src/net/instaweb/rewriter/add_ids_filter.cc - incubator-pagespeed-debian - Git at Google

 /*
  * Copyright 2011 Google Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
  * You may obtain a copy of the License at
  *
  *      http://www.apache.org/licenses/LICENSE-2.0
  *
  * Unless required by applicable law or agreed to in writing, software
  * distributed under the License is distributed on an "AS IS" BASIS,
  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  * See the License for the specific language governing permissions and
  * limitations under the License.
  */

 // Author: jhoch@google.com (Jason R. Hoch)
 // Author: jmaessen@google.com (Jan-Willem Maessen)

 #include "net/instaweb/rewriter/public/add_ids_filter.h"

 #include <algorithm>

 #include "base/logging.h"
 #include "pagespeed/kernel/base/basictypes.h"
 #include "pagespeed/kernel/base/string.h"
 #include "pagespeed/kernel/base/string_util.h"
 #include "pagespeed/kernel/html/html_element.h"
 #include "pagespeed/kernel/html/html_name.h"

 namespace net_instaweb {

 namespace {

 // TODO(jmaessen): perhaps this should go somewhere central?  It needs to be a
 // subset of the tags considered divlike by mobilize_label_filter at least.
 const HtmlName::Keyword kDivLikeTags[] = {
   HtmlName::kArticle,
   HtmlName::kAside,
   HtmlName::kContent,
   HtmlName::kDiv,
   HtmlName::kFooter,
   HtmlName::kForm,
   HtmlName::kHeader,
   HtmlName::kMain,
   HtmlName::kMenu,
   HtmlName::kNav,
   HtmlName::kSection,
   HtmlName::kTable,
   HtmlName::kTr,
   HtmlName::kUl
 };

 #ifndef NDEBUG
 // For invariant-checking the static data above.
 void CheckKeywordsSorted(const HtmlName::Keyword* list, int len) {
   for (int i = 1; i < len; ++i) {
     DCHECK_LT(list[i - 1], list[i]);
   }
 }
 #endif  // #ifndef NDEBUG

 bool IsDivLike(HtmlName::Keyword tag) {
   return std::binary_search(
       kDivLikeTags, kDivLikeTags + arraysize(kDivLikeTags), tag);
 }

 bool NeedsExplicitId(HtmlName::Keyword tag) {
   return IsDivLike(tag);
 }

 bool IsIgnored(HtmlName::Keyword tag) {
   return (tag == HtmlName::kHtml || tag == HtmlName::kBody);
 }

 }  // namespace

 // We don't want this to conflict with another id name, and length
 // also matters (shorter is better).
 const char AddIdsFilter::kIdPrefix[] = "PageSpeed";

 const int AddIdsFilter::kIsId = -1;

 AddIdsFilter::AddIdsFilter(RewriteDriver* driver)
     : driver_(driver) {}

 AddIdsFilter::~AddIdsFilter() {}

 void AddIdsFilter::StartDocument() {
   // Push an initial top-level count.
   div_count_stack_.clear();
   div_count_stack_.push_back(0);
   id_stack_.clear();
 #ifndef NDEBUG
   CheckKeywordsSorted(kDivLikeTags, arraysize(kDivLikeTags));
 #endif  // #ifndef NDEBUG
 }

 // As we parse outside head we maintain a stack of tag locations, and at each
 // tag for which TagRequiresId we add an encoded version of the stack as a query
 // param.  Note that the stack is incremented immediately after its encoded
 // value is added as a query param.
 //
 // An explicit id adds a kIsId entry to the stack before the entry for that
 // tag's children, and pushes the id onto the id_stack_.
 //
 // Example HTML:                   | Stack as we go:
 //                                 |
 // <html>                          | 0
 //   <head>                        | 0
 //   </head>                       | 0
 //   <body>                        | 0
 //     <div>                       | 0, 0 (id="...-0")
 //       <p>Toolbar link 1.</p>    | 0, 0
 //       <p>Toolbar link 2.</p>    | 0, 1
 //     </div>                      | 1             id stack
 //     <div id=foo>                | 1, -1, 0      foo
 //       <div>                     | 1, -1, 0, 0   foo (id="...-foo-0")
 //         <p>Main page link.</p>  | 1, -1, 0, 0   foo
 //       </div>                    | 1, 1
 //       <div>Secondary link.      | 1, -1, 1, 0   foo (id="...-foo-1")
 //       </div>                    | 1, -1, 2      foo
 //     </div>                      | 2
 //   </body>                       | 2
 // </html>                         | 2
 void AddIdsFilter::StartElement(HtmlElement* element) {
   HtmlName::Keyword tag = element->keyword();
   const HtmlElement::Attribute* id =
       element->FindAttribute(HtmlName::kId);
   if (id != NULL) {
     id_stack_.push_back(id);
     div_count_stack_.push_back(kIsId);
   } else if (IsIgnored(tag)) {
     // Don't touch stack in this case.
     return;
   } else if (NeedsExplicitId(tag) ||
              element->FindAttribute(HtmlName::kClass) != NULL) {
     driver_->AddAttribute(element, HtmlName::kId, GetDivCountStackEncoding());
   }
   div_count_stack_.push_back(0);
 }

 void AddIdsFilter::EndElement(HtmlElement* element) {
   DCHECK(!div_count_stack_.empty());
   DCHECK_NE(kIsId, div_count_stack_.back());
   if (!id_stack_.empty() &&
       id_stack_.back() == element->FindAttribute(HtmlName::kId)) {
     DCHECK_LT(2, div_count_stack_.size());
     // For an element with an id the stack looks like:
     // ... my_count_in_parent kIsId child_count
     // If so, pop both along with the back of id_stack_.
     div_count_stack_.pop_back();
     id_stack_.pop_back();
     // Now stack is ... my_count_in_parent kIsId
   } else if (IsIgnored(element->keyword())) {
     // Again, don't touch the stack in this case.
     return;
   } else {
     // stack is:
     // ... my_count_in_parent child_count
   }
   div_count_stack_.pop_back();
   // Stack is ... my_count_in_parent
   div_count_stack_.back()++;
   // Stack is ... my_count_in_parent+1
   DCHECK(!div_count_stack_.empty());
   DCHECK_NE(kIsId, div_count_stack_.back());
 }

 GoogleString AddIdsFilter::GetDivCountStackEncoding() {
   DCHECK(!div_count_stack_.empty());
   DCHECK_NE(kIsId, div_count_stack_.back());
   GoogleString result(kIdPrefix);
   if (!id_stack_.empty()) {
     // Note: we make use of StringPiece(NULL) -> "" in this call.
     StrAppend(&result, "-", id_stack_.back()->escaped_value());
   }
   int size = div_count_stack_.size();
   int count_index = size - 1;
   while (count_index > 0 && div_count_stack_[count_index - 1] != kIsId) {
     --count_index;
   }
   for (; count_index < size; ++count_index) {
     StrAppend(&result, "-", IntegerToString(div_count_stack_[count_index]));
   }
   return result;
 }

 }  // namespace net_instaweb
	/*
	* Copyright 2011 Google Inc.
	*
	* Licensed under the Apache License, Version 2.0 (the "License");
	* you may not use this file except in compliance with the License.
	* You may obtain a copy of the License at
	*
	* http://www.apache.org/licenses/LICENSE-2.0
	*
	* Unless required by applicable law or agreed to in writing, software
	* distributed under the License is distributed on an "AS IS" BASIS,
	* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
	* See the License for the specific language governing permissions and
	* limitations under the License.
	*/

	// Author: jhoch@google.com (Jason R. Hoch)
	// Author: jmaessen@google.com (Jan-Willem Maessen)

	#include "net/instaweb/rewriter/public/add_ids_filter.h"

	#include <algorithm>

	#include "base/logging.h"
	#include "pagespeed/kernel/base/basictypes.h"
	#include "pagespeed/kernel/base/string.h"
	#include "pagespeed/kernel/base/string_util.h"
	#include "pagespeed/kernel/html/html_element.h"
	#include "pagespeed/kernel/html/html_name.h"

	namespace net_instaweb {

	namespace {

	// TODO(jmaessen): perhaps this should go somewhere central? It needs to be a
	// subset of the tags considered divlike by mobilize_label_filter at least.
	const HtmlName::Keyword kDivLikeTags[] = {
	HtmlName::kArticle,
	HtmlName::kAside,
	HtmlName::kContent,
	HtmlName::kDiv,
	HtmlName::kFooter,
	HtmlName::kForm,
	HtmlName::kHeader,
	HtmlName::kMain,
	HtmlName::kMenu,
	HtmlName::kNav,
	HtmlName::kSection,
	HtmlName::kTable,
	HtmlName::kTr,
	HtmlName::kUl
	};

	#ifndef NDEBUG
	// For invariant-checking the static data above.
	void CheckKeywordsSorted(const HtmlName::Keyword* list, int len) {
	for (int i = 1; i < len; ++i) {
	DCHECK_LT(list[i - 1], list[i]);
	}
	}
	#endif // #ifndef NDEBUG

	bool IsDivLike(HtmlName::Keyword tag) {
	return std::binary_search(
	kDivLikeTags, kDivLikeTags + arraysize(kDivLikeTags), tag);
	}

	bool NeedsExplicitId(HtmlName::Keyword tag) {
	return IsDivLike(tag);
	}

	bool IsIgnored(HtmlName::Keyword tag) {
	return (tag == HtmlName::kHtml \|\| tag == HtmlName::kBody);
	}

	} // namespace

	// We don't want this to conflict with another id name, and length
	// also matters (shorter is better).
	const char AddIdsFilter::kIdPrefix[] = "PageSpeed";

	const int AddIdsFilter::kIsId = -1;

	AddIdsFilter::AddIdsFilter(RewriteDriver* driver)
	: driver_(driver) {}

	AddIdsFilter::~AddIdsFilter() {}

	void AddIdsFilter::StartDocument() {
	// Push an initial top-level count.
	div_count_stack_.clear();
	div_count_stack_.push_back(0);
	id_stack_.clear();
	#ifndef NDEBUG
	CheckKeywordsSorted(kDivLikeTags, arraysize(kDivLikeTags));
	#endif // #ifndef NDEBUG
	}

	// As we parse outside head we maintain a stack of tag locations, and at each
	// tag for which TagRequiresId we add an encoded version of the stack as a query
	// param. Note that the stack is incremented immediately after its encoded
	// value is added as a query param.
	//
	// An explicit id adds a kIsId entry to the stack before the entry for that
	// tag's children, and pushes the id onto the id_stack_.
	//
	// Example HTML: \| Stack as we go:
	// \|
	// <html> \| 0
	// <head> \| 0
	// </head> \| 0
	// <body> \| 0
	// <div> \| 0, 0 (id="...-0")
	// <p>Toolbar link 1.</p> \| 0, 0
	// <p>Toolbar link 2.</p> \| 0, 1
	// </div> \| 1 id stack
	// <div id=foo> \| 1, -1, 0 foo
	// <div> \| 1, -1, 0, 0 foo (id="...-foo-0")
	// <p>Main page link.</p> \| 1, -1, 0, 0 foo
	// </div> \| 1, 1
	// <div>Secondary link. \| 1, -1, 1, 0 foo (id="...-foo-1")
	// </div> \| 1, -1, 2 foo
	// </div> \| 2
	// </body> \| 2
	// </html> \| 2
	void AddIdsFilter::StartElement(HtmlElement* element) {
	HtmlName::Keyword tag = element->keyword();
	const HtmlElement::Attribute* id =
	element->FindAttribute(HtmlName::kId);
	if (id != NULL) {
	id_stack_.push_back(id);
	div_count_stack_.push_back(kIsId);
	} else if (IsIgnored(tag)) {
	// Don't touch stack in this case.
	return;
	} else if (NeedsExplicitId(tag) \|\|
	element->FindAttribute(HtmlName::kClass) != NULL) {
	driver_->AddAttribute(element, HtmlName::kId, GetDivCountStackEncoding());
	}
	div_count_stack_.push_back(0);
	}

	void AddIdsFilter::EndElement(HtmlElement* element) {
	DCHECK(!div_count_stack_.empty());
	DCHECK_NE(kIsId, div_count_stack_.back());
	if (!id_stack_.empty() &&
	id_stack_.back() == element->FindAttribute(HtmlName::kId)) {
	DCHECK_LT(2, div_count_stack_.size());
	// For an element with an id the stack looks like:
	// ... my_count_in_parent kIsId child_count
	// If so, pop both along with the back of id_stack_.
	div_count_stack_.pop_back();
	id_stack_.pop_back();
	// Now stack is ... my_count_in_parent kIsId
	} else if (IsIgnored(element->keyword())) {
	// Again, don't touch the stack in this case.
	return;
	} else {
	// stack is:
	// ... my_count_in_parent child_count
	}
	div_count_stack_.pop_back();
	// Stack is ... my_count_in_parent
	div_count_stack_.back()++;
	// Stack is ... my_count_in_parent+1
	DCHECK(!div_count_stack_.empty());
	DCHECK_NE(kIsId, div_count_stack_.back());
	}

	GoogleString AddIdsFilter::GetDivCountStackEncoding() {
	DCHECK(!div_count_stack_.empty());
	DCHECK_NE(kIsId, div_count_stack_.back());
	GoogleString result(kIdPrefix);
	if (!id_stack_.empty()) {
	// Note: we make use of StringPiece(NULL) -> "" in this call.
	StrAppend(&result, "-", id_stack_.back()->escaped_value());
	}
	int size = div_count_stack_.size();
	int count_index = size - 1;
	while (count_index > 0 && div_count_stack_[count_index - 1] != kIsId) {
	--count_index;
	}
	for (; count_index < size; ++count_index) {
	StrAppend(&result, "-", IntegerToString(div_count_stack_[count_index]));
	}
	return result;
	}

	} // namespace net_instaweb