1/*
2 * Copyright (C) 2008, 2010, 2013 Apple Inc. All Rights Reserved.
3 * Copyright (C) 2009 Torch Mobile, Inc. http://www.torchmobile.com/
4 * Copyright (C) 2010 Google Inc. All Rights Reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
8 * are met:
9 * 1. Redistributions of source code must retain the above copyright
10 *    notice, this list of conditions and the following disclaimer.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 *    notice, this list of conditions and the following disclaimer in the
13 *    documentation and/or other materials provided with the distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
16 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
18 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
19 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
20 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
21 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
22 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
23 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
25 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26 */
27
28#include "config.h"
29#include "CSSPreloadScanner.h"
30
31#include "CachedResourceRequestInitiators.h"
32#include "HTMLIdentifier.h"
33#include "HTMLParserIdioms.h"
34
35namespace WebCore {
36
37CSSPreloadScanner::CSSPreloadScanner()
38    : m_state(Initial)
39    , m_requests(0)
40{
41}
42
43CSSPreloadScanner::~CSSPreloadScanner()
44{
45}
46
47void CSSPreloadScanner::reset()
48{
49    m_state = Initial;
50    m_rule.clear();
51    m_ruleValue.clear();
52}
53
54template<typename Char>
55void CSSPreloadScanner::scanCommon(const Char* begin, const Char* end, PreloadRequestStream& requests)
56{
57    m_requests = &requests;
58    for (const Char* it = begin; it != end && m_state != DoneParsingImportRules; ++it)
59        tokenize(*it);
60    m_requests = 0;
61}
62
63void CSSPreloadScanner::scan(const HTMLToken::DataVector& data, PreloadRequestStream& requests)
64{
65    scanCommon(data.data(), data.data() + data.size(), requests);
66}
67
68#if ENABLE(THREADED_HTML_PARSER)
69void CSSPreloadScanner::scan(const HTMLIdentifier& identifier, PreloadRequestStream& requests)
70{
71    const StringImpl* data = identifier.asStringImpl();
72    if (data->is8Bit()) {
73        const LChar* begin = data->characters8();
74        scanCommon(begin, begin + data->length(), requests);
75        return;
76    }
77    const UChar* begin = data->characters16();
78    scanCommon(begin, begin + data->length(), requests);
79}
80#endif
81
82inline void CSSPreloadScanner::tokenize(UChar c)
83{
84    // We are just interested in @import rules, no need for real tokenization here
85    // Searching for other types of resources is probably low payoff.
86    switch (m_state) {
87    case Initial:
88        if (isHTMLSpace(c))
89            break;
90        if (c == '@')
91            m_state = RuleStart;
92        else if (c == '/')
93            m_state = MaybeComment;
94        else
95            m_state = DoneParsingImportRules;
96        break;
97    case MaybeComment:
98        if (c == '*')
99            m_state = Comment;
100        else
101            m_state = Initial;
102        break;
103    case Comment:
104        if (c == '*')
105            m_state = MaybeCommentEnd;
106        break;
107    case MaybeCommentEnd:
108        if (c == '*')
109            break;
110        if (c == '/')
111            m_state = Initial;
112        else
113            m_state = Comment;
114        break;
115    case RuleStart:
116        if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')) {
117            m_rule.clear();
118            m_ruleValue.clear();
119            m_rule.append(c);
120            m_state = Rule;
121        } else
122            m_state = Initial;
123        break;
124    case Rule:
125        if (isHTMLSpace(c))
126            m_state = AfterRule;
127        else if (c == ';')
128            m_state = Initial;
129        else
130            m_rule.append(c);
131        break;
132    case AfterRule:
133        if (isHTMLSpace(c))
134            break;
135        if (c == ';')
136            m_state = Initial;
137        else if (c == '{')
138            m_state = DoneParsingImportRules;
139        else {
140            m_state = RuleValue;
141            m_ruleValue.append(c);
142        }
143        break;
144    case RuleValue:
145        if (isHTMLSpace(c))
146            m_state = AfterRuleValue;
147        else if (c == ';')
148            emitRule();
149        else
150            m_ruleValue.append(c);
151        break;
152    case AfterRuleValue:
153        if (isHTMLSpace(c))
154            break;
155        if (c == ';')
156            emitRule();
157        else if (c == '{')
158            m_state = DoneParsingImportRules;
159        else {
160            // FIXME: media rules
161            m_state = Initial;
162        }
163        break;
164    case DoneParsingImportRules:
165        ASSERT_NOT_REACHED();
166        break;
167    }
168}
169
170static String parseCSSStringOrURL(const UChar* characters, size_t length)
171{
172    size_t offset = 0;
173    size_t reducedLength = length;
174
175    while (reducedLength && isHTMLSpace(characters[offset])) {
176        ++offset;
177        --reducedLength;
178    }
179    while (reducedLength && isHTMLSpace(characters[offset + reducedLength - 1]))
180        --reducedLength;
181
182    if (reducedLength >= 5
183            && (characters[offset] == 'u' || characters[offset] == 'U')
184            && (characters[offset + 1] == 'r' || characters[offset + 1] == 'R')
185            && (characters[offset + 2] == 'l' || characters[offset + 2] == 'L')
186            && characters[offset + 3] == '('
187            && characters[offset + reducedLength - 1] == ')') {
188        offset += 4;
189        reducedLength -= 5;
190    }
191
192    while (reducedLength && isHTMLSpace(characters[offset])) {
193        ++offset;
194        --reducedLength;
195    }
196    while (reducedLength && isHTMLSpace(characters[offset + reducedLength - 1]))
197        --reducedLength;
198
199    if (reducedLength < 2 || characters[offset] != characters[offset + reducedLength - 1] || !(characters[offset] == '\'' || characters[offset] == '"'))
200        return String();
201    offset++;
202    reducedLength -= 2;
203
204    while (reducedLength && isHTMLSpace(characters[offset])) {
205        ++offset;
206        --reducedLength;
207    }
208    while (reducedLength && isHTMLSpace(characters[offset + reducedLength - 1]))
209        --reducedLength;
210
211    return String(characters + offset, reducedLength);
212}
213
214template<unsigned referenceLength>
215static inline bool ruleEqualIgnoringCase(const Vector<UChar>& rule, const char (&reference)[referenceLength])
216{
217    unsigned referenceCharactersLength = referenceLength - 1;
218    return rule.size() == referenceCharactersLength && equalIgnoringCase(reference, rule.data(), referenceCharactersLength);
219}
220
221void CSSPreloadScanner::emitRule()
222{
223    if (ruleEqualIgnoringCase(m_rule, "import")) {
224        String url = parseCSSStringOrURL(m_ruleValue.data(), m_ruleValue.size());
225        if (!url.isEmpty()) {
226            KURL baseElementURL; // FIXME: This should be passed in from the HTMLPreloadScaner via scan()!
227            OwnPtr<PreloadRequest> request = PreloadRequest::create("css", url, baseElementURL, CachedResource::CSSStyleSheet);
228            // FIXME: Should this be including the charset in the preload request?
229            m_requests->append(request.release());
230        }
231        m_state = Initial;
232    } else if (ruleEqualIgnoringCase(m_rule, "charset"))
233        m_state = Initial;
234    else
235        m_state = DoneParsingImportRules;
236    m_rule.clear();
237    m_ruleValue.clear();
238}
239
240}
241