClassTable[256] = {
    /* 0 nul */ PathSegmentEndChar,    /* 1 soh */ BadChar,
    /* 2 stx */ BadChar,    /* 3 etx */ BadChar,
    /* 4 eot */ BadChar,    /* 5 enq */ BadChar,    /* 6 ack */ BadChar,    /* 7 bel */ BadChar,
    /* 8 bs */ BadChar,     /* 9 ht */ BadChar,     /* 10 nl */ BadChar,    /* 11 vt */ BadChar,
    /* 12 np */ BadChar,    /* 13 cr */ BadChar,    /* 14 so */ BadChar,    /* 15 si */ BadChar,
    /* 16 dle */ BadChar,   /* 17 dc1 */ BadChar,   /* 18 dc2 */ BadChar,   /* 19 dc3 */ BadChar,
    /* 20 dc4 */ BadChar,   /* 21 nak */ BadChar,   /* 22 syn */ BadChar,   /* 23 etb */ BadChar,
    /* 24 can */ BadChar,   /* 25 em */ BadChar,    /* 26 sub */ BadChar,   /* 27 esc */ BadChar,
    /* 28 fs */ BadChar,    /* 29 gs */ BadChar,    /* 30 rs */ BadChar,    /* 31 us */ BadChar,
    /* 32 sp */ BadChar,    /* 33  ! */ UserInfoChar,
    /* 34  " */ BadChar,    /* 35  # */ PathSegmentEndChar | BadChar,
    /* 36  $ */ UserInfoChar,    /* 37  % */ UserInfoChar | HostnameChar | IPv6Char | BadChar,
    /* 38  & */ UserInfoChar,    /* 39  ' */ UserInfoChar,
    /* 40  ( */ UserInfoChar,    /* 41  ) */ UserInfoChar,
    /* 42  * */ UserInfoChar,    /* 43  + */ SchemeChar | UserInfoChar,
    /* 44  , */ UserInfoChar,
    /* 45  - */ SchemeChar | UserInfoChar | HostnameChar,
    /* 46  . */ SchemeChar | UserInfoChar | HostnameChar,
    /* 47  / */ PathSegmentEndChar,
    /* 48  0 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 49  1 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char,    
    /* 50  2 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 51  3 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 52  4 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 53  5 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 54  6 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 55  7 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 56  8 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 57  9 */ SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 58  : */ UserInfoChar | IPv6Char,    /* 59  ; */ UserInfoChar,
    /* 60  < */ BadChar,    /* 61  = */ UserInfoChar,
    /* 62  > */ BadChar,    /* 63  ? */ PathSegmentEndChar | BadChar,
    /* 64  @ */ 0,
    /* 65  A */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,    
    /* 66  B */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 67  C */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 68  D */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 69  E */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 70  F */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 71  G */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 72  H */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 73  I */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 74  J */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 75  K */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 76  L */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 77  M */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 78  N */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 79  O */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 80  P */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 81  Q */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 82  R */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 83  S */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 84  T */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 85  U */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 86  V */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 87  W */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 88  X */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 89  Y */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 90  Z */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 91  [ */ 0,
    /* 92  \ */ 0,    /* 93  ] */ 0,
    /* 94  ^ */ 0,
    /* 95  _ */ UserInfoChar | HostnameChar,
    /* 96  ` */ 0,
    /* 97  a */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 98  b */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 99  c */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 100  d */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 101  e */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char,
    /* 102  f */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar | IPv6Char, 
    /* 103  g */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 104  h */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 105  i */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 106  j */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 107  k */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 108  l */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 109  m */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 110  n */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 111  o */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 112  p */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 113  q */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 114  r */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 115  s */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 116  t */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 117  u */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 118  v */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 119  w */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 120  x */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 121  y */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar,
    /* 122  z */ SchemeFirstChar | SchemeChar | UserInfoChar | HostnameChar, 
    /* 123  { */ 0,
    /* 124  | */ 0,   /* 125  } */ 0,   /* 126  ~ */ UserInfoChar,   /* 127 del */ BadChar,
    /* 128 */ BadChar, /* 129 */ BadChar, /* 130 */ BadChar, /* 131 */ BadChar,
    /* 132 */ BadChar, /* 133 */ BadChar, /* 134 */ BadChar, /* 135 */ BadChar,
    /* 136 */ BadChar, /* 137 */ BadChar, /* 138 */ BadChar, /* 139 */ BadChar,
    /* 140 */ BadChar, /* 141 */ BadChar, /* 142 */ BadChar, /* 143 */ BadChar,
    /* 144 */ BadChar, /* 145 */ BadChar, /* 146 */ BadChar, /* 147 */ BadChar,
    /* 148 */ BadChar, /* 149 */ BadChar, /* 150 */ BadChar, /* 151 */ BadChar,
    /* 152 */ BadChar, /* 153 */ BadChar, /* 154 */ BadChar, /* 155 */ BadChar,
    /* 156 */ BadChar, /* 157 */ BadChar, /* 158 */ BadChar, /* 159 */ BadChar,
    /* 160 */ BadChar, /* 161 */ BadChar, /* 162 */ BadChar, /* 163 */ BadChar,
    /* 164 */ BadChar, /* 165 */ BadChar, /* 166 */ BadChar, /* 167 */ BadChar,
    /* 168 */ BadChar, /* 169 */ BadChar, /* 170 */ BadChar, /* 171 */ BadChar,
    /* 172 */ BadChar, /* 173 */ BadChar, /* 174 */ BadChar, /* 175 */ BadChar,
    /* 176 */ BadChar, /* 177 */ BadChar, /* 178 */ BadChar, /* 179 */ BadChar,
    /* 180 */ BadChar, /* 181 */ BadChar, /* 182 */ BadChar, /* 183 */ BadChar,
    /* 184 */ BadChar, /* 185 */ BadChar, /* 186 */ BadChar, /* 187 */ BadChar,
    /* 188 */ BadChar, /* 189 */ BadChar, /* 190 */ BadChar, /* 191 */ BadChar,
    /* 192 */ BadChar, /* 193 */ BadChar, /* 194 */ BadChar, /* 195 */ BadChar,
    /* 196 */ BadChar, /* 197 */ BadChar, /* 198 */ BadChar, /* 199 */ BadChar,
    /* 200 */ BadChar, /* 201 */ BadChar, /* 202 */ BadChar, /* 203 */ BadChar,
    /* 204 */ BadChar, /* 205 */ BadChar, /* 206 */ BadChar, /* 207 */ BadChar,
    /* 208 */ BadChar, /* 209 */ BadChar, /* 210 */ BadChar, /* 211 */ BadChar,
    /* 212 */ BadChar, /* 213 */ BadChar, /* 214 */ BadChar, /* 215 */ BadChar,
    /* 216 */ BadChar, /* 217 */ BadChar, /* 218 */ BadChar, /* 219 */ BadChar,
    /* 220 */ BadChar, /* 221 */ BadChar, /* 222 */ BadChar, /* 223 */ BadChar,
    /* 224 */ BadChar, /* 225 */ BadChar, /* 226 */ BadChar, /* 227 */ BadChar,
    /* 228 */ BadChar, /* 229 */ BadChar, /* 230 */ BadChar, /* 231 */ BadChar,
    /* 232 */ BadChar, /* 233 */ BadChar, /* 234 */ BadChar, /* 235 */ BadChar,
    /* 236 */ BadChar, /* 237 */ BadChar, /* 238 */ BadChar, /* 239 */ BadChar,
    /* 240 */ BadChar, /* 241 */ BadChar, /* 242 */ BadChar, /* 243 */ BadChar,
    /* 244 */ BadChar, /* 245 */ BadChar, /* 246 */ BadChar, /* 247 */ BadChar,
    /* 248 */ BadChar, /* 249 */ BadChar, /* 250 */ BadChar, /* 251 */ BadChar,
    /* 252 */ BadChar, /* 253 */ BadChar, /* 254 */ BadChar, /* 255 */ BadChar
};

static int copyPathRemovingDots(char* dst, const char* src, int srcStart, int srcEnd);
static void encodeRelativeString(const String& rel, const TextEncoding&, CharBuffer& ouput);
static String substituteBackslashes(const String&);

static inline bool isSchemeFirstChar(char c) { return characterClassTable[static_cast<unsigned char>(c)] & SchemeFirstChar; }
static inline bool isSchemeFirstChar(UChar c) { return c <= 0xff && (characterClassTable[c] & SchemeFirstChar); }
static inline bool isSchemeChar(char c) { return characterClassTable[static_cast<unsigned char>(c)] & SchemeChar; }
static inline bool isSchemeChar(UChar c) { return c <= 0xff && (characterClassTable[c] & SchemeChar); }
static inline bool isUserInfoChar(unsigned char c) { return characterClassTable[c] & UserInfoChar; }
static inline bool isHostnameChar(unsigned char c) { return characterClassTable[c] & HostnameChar; }
static inline bool isIPv6Char(unsigned char c) { return characterClassTable[c] & IPv6Char; }
static inline bool isPathSegmentEndChar(char c) { return characterClassTable[static_cast<unsigned char>(c)] & PathSegmentEndChar; }
static inline bool isPathSegmentEndChar(UChar c) { return c <= 0xff && (characterClassTable[c] & PathSegmentEndChar); }
static inline bool isBadChar(unsigned char c) { return characterClassTable[c] & BadChar; }

static inline int hexDigitValue(UChar c)
{
    ASSERT(isASCIIHexDigit(c));
    if (c < 'A')
        return c - '0';
    return (c - 'A' + 10) & 0xF; // handle both upper and lower case without a branch
}

// Copies the source to the destination, assuming all the source characters are
// ASCII. The destination buffer must be large enough. Null characters are allowed
// in the source string, and no attempt is made to null-terminate the result.
static void copyASCII(const UChar* src, int length, char* dest)
{
    for (int i = 0; i < length; i++)
        dest[i] = static_cast<char>(src[i]);
}

static void appendASCII(const String& base, const char* rel, size_t len, CharBuffer& buffer)
{
    buffer.resize(base.length() + len + 1);
    copyASCII(base.characters(), base.length(), buffer.data());
    memcpy(buffer.data() + base.length(), rel, len);
    buffer[buffer.size() - 1] = '\0';
}

// FIXME: Move to PlatformString.h eventually.
// Returns the index of the first index in string |s| of any of the characters
// in |toFind|. |toFind| should be a null-terminated string, all characters up
// to the null will be searched. Returns int if not found.
static int findFirstOf(const UChar* s, int sLen, int startPos, const char* toFind)
{
    for (int i = startPos; i < sLen; i++) {
        const char* cur = toFind;
        while (*cur) {
            if (s[i] == *(cur++))
                return i;
        }
    }
    return -1;
}

#ifndef NDEBUG
static void checkEncodedString(const String& url)
{
    for (unsigned i = 0; i < url.length(); ++i)
        ASSERT(!(url[i] & ~0x7F));

    ASSERT(!url.length() || isSchemeFirstChar(url[0]));
}
#else
static inline void checkEncodedString(const String&)
{
}
#endif

inline bool KURL::protocolIs(const String& string, const char* protocol)
{
    return WebCore::protocolIs(string, protocol);
}

void KURL::invalidate()
{
    m_isValid = false;
    m_protocolInHTTPFamily = false;
    m_schemeEnd = 0;
    m_userStart = 0;
    m_userEnd = 0;
    m_passwordEnd = 0;
    m_hostEnd = 0;
    m_portEnd = 0;
    m_pathEnd = 0;
    m_pathAfterLastSlash = 0;
    m_queryEnd = 0;
    m_fragmentEnd = 0;
}

KURL::KURL(const char* url)
{
    parse(url, 0);
    ASSERT(url == m_string);
}

KURL::KURL(const String& url)
{
    checkEncodedString(url);

    parse(url);
    ASSERT(url == m_string);
}

KURL::KURL(const KURL& base, const String& relative)
{
    init(base, relative, UTF8Encoding());
}

KURL::KURL(const KURL& base, const String& relative, const TextEncoding& encoding)
{
    // For UTF-{7,16,32}, we want to use UTF-8 for the query part as 
    // we do when submitting a form. A form with GET method
    // has its contents added to a URL as query params and it makes sense
    // to be consistent.
    init(base, relative, encoding.encodingForFormSubmission());
}

void KURL::init(const KURL& base, const String& relative, const TextEncoding& encoding)
{
    // Allow resolutions with a null or empty base URL, but not with any other invalid one.
    // FIXME: Is this a good rule?
    if (!base.m_isValid && !base.isEmpty()) {
        m_string = relative;
        invalidate();
        return;
    }

    // For compatibility with Win IE, treat backslashes as if they were slashes,
    // as long as we're not dealing with javascript: or data: URLs.
    String rel = relative;
    if (rel.contains('\\') && !(protocolIs(rel, "javascript") || protocolIs(rel, "data")))
        rel = substituteBackslashes(rel);

    String* originalString = &rel;

    bool allASCII = charactersAreAllASCII(rel.characters(), rel.length());
    CharBuffer strBuffer;
    char* str;
    size_t len;
    if (allASCII) {
        len = rel.length();
        strBuffer.resize(len + 1);
        copyASCII(rel.characters(), len, strBuffer.data());
        strBuffer[len] = 0;
        str = strBuffer.data();
    } else {
        originalString = 0;
        encodeRelativeString(rel, encoding, strBuffer);
        str = strBuffer.data();
        len = strlen(str);
    }

    // Get rid of leading whitespace.
    while (*str == ' ') {
        originalString = 0;
        str++;
        --len;
    }

    // Get rid of trailing whitespace.
    while (len && str[len - 1] == ' ') {
        originalString = 0;
        str[--len] = '\0';
    }

    // According to the RFC, the reference should be interpreted as an
    // absolute URI if possible, using the "leftmost, longest"
    // algorithm. If the URI reference is absolute it will have a
    // scheme, meaning that it will have a colon before the first
    // non-scheme element.
    bool absolute = false;
    char* p = str;
    if (isSchemeFirstChar(*p)) {
        ++p;
        while (isSchemeChar(*p)) {
            ++p;
        }
        if (*p == ':') {
            if (p[1] != '/' && equalIgnoringCase(base.protocol(), String(str, p - str)) && base.isHierarchical()) {
                str = p + 1;
                originalString = 0;
            } else
                absolute = true;
        }
    }

    CharBuffer parseBuffer;

    if (absolute) {
        parse(str, originalString);
    } else {
        // If the base is empty or opaque (e.g. data: or javascript:), then the URL is invalid
        // unless the relative URL is a single fragment.
        if (!base.isHierarchical()) {
            if (str[0] == '#') {
                appendASCII(base.m_string.left(base.m_queryEnd), str, len, parseBuffer);