The Tor Browser / file comparison
comparison: parser/html/nsParserUtils.cpp
parser/html/nsParserUtils.cpp
- branch
- TOR_BUG_9701
- changeset 15
- b8a032363ba2
equal
deleted
inserted
replaced
| -1:000000000000 | 0:4cfdb51ec3a3 |
|---|---|
| 1 /* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ | |
| 2 /* This Source Code Form is subject to the terms of the Mozilla Public | |
| 3 * License, v. 2.0. If a copy of the MPL was not distributed with this | |
| 4 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ | |
| 5 | |
| 6 #include "nsString.h" | |
| 7 #include "nsIComponentManager.h" | |
| 8 #include "nsCOMPtr.h" | |
| 9 #include "nsXPCOM.h" | |
| 10 #include "nsISupportsPrimitives.h" | |
| 11 #include "nsXPIDLString.h" | |
| 12 #include "nsScriptLoader.h" | |
| 13 #include "nsEscape.h" | |
| 14 #include "nsIParser.h" | |
| 15 #include "nsIDTD.h" | |
| 16 #include "nsNetCID.h" | |
| 17 #include "nsNetUtil.h" | |
| 18 #include "nsParserCIID.h" | |
| 19 #include "nsContentUtils.h" | |
| 20 #include "nsIContentSink.h" | |
| 21 #include "nsIDocumentEncoder.h" | |
| 22 #include "nsIDOMDocumentFragment.h" | |
| 23 #include "nsIFragmentContentSink.h" | |
| 24 #include "nsIDOMDocument.h" | |
| 25 #include "nsIDOMNodeList.h" | |
| 26 #include "nsIDOMNode.h" | |
| 27 #include "nsIDOMElement.h" | |
| 28 #include "nsIDocument.h" | |
| 29 #include "nsIContent.h" | |
| 30 #include "nsAttrName.h" | |
| 31 #include "nsHTMLParts.h" | |
| 32 #include "nsContentCID.h" | |
| 33 #include "nsIScriptableUnescapeHTML.h" | |
| 34 #include "nsParserUtils.h" | |
| 35 #include "nsAutoPtr.h" | |
| 36 #include "nsTreeSanitizer.h" | |
| 37 #include "nsHtml5Module.h" | |
| 38 #include "mozilla/dom/DocumentFragment.h" | |
| 39 | |
| 40 #define XHTML_DIV_TAG "div xmlns=\"http://www.w3.org/1999/xhtml\"" | |
| 41 | |
| 42 using namespace mozilla::dom; | |
| 43 | |
| 44 NS_IMPL_ISUPPORTS(nsParserUtils, | |
| 45 nsIScriptableUnescapeHTML, | |
| 46 nsIParserUtils) | |
| 47 | |
| 48 static NS_DEFINE_CID(kCParserCID, NS_PARSER_CID); | |
| 49 | |
| 50 | |
| 51 | |
| 52 NS_IMETHODIMP | |
| 53 nsParserUtils::ConvertToPlainText(const nsAString& aFromStr, | |
| 54 uint32_t aFlags, | |
| 55 uint32_t aWrapCol, | |
| 56 nsAString& aToStr) | |
| 57 { | |
| 58 return nsContentUtils::ConvertToPlainText(aFromStr, | |
| 59 aToStr, | |
| 60 aFlags, | |
| 61 aWrapCol); | |
| 62 } | |
| 63 | |
| 64 NS_IMETHODIMP | |
| 65 nsParserUtils::Unescape(const nsAString& aFromStr, | |
| 66 nsAString& aToStr) | |
| 67 { | |
| 68 return nsContentUtils::ConvertToPlainText(aFromStr, | |
| 69 aToStr, | |
| 70 nsIDocumentEncoder::OutputSelectionOnly | | |
| 71 nsIDocumentEncoder::OutputAbsoluteLinks, | |
| 72 0); | |
| 73 } | |
| 74 | |
| 75 NS_IMETHODIMP | |
| 76 nsParserUtils::Sanitize(const nsAString& aFromStr, | |
| 77 uint32_t aFlags, | |
| 78 nsAString& aToStr) | |
| 79 { | |
| 80 nsCOMPtr<nsIURI> uri; | |
| 81 NS_NewURI(getter_AddRefs(uri), "about:blank"); | |
| 82 nsCOMPtr<nsIPrincipal> principal = | |
| 83 do_CreateInstance("@mozilla.org/nullprincipal;1"); | |
| 84 nsCOMPtr<nsIDOMDocument> domDocument; | |
| 85 nsresult rv = NS_NewDOMDocument(getter_AddRefs(domDocument), | |
| 86 EmptyString(), | |
| 87 EmptyString(), | |
| 88 nullptr, | |
| 89 uri, | |
| 90 uri, | |
| 91 principal, | |
| 92 true, | |
| 93 nullptr, | |
| 94 DocumentFlavorHTML); | |
| 95 NS_ENSURE_SUCCESS(rv, rv); | |
| 96 | |
| 97 nsCOMPtr<nsIDocument> document = do_QueryInterface(domDocument); | |
| 98 rv = nsContentUtils::ParseDocumentHTML(aFromStr, document, false); | |
| 99 NS_ENSURE_SUCCESS(rv, rv); | |
| 100 | |
| 101 nsTreeSanitizer sanitizer(aFlags); | |
| 102 sanitizer.Sanitize(document); | |
| 103 | |
| 104 nsCOMPtr<nsIDocumentEncoder> encoder = | |
| 105 do_CreateInstance(NS_DOC_ENCODER_CONTRACTID_BASE "text/html"); | |
| 106 | |
| 107 encoder->NativeInit(document, | |
| 108 NS_LITERAL_STRING("text/html"), | |
| 109 nsIDocumentEncoder::OutputDontRewriteEncodingDeclaration | | |
| 110 nsIDocumentEncoder::OutputNoScriptContent | | |
| 111 nsIDocumentEncoder::OutputEncodeBasicEntities | | |
| 112 nsIDocumentEncoder::OutputLFLineBreak | | |
| 113 nsIDocumentEncoder::OutputRaw); | |
| 114 | |
| 115 return encoder->EncodeToString(aToStr); | |
| 116 } | |
| 117 | |
| 118 NS_IMETHODIMP | |
| 119 nsParserUtils::ParseFragment(const nsAString& aFragment, | |
| 120 bool aIsXML, | |
| 121 nsIURI* aBaseURI, | |
| 122 nsIDOMElement* aContextElement, | |
| 123 nsIDOMDocumentFragment** aReturn) | |
| 124 { | |
| 125 return nsParserUtils::ParseFragment(aFragment, | |
| 126 0, | |
| 127 aIsXML, | |
| 128 aBaseURI, | |
| 129 aContextElement, | |
| 130 aReturn); | |
| 131 } | |
| 132 | |
| 133 NS_IMETHODIMP | |
| 134 nsParserUtils::ParseFragment(const nsAString& aFragment, | |
| 135 uint32_t aFlags, | |
| 136 bool aIsXML, | |
| 137 nsIURI* aBaseURI, | |
| 138 nsIDOMElement* aContextElement, | |
| 139 nsIDOMDocumentFragment** aReturn) | |
| 140 { | |
| 141 NS_ENSURE_ARG(aContextElement); | |
| 142 *aReturn = nullptr; | |
| 143 | |
| 144 nsCOMPtr<nsIDocument> document; | |
| 145 nsCOMPtr<nsIDOMDocument> domDocument; | |
| 146 nsCOMPtr<nsIDOMNode> contextNode; | |
| 147 contextNode = do_QueryInterface(aContextElement); | |
| 148 contextNode->GetOwnerDocument(getter_AddRefs(domDocument)); | |
| 149 document = do_QueryInterface(domDocument); | |
| 150 NS_ENSURE_TRUE(document, NS_ERROR_NOT_AVAILABLE); | |
| 151 | |
| 152 nsAutoScriptBlockerSuppressNodeRemoved autoBlocker; | |
| 153 | |
| 154 // stop scripts | |
| 155 nsRefPtr<nsScriptLoader> loader; | |
| 156 bool scripts_enabled = false; | |
| 157 if (document) { | |
| 158 loader = document->ScriptLoader(); | |
| 159 scripts_enabled = loader->GetEnabled(); | |
| 160 } | |
| 161 if (scripts_enabled) { | |
| 162 loader->SetEnabled(false); | |
| 163 } | |
| 164 | |
| 165 // Wrap things in a div or body for parsing, but it won't show up in | |
| 166 // the fragment. | |
| 167 nsAutoTArray<nsString, 2> tagStack; | |
| 168 nsAutoCString base, spec; | |
| 169 if (aIsXML) { | |
| 170 // XHTML | |
| 171 if (aBaseURI) { | |
| 172 base.Append(NS_LITERAL_CSTRING(XHTML_DIV_TAG)); | |
| 173 base.Append(NS_LITERAL_CSTRING(" xml:base=\"")); | |
| 174 aBaseURI->GetSpec(spec); | |
| 175 // nsEscapeHTML is good enough, because we only need to get | |
| 176 // quotes, ampersands, and angle brackets | |
| 177 char* escapedSpec = nsEscapeHTML(spec.get()); | |
| 178 if (escapedSpec) | |
| 179 base += escapedSpec; | |
| 180 NS_Free(escapedSpec); | |
| 181 base.Append(NS_LITERAL_CSTRING("\"")); | |
| 182 tagStack.AppendElement(NS_ConvertUTF8toUTF16(base)); | |
| 183 } else { | |
| 184 tagStack.AppendElement(NS_LITERAL_STRING(XHTML_DIV_TAG)); | |
| 185 } | |
| 186 } | |
| 187 | |
| 188 nsresult rv = NS_OK; | |
| 189 nsCOMPtr<nsIContent> fragment; | |
| 190 if (aIsXML) { | |
| 191 rv = nsContentUtils::ParseFragmentXML(aFragment, | |
| 192 document, | |
| 193 tagStack, | |
| 194 true, | |
| 195 aReturn); | |
| 196 fragment = do_QueryInterface(*aReturn); | |
| 197 } else { | |
| 198 NS_ADDREF(*aReturn = new DocumentFragment(document->NodeInfoManager())); | |
| 199 fragment = do_QueryInterface(*aReturn); | |
| 200 rv = nsContentUtils::ParseFragmentHTML(aFragment, | |
| 201 fragment, | |
| 202 nsGkAtoms::body, | |
| 203 kNameSpaceID_XHTML, | |
| 204 false, | |
| 205 true); | |
| 206 // Now, set the base URI on all subtree roots. | |
| 207 if (aBaseURI) { | |
| 208 aBaseURI->GetSpec(spec); | |
| 209 nsAutoString spec16; | |
| 210 CopyUTF8toUTF16(spec, spec16); | |
| 211 nsIContent* node = fragment->GetFirstChild(); | |
| 212 while (node) { | |
| 213 if (node->IsElement()) { | |
| 214 node->SetAttr(kNameSpaceID_XML, | |
| 215 nsGkAtoms::base, | |
| 216 nsGkAtoms::xml, | |
| 217 spec16, | |
| 218 false); | |
| 219 } | |
| 220 node = node->GetNextSibling(); | |
| 221 } | |
| 222 } | |
| 223 } | |
| 224 if (fragment) { | |
| 225 nsTreeSanitizer sanitizer(aFlags); | |
| 226 sanitizer.Sanitize(fragment); | |
| 227 } | |
| 228 | |
| 229 if (scripts_enabled) { | |
| 230 loader->SetEnabled(true); | |
| 231 } | |
| 232 | |
| 233 return rv; | |
| 234 } |
