third_party/blink/renderer/core/html/parser/css_preload_scanner.cc - chromium/src - Git at Google

 /*
  * Copyright (C) 2008, 2010 Apple Inc. All Rights Reserved.
  * Copyright (C) 2009 Torch Mobile, Inc. http://www.torchmobile.com/
  * Copyright (C) 2010 Google 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 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 "third_party/blink/renderer/core/html/parser/css_preload_scanner.h"

 #include <memory>

 #include "third_party/blink/renderer/core/html/parser/html_parser_idioms.h"
 #include "third_party/blink/renderer/platform/loader/fetch/fetch_initiator_type_names.h"
 #include "third_party/blink/renderer/platform/network/http_names.h"
 #include "third_party/blink/renderer/platform/text/segmented_string.h"

 namespace blink {

 CSSPreloadScanner::CSSPreloadScanner() = default;

 CSSPreloadScanner::~CSSPreloadScanner() = default;

 void CSSPreloadScanner::Reset() {
   state_ = kInitial;
   rule_.Clear();
   rule_value_.Clear();
 }

 template <typename Char>
 void CSSPreloadScanner::ScanCommon(const Char* begin,
                                    const Char* end,
                                    const SegmentedString& source,
                                    PreloadRequestStream& requests,
                                    const KURL& predicted_base_element_url) {
   requests_ = &requests;
   predicted_base_element_url_ = &predicted_base_element_url;

   for (const Char* it = begin; it != end && state_ != kDoneParsingImportRules;
        ++it)
     Tokenize(*it, source);

   requests_ = nullptr;
   predicted_base_element_url_ = nullptr;
 }

 void CSSPreloadScanner::Scan(const HTMLToken::DataVector& data,
                              const SegmentedString& source,
                              PreloadRequestStream& requests,
                              const KURL& predicted_base_element_url) {
   ScanCommon(data.data(), data.data() + data.size(), source, requests,
              predicted_base_element_url);
 }

 void CSSPreloadScanner::Scan(const String& tag_name,
                              const SegmentedString& source,
                              PreloadRequestStream& requests,
                              const KURL& predicted_base_element_url) {
   if (tag_name.Is8Bit()) {
     const LChar* begin = tag_name.Characters8();
     ScanCommon(begin, begin + tag_name.length(), source, requests,
                predicted_base_element_url);
     return;
   }
   const UChar* begin = tag_name.Characters16();
   ScanCommon(begin, begin + tag_name.length(), source, requests,
              predicted_base_element_url);
 }

 void CSSPreloadScanner::SetReferrerPolicy(const ReferrerPolicy policy) {
   referrer_policy_ = policy;
 }

 inline void CSSPreloadScanner::Tokenize(UChar c,
                                         const SegmentedString& source) {
   // We are just interested in @import rules, no need for real tokenization here
   // Searching for other types of resources is probably low payoff.
   // If we ever decide to preload fonts, we also need to change
   // ResourceFetcher::resourceNeedsLoad to immediately load speculative font
   // preloads.
   switch (state_) {
     case kInitial:
       if (IsHTMLSpace<UChar>(c))
         break;
       if (c == '@')
         state_ = kRuleStart;
       else if (c == '/')
         state_ = kMaybeComment;
       else
         state_ = kDoneParsingImportRules;
       break;
     case kMaybeComment:
       if (c == '*')
         state_ = kComment;
       else
         state_ = kInitial;
       break;
     case kComment:
       if (c == '*')
         state_ = kMaybeCommentEnd;
       break;
     case kMaybeCommentEnd:
       if (c == '*')
         break;
       if (c == '/')
         state_ = kInitial;
       else
         state_ = kComment;
       break;
     case kRuleStart:
       if ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')) {
         rule_.Clear();
         rule_value_.Clear();
         rule_.Append(c);
         state_ = kRule;
       } else
         state_ = kInitial;
       break;
     case kRule:
       if (IsHTMLSpace<UChar>(c))
         state_ = kAfterRule;
       else if (c == ';')
         state_ = kInitial;
       else
         rule_.Append(c);
       break;
     case kAfterRule:
       if (IsHTMLSpace<UChar>(c))
         break;
       if (c == ';')
         state_ = kInitial;
       else if (c == '{')
         state_ = kDoneParsingImportRules;
       else {
         state_ = kRuleValue;
         rule_value_.Append(c);
       }
       break;
     case kRuleValue:
       if (IsHTMLSpace<UChar>(c)) {
         state_ = kAfterRuleValue;
       } else if (c == ';') {
         EmitRule(source);
       } else {
         rule_value_.Append(c);
         // When reading the rule and hitting ')', which signifies the URL end,
         // emit the rule.
         if (c == ')') {
           EmitRule(source);
         }
       }
       break;
     case kAfterRuleValue:
       if (IsHTMLSpace<UChar>(c))
         break;
       if (c == ';')
         EmitRule(source);
       else if (c == '{')
         state_ = kDoneParsingImportRules;
       else {
         // FIXME: media rules
         state_ = kInitial;
       }
       break;
     case kDoneParsingImportRules:
       NOTREACHED();
       break;
   }
 }

 static String ParseCSSStringOrURL(const String& string) {
   wtf_size_t offset = 0;
   wtf_size_t reduced_length = string.length();

   // Remove whitespace from the rule start
   while (reduced_length && IsHTMLSpace<UChar>(string[offset])) {
     ++offset;
     --reduced_length;
   }
   // Remove whitespace from the rule end
   // TODO(yoav): Evaluate performance benefits of using raw string operations.
   // TODO(yoav): Look into moving parsing to use better parsing primitives.
   while (reduced_length &&
          IsHTMLSpace<UChar>(string[offset + reduced_length - 1])) {
     --reduced_length;
   }

   // Skip the "url(" prefix and the ")" suffix
   if (reduced_length >= 5 && (string[offset] == 'u' || string[offset] == 'U') &&
       (string[offset + 1] == 'r' || string[offset + 1] == 'R') &&
       (string[offset + 2] == 'l' || string[offset + 2] == 'L') &&
       string[offset + 3] == '(' && string[offset + reduced_length - 1] == ')') {
     offset += 4;
     reduced_length -= 5;
   }

   // Skip whitespace before and after the URL inside the "url()" parenthesis.
   while (reduced_length && IsHTMLSpace<UChar>(string[offset])) {
     ++offset;
     --reduced_length;
   }
   while (reduced_length &&
          IsHTMLSpace<UChar>(string[offset + reduced_length - 1])) {
     --reduced_length;
   }

   // Remove single-quotes or double-quotes from the URL
   if ((reduced_length >= 2) &&
       (string[offset] == string[offset + reduced_length - 1]) &&
       (string[offset] == '\'' || string[offset] == '"')) {
     offset++;
     reduced_length -= 2;
   }

   return string.Substring(offset, reduced_length);
 }

 void CSSPreloadScanner::EmitRule(const SegmentedString& source) {
   if (DeprecatedEqualIgnoringCase(rule_, "import")) {
     String url = ParseCSSStringOrURL(rule_value_.ToString());
     TextPosition position =
         TextPosition(source.CurrentLine(), source.CurrentColumn());
     auto request = PreloadRequest::CreateIfNeeded(
         fetch_initiator_type_names::kCSS, position, url,
         *predicted_base_element_url_, ResourceType::kCSSStyleSheet,
         referrer_policy_, PreloadRequest::kBaseUrlIsReferrer,
         ResourceFetcher::kImageNotImageSet);
     if (request) {
       // FIXME: Should this be including the charset in the preload request?
       requests_->push_back(std::move(request));
     }
     state_ = kInitial;
   } else if (DeprecatedEqualIgnoringCase(rule_, "charset"))
     state_ = kInitial;
   else
     state_ = kDoneParsingImportRules;
   rule_.Clear();
   rule_value_.Clear();
 }

 }  // namespace blink
	/*
	* Copyright (C) 2008, 2010 Apple Inc. All Rights Reserved.
	* Copyright (C) 2009 Torch Mobile, Inc. http://www.torchmobile.com/
	* Copyright (C) 2010 Google 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 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 "third_party/blink/renderer/core/html/parser/css_preload_scanner.h"

	#include <memory>

	#include "third_party/blink/renderer/core/html/parser/html_parser_idioms.h"
	#include "third_party/blink/renderer/platform/loader/fetch/fetch_initiator_type_names.h"
	#include "third_party/blink/renderer/platform/network/http_names.h"
	#include "third_party/blink/renderer/platform/text/segmented_string.h"

	namespace blink {

	CSSPreloadScanner::CSSPreloadScanner() = default;

	CSSPreloadScanner::~CSSPreloadScanner() = default;

	void CSSPreloadScanner::Reset() {
	state_ = kInitial;
	rule_.Clear();
	rule_value_.Clear();
	}

	template <typename Char>
	void CSSPreloadScanner::ScanCommon(const Char* begin,
	const Char* end,
	const SegmentedString& source,
	PreloadRequestStream& requests,
	const KURL& predicted_base_element_url) {
	requests_ = &requests;
	predicted_base_element_url_ = &predicted_base_element_url;

	for (const Char* it = begin; it != end && state_ != kDoneParsingImportRules;
	++it)
	Tokenize(*it, source);

	requests_ = nullptr;
	predicted_base_element_url_ = nullptr;
	}

	void CSSPreloadScanner::Scan(const HTMLToken::DataVector& data,
	const SegmentedString& source,
	PreloadRequestStream& requests,
	const KURL& predicted_base_element_url) {
	ScanCommon(data.data(), data.data() + data.size(), source, requests,
	predicted_base_element_url);
	}

	void CSSPreloadScanner::Scan(const String& tag_name,
	const SegmentedString& source,
	PreloadRequestStream& requests,
	const KURL& predicted_base_element_url) {
	if (tag_name.Is8Bit()) {
	const LChar* begin = tag_name.Characters8();
	ScanCommon(begin, begin + tag_name.length(), source, requests,
	predicted_base_element_url);
	return;
	}
	const UChar* begin = tag_name.Characters16();
	ScanCommon(begin, begin + tag_name.length(), source, requests,
	predicted_base_element_url);
	}

	void CSSPreloadScanner::SetReferrerPolicy(const ReferrerPolicy policy) {
	referrer_policy_ = policy;
	}

	inline void CSSPreloadScanner::Tokenize(UChar c,
	const SegmentedString& source) {
	// We are just interested in @import rules, no need for real tokenization here
	// Searching for other types of resources is probably low payoff.
	// If we ever decide to preload fonts, we also need to change
	// ResourceFetcher::resourceNeedsLoad to immediately load speculative font
	// preloads.
	switch (state_) {
	case kInitial:
	if (IsHTMLSpace<UChar>(c))
	break;
	if (c == '@')
	state_ = kRuleStart;
	else if (c == '/')
	state_ = kMaybeComment;
	else
	state_ = kDoneParsingImportRules;
	break;
	case kMaybeComment:
	if (c == '*')
	state_ = kComment;
	else
	state_ = kInitial;
	break;
	case kComment:
	if (c == '*')
	state_ = kMaybeCommentEnd;
	break;
	case kMaybeCommentEnd:
	if (c == '*')
	break;
	if (c == '/')
	state_ = kInitial;
	else
	state_ = kComment;
	break;
	case kRuleStart:
	if ((c >= 'A' && c <= 'Z') \|\| (c >= 'a' && c <= 'z')) {
	rule_.Clear();
	rule_value_.Clear();
	rule_.Append(c);
	state_ = kRule;
	} else
	state_ = kInitial;
	break;
	case kRule:
	if (IsHTMLSpace<UChar>(c))
	state_ = kAfterRule;
	else if (c == ';')
	state_ = kInitial;
	else
	rule_.Append(c);
	break;
	case kAfterRule:
	if (IsHTMLSpace<UChar>(c))
	break;
	if (c == ';')
	state_ = kInitial;
	else if (c == '{')
	state_ = kDoneParsingImportRules;
	else {
	state_ = kRuleValue;
	rule_value_.Append(c);
	}
	break;
	case kRuleValue:
	if (IsHTMLSpace<UChar>(c)) {
	state_ = kAfterRuleValue;
	} else if (c == ';') {
	EmitRule(source);
	} else {
	rule_value_.Append(c);
	// When reading the rule and hitting ')', which signifies the URL end,
	// emit the rule.
	if (c == ')') {
	EmitRule(source);
	}
	}
	break;
	case kAfterRuleValue:
	if (IsHTMLSpace<UChar>(c))
	break;
	if (c == ';')
	EmitRule(source);
	else if (c == '{')
	state_ = kDoneParsingImportRules;
	else {
	// FIXME: media rules
	state_ = kInitial;
	}
	break;
	case kDoneParsingImportRules:
	NOTREACHED();
	break;
	}
	}

	static String ParseCSSStringOrURL(const String& string) {
	wtf_size_t offset = 0;
	wtf_size_t reduced_length = string.length();

	// Remove whitespace from the rule start
	while (reduced_length && IsHTMLSpace<UChar>(string[offset])) {
	++offset;
	--reduced_length;
	}
	// Remove whitespace from the rule end
	// TODO(yoav): Evaluate performance benefits of using raw string operations.
	// TODO(yoav): Look into moving parsing to use better parsing primitives.
	while (reduced_length &&
	IsHTMLSpace<UChar>(string[offset + reduced_length - 1])) {
	--reduced_length;
	}

	// Skip the "url(" prefix and the ")" suffix
	if (reduced_length >= 5 && (string[offset] == 'u' \|\| string[offset] == 'U') &&
	(string[offset + 1] == 'r' \|\| string[offset + 1] == 'R') &&
	(string[offset + 2] == 'l' \|\| string[offset + 2] == 'L') &&
	string[offset + 3] == '(' && string[offset + reduced_length - 1] == ')') {
	offset += 4;
	reduced_length -= 5;
	}

	// Skip whitespace before and after the URL inside the "url()" parenthesis.
	while (reduced_length && IsHTMLSpace<UChar>(string[offset])) {
	++offset;
	--reduced_length;
	}
	while (reduced_length &&
	IsHTMLSpace<UChar>(string[offset + reduced_length - 1])) {
	--reduced_length;
	}

	// Remove single-quotes or double-quotes from the URL
	if ((reduced_length >= 2) &&
	(string[offset] == string[offset + reduced_length - 1]) &&
	(string[offset] == '\'' \|\| string[offset] == '"')) {
	offset++;
	reduced_length -= 2;
	}

	return string.Substring(offset, reduced_length);
	}

	void CSSPreloadScanner::EmitRule(const SegmentedString& source) {
	if (DeprecatedEqualIgnoringCase(rule_, "import")) {
	String url = ParseCSSStringOrURL(rule_value_.ToString());
	TextPosition position =
	TextPosition(source.CurrentLine(), source.CurrentColumn());
	auto request = PreloadRequest::CreateIfNeeded(
	fetch_initiator_type_names::kCSS, position, url,
	*predicted_base_element_url_, ResourceType::kCSSStyleSheet,
	referrer_policy_, PreloadRequest::kBaseUrlIsReferrer,
	ResourceFetcher::kImageNotImageSet);
	if (request) {
	// FIXME: Should this be including the charset in the preload request?
	requests_->push_back(std::move(request));
	}
	state_ = kInitial;
	} else if (DeprecatedEqualIgnoringCase(rule_, "charset"))
	state_ = kInitial;
	else
	state_ = kDoneParsingImportRules;
	rule_.Clear();
	rule_value_.Clear();
	}

	} // namespace blink