mirrors/zotero
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions
zotero/chrome/content/zotero/webpagedump/domsaver.js
Dan Stillman a97e27a9f2 Fix CSS in snapshots of Reader View 
Allow saving of CSS links over chrome://. We can scope this to
chrome://global/skin/aboutReader.css if anyone thinks of a reason why
this was disabled to begin with, but I'm not sure in what other
situations CSS it would apply.
2015-07-17 16:22:15 -04:00

1091 lines
No EOL
42 KiB
JavaScript
Raw Blame History

/* ***** BEGIN LICENSE BLOCK *****
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
*
* The contents of this file are subject to the Mozilla Public License Version
* 1.1 (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.mozilla.org/MPL/
*
* Software distributed under the License is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
* for the specific language governing rights and limitations under the
* License.
*
* The Original Code is ScrapBook.
*
* The Initial Developer of the Original Code is Gomita.
* Portions created by the Initial Developer are Copyright (C) 2004
* the Initial Developer. All Rights Reserved.
*
* Contributor(s):
* Bernhard Pollak <pollak@dbai.tuwien.ac.at> (WebPageDump Fork)
*
* Alternatively, the contents of this file may be used under the terms of
* either the GNU Affero General Public License Version 2 or later (the "GPL"), or
* the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
* in which case the provisions of the GPL or the LGPL are applicable instead
* of those above. If you wish to allow use of your version of this file only
* under the terms of either the GPL or the LGPL, and not to allow others to
* use your version of this file under the terms of the MPL, indicate your
* decision by deleting the provisions above and replace them with the notice
* and other provisions required by the GPL or the LGPL. If you do not delete
* the provisions above, a recipient may use your version of this file under
* the terms of any one of the MPL, the GPL or the LGPL.
*
* ***** END LICENSE BLOCK ***** */
// --------------------------------------------------------------------------------
// "WebPageDump" Firefox Extension
// --------------------------------------------------------------------------------
// - File: "domsaver.js" -
// - Description:
// Makes a (hopefully perfect) local copy of the actual open webpage.
// Current Browsers make sometimes errors when saving a webpage. The files
// will be saved in one flat directory (without subdirs)
// - Using:
// 1. call "wpdDOMSaver.init(filePath)" and pass the full destination path
// 2. afterwards call "wpdDOMSaver.saveHTMLDocument" for saving the (active) window
// --------------------------------------------------------------------------------
// Call Tree Overview - wpdDOMSaver
//
// saveHTMLDocument
// saveDocumentEx (decide if we have a HTML or another file)
// saveDocumentFile (we have a non HTML file (e.g for embedded objects - images, movies,...))
// download (we download the file and ...)
// writefile (... make a HTML wrapper file)
// saveDocumentHTML (we have a HTML File)
// processDOMRecursively (go through the DOM nodes)
// processDOMNode (for each node we do extensive processing (links, javascript,...))
// download (for image,flash,... references)
// saveDocumentEx ... (starting again with "saveDocumentEx" for frame documents)
// saveDocumentCSS (save CSS File)
// processCSSRecursively (process the CSS text)
// processCSSText (do some replacement stuff and link processing)
// download (download CSS image references)
// generateHTMLString (create the HTML string)
//
//
// --------------------------------------------------------------------------------
//
//
// TO DO: use version information from rdf file...
var WPD_VERSION = "0.2";
// Bug variables: set to false if the bug is not present anymore
// CRLFBUG: Innerhtml trims the text inside a tag. This lead
// to problems with the PRE Tag where sometimes one starting
// carriage return is lost...
var WPD_CRLFBUG = true;
// ENTITYBUG: HTML Entities are lost inside the DOM Tree (they
// are converted to corresponding unicode characters) which
// results in problems when using a non unicode charset as output
// target where this values/symbols do not exist. So we call
// the ConvertToEntities XPCOM function for generating usual
// HTML Entities...
// (this is precisely not a bug but a concept failure)
var WPD_ENTITYBUG = false;
// CSSSCROLLBUG: The css "scroll" property of "background" is
// loosing the zero vertical position leading to a false
// positioned background (centered by default)...
var WPD_CSSSCROLLBUG = true;
// CSSBACKGROUNDPOSITIONBUG: "background-position 0 0" is
// loosing the zero vertical position
var WPD_CSSBACKGROUNDPOSITIONBUG = true;
// DOCTYPEBUG: If the doctype is inserted before
// the <HTML> tag there would be rendering errors with the
// right to left writing direction, because there are problems
// with the DIR attribute (text direction (rtl,ltr,lro,rlo))
// Positioning the doctype below the <HTML> Tag would fix the
// problem.
// But: inserting the docctype only below the <HTML> tag
// results in small layout changes in some tables. So we
// leave the doctype at the original position before the
// HTML tag <HTML> and insert the doctype entry a second
// time below the <HTML> tag...
var WPD_DOCTYPEBUG = false;
// JAVASCRIPTSRCBUG: Deleting the "src" attribute together
// with the whole <SCRIPT> tag may result in unexpected
// layout changes (table width is changed). So we set the
// "src" attribute of the <SCRIPT> tag to an empty string
// and don<6F>t delete the whole tag...
// Remark: it may be necessary to use an invalid ip address
// (e.g. http://0.0.0.0) but this may lead to other strange
// layout dependencies...
var WPD_JAVASCRIPTSRCBUG = true;
// CLONENODEBUG: CloneNode copies only the initial state of
// the INPUT fields and ignores the actual values of the fields
// We introduced this.curBody and the getCurrentNodeValue function.
var WPD_CLONENODEBUG = true;
var wpdDOMSaver = {
name: "",
document: null, // the original document
curDocument: null, // the current document
curCharacterSet: "", // the current characterset
curBody: null, // the current body node (inclusive child nodes)
currentDir: "",
baseURL: "", // the original base url
currentURL: "", // the current url (necessary for frames)
fileInfo: [], // for saving already processed files and double name checking
// (cause we use one flat directory for all files)
option: {},
frameList: [],
frameNumber: 0,
dateObj: null,
// initialize the properties (set document, URL, Directory, ...)
init: function (fileName, document) {
Zotero.debug("[wpdDOMSaver.init] ...");
this.name = "";
this.document = null;
this.curDocument = null;
this.curCharacterSet = "";
this.curBody = null;
this.currentDir = "";
this.baseURL = "";
this.currentURL = "";
this.fileInfo = []; // clear registered downloaded files...
this.option = {};
this.frameList = []; // clear frame list
this.frameNumber = 0;
this.dateObj = new Date();
// Split fileName in Path and Name
this.name = wpdCommon.getValidFileName(
wpdCommon.getFileLeafName(fileName)); // extract fileName from filePath
this.currentDir = wpdCommon.getFilePath(fileName); // only directory
this.name = wpdCommon.splitFileName(this.name)[0]; // no extension!
// Added by Dan S. for Zotero, replacing three lines below
this.document = document;
this.setFrameList(document.defaultView);
this.baseURL = document.location.href;
// Set the document and frames
//this.document = top.window._content.document;
//this.setFrameList(top.window._content);
// set the urls
//this.baseURL = wpdCommon.getURL(); // initial base url
this.currentURL = this.baseURL; // current base url - needed for frame processing
// (without frames this property will always be like the baseURL)
// default options - for the files which should be downloaded
// (this is only for external link references not for the embedded files)
this.option = {
"image": false,
"sound": false,
"movie": false,
"archive": false,
"custom": "", // comma delimited custom extensions (e.g. doc,xls,...)
"format": true, // when false we get only naked html without images
// Changed by Dan for Zotero
"script": true, // no scripts
"encodeUTF8": true, // write the DOM Tree as UTF-8 and change the charset entry of the document
"metainfo": true, // include meta tags with URL and date/time information
"metacharset": false // if the meta charset is defined inside html override document charset
//"xtagging" : true // include a x tag around each word
};
},
// get all frames in the document (recursively) and save in this.frameList
setFrameList: function (aDocument) {
try {
for (var f = 0; f < aDocument.frames.length; f++) {
this.frameList.push(aDocument.frames[f]);
this.setFrameList(aDocument.frames[f]);
}
} catch (ex) {}
},
// resolve the javascript links inside the attributes (e.g. onclick,...)
normalizeJavaScriptLink: function (aNode, aAttr) {
var val = aNode.getAttribute(aAttr); // get the attribute value and check for link stuff
if (!val || !val.match(/\(\'([^\']+)\'/)) return aNode;
val = RegExp.$1;
if (val.indexOf("/") == -1 && val.indexOf(".") == -1) return aNode;
val = wpdCommon.resolveURL(this.currentURL, val); // it is a link -> resolve and set the URL to the local URL
if (aNode.nodeName.toLowerCase() == "img") {
if (aNode.parentNode.nodeName.toLowerCase() == "a") {
aNode.parentNode.setAttribute("href", val); // change the href of img to the onclick url
aNode.removeAttribute("onclick");
} else {
val = "window.open('" + val + "');"; // if this is not a reference make a window open function for the img
aNode.setAttribute(aAttr, val);
}
} else {
if (aNode.hasAttribute("href") && aNode.getAttribute("href").indexOf("http://") != 0) {
aNode.setAttribute("href", val);
aNode.removeAttribute("onclick");
}
}
return aNode;
},
// check if the file extension of the url is specified in the options array
checkFileTypeOptions: function (aURL) {
var ext = wpdCommon.splitFileName(wpdCommon.getFileName(aURL))[1].toLowerCase();
var flag = false;
switch (ext) {
case "jpg":
case "jpeg":
case "png":
case "gif":
flag = this.option["image"];
break;
case "mp3":
case "wav":
case "ram":
case "wma":
flag = this.option["sound"];
break;
case "mpg":
case "mpeg":
case "avi":
case "ram":
case "rm":
case "mov":
case "wmv":
flag = this.option["movie"];
break;
case "zip":
case "lzh":
case "rar":
case "xpi":
flag = this.option["archive"];
break;
default:
if (ext && this.option["custom"]) {
if ((", " + this.option["custom"] + ", ").indexOf(", " + ext + ", ") != -1) flag = true;
}
}
if (aURL.indexOf("file://") == 0 && !aURL.match(/\.html$/)) flag = true;
return flag;
},
// do the conversion from the DOM Text to the destination Charset
convertEntity: function (aText) {
if (this.option["encodeUTF8"]) {
return wpdCommon.unicodeToEntity(aText, "UTF-8");
} else {
return wpdCommon.unicodeToEntity(aText, this.curCharacterSet);
}
},
// we only can manage GIF animations - Flash does not work...
disableAnimation: function (aNode) {
// thanx to pageanimator extension...
/* try {
//dump("inspecting "+aNode.nodeName+"\n");
//aNode.setAttribute("swLiveConnect", "true");
aNode.StopPlay();
dump ("prepare flash deanimation ... ");
if ( aNode.hasAttribute("play") ) aNode.setAttribute("play", "false");
dump ("flash deanimation ... ");
aNode.Rewind(); // seems to be the key for some obnoxious instances
aNode.StopPlay();
dump ("ready! \n");
} catch (e) {} */
try {
var container = aNode.QueryInterface(Components.interfaces.nsIImageLoadingContent)
.getRequest(Components.interfaces.nsIImageLoadingContent.CURRENT_REQUEST)
.image;
container.animationMode = Components.interfaces.imgIContainer.kDontAnimMode;
} catch (e) {}
},
// get the node value of aNode directly from the actual DOM tree (WPD_CLONENODEBUG)
getCurrentNodeValue: function (aNode) {
try {
this.curDocument.body.cloneNode(false);
var body = this.curDocument.body;
} catch (ex) {
var body = this.curDocument.getElementsByTagName("body")[0];
}
var refnodes = body.getElementsByTagName(aNode.nodeName);
var nodes = this.curBody.getElementsByTagName(aNode.nodeName);
if (refnodes.length != nodes.length) return aNode.value;
for (var i = 0; i < refnodes.length; i++) {
if ((nodes[i] == aNode) && (refnodes[i].name == aNode.name) && (refnodes[i].defaultValue == aNode.defaultValue)) {
return refnodes[i].value;
}
}
return aNode.value;
},
// process the DOM Node (update the links, remove attributes and process the options)
processDOMNode: function (aNode) {
this.disableAnimation(aNode);
try {
switch (aNode.nodeName.toLowerCase()) {
case "img":
case "embed":
// "embed": embedding multimedia content
if (this.option["format"]) {
if (aNode.hasAttribute("onclick")) aNode = this.normalizeJavaScriptLink(aNode, "onclick");
var aDownload = true;
if (aNode.nodeName.toLowerCase() == "img") {
try {
aDownload = aNode.complete;
} catch (ex) {}
}
var aFileName = this.download(aNode.src, aDownload);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("src", this.relativeLinkFix(aFileName));
} else {
return wpdCommon.removeNodeFromParent(aNode);
}
break;
case "object":
// for embedding different data sources in the html page
if (this.option["format"]) {
var aFileName = this.download(aNode.data, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("data", this.relativeLinkFix(aFileName));
} else {
return wpdCommon.removeNodeFromParent(aNode);
}
break;
case "body":
if (this.option["format"]) {
var aFileName = this.download(aNode.background, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("background", this.relativeLinkFix(aFileName));
} else {
aNode.removeAttribute("background");
aNode.removeAttribute("bgcolor");
aNode.removeAttribute("text");
}
break;
case "table":
case "tr":
case "th":
case "td":
if (this.option["format"]) {
var aFileName = this.download(aNode.getAttribute("background"), true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("background", this.relativeLinkFix(aFileName));
} else {
aNode.removeAttribute("background");
aNode.removeAttribute("bgcolor");
}
break;
case "input":
if (aNode.type.toLowerCase() == "image") {
if (this.option["format"]) {
var aFileName = this.download(aNode.src, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("src", this.relativeLinkFix(aFileName));
} else {
aNode.setAttribute("type", "button");
aNode.removeAttribute("src");
}
} else if ((aNode.type.toLowerCase() != "hidden") && (aNode.hasAttribute("value"))) {
if (WPD_CLONENODEBUG) aNode.setAttribute("value", this.getCurrentNodeValue(aNode));
if (WPD_ENTITYBUG) aNode.setAttribute("value", this.convertEntity(aNode.getAttribute("value")));
}
break;
case "link":
// could containt urls (icon, stylesheet and fontdef)
// We have to remove nodes with the stylesheet attribute because they will be added later
if(!aNode.hasAttribute("rel")) return aNode;
if (aNode.getAttribute("rel").toLowerCase() == "stylesheet"
&& (aNode.hasAttribute("href") && aNode.getAttribute("href").indexOf("chrome://") == -1)) {
return wpdCommon.removeNodeFromParent(aNode);
} else if (aNode.getAttribute("rel").toLowerCase() == "shortcut icon"
|| aNode.getAttribute("rel").toLowerCase() == "icon") {
var aFileName = this.download(aNode.href, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("href", this.relativeLinkFix(aFileName));
} else if (aNode.getAttribute("rel").toLowerCase() == "fontdef") {
var aFileName = this.download(aNode.src, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("src", this.relativeLinkFix(aFileName));
} else {
aNode.setAttribute("href", aNode.href);
}
break;
case "style":
return wpdCommon.removeNodeFromParent(aNode);
break;
case "applet":
if (aNode.hasAttribute("code")) aNode.setAttribute("code", "");
if (aNode.hasAttribute("codebase")) aNode.setAttribute("codebase", "");
if (aNode.hasAttribute("archive")) aNode.setAttribute("archive", "");
break;
case "script":
if (this.option["script"]) {
if (aNode.hasAttribute("src")) {
var aFileName = this.download(aNode.src, true);
// Changed by Dan S. for Zotero -- see this.repairRelativeLinks()
if (aFileName) aNode.setAttribute("src", this.relativeLinkFix(aFileName));
}
} else {
if (WPD_JAVASCRIPTSRCBUG && aNode.hasAttribute("src")) {
//if ( aNode.getAttribute("src").indexOf("http://")!=-1 ) {
// aNode.setAttribute("src", "http://0.0.0.0");
//} else {
aNode.setAttribute("src", "");
//}
} else {
return wpdCommon.removeNodeFromParent(aNode);
}
}
break;
case "noscript":
if (!WPD_JAVASCRIPTSRCBUG) return wpdCommon.removeNodeFromParent(aNode);
break;
case "a":
case "area":
if (aNode.hasAttribute("onclick")) aNode = this.normalizeJavaScriptLink(aNode, "onclick");
if (!aNode.hasAttribute("href")) return aNode;
if (aNode.target == "_blank") aNode.setAttribute("target", "_top");
if (aNode.href.match(/^javascript:/i)) aNode = this.normalizeJavaScriptLink(aNode, "href");
if (!this.selection && aNode.getAttribute("href").charAt(0) == "#") return aNode;
// download file depending on option settings and file extension
if (this.checkFileTypeOptions(aNode.href)) {
var aFileName = this.download(aNode.href, true);
if (aFileName) aNode.setAttribute("href", aFileName);
} else {
aNode.setAttribute("href", aNode.href);
}
break;
case "form":
aNode.setAttribute("action", wpdCommon.resolveURL(this.currentURL, aNode.action));
break;
case "meta":
if ((aNode.hasAttribute("http-equiv") && aNode.hasAttribute("content")) && (aNode.getAttribute("http-equiv").toLowerCase() == "content-type") && (aNode.getAttribute("content").match(/charset\=/i))) {
// we remove possible charset definitions because they will be added later
return wpdCommon.removeNodeFromParent(aNode);
}
if ((aNode.hasAttribute("http-equiv") && aNode.hasAttribute("content")) && (aNode.getAttribute("http-equiv").toLowerCase() == "refresh") && (aNode.getAttribute("content").match(/URL\=/i))) {
// there should be no refresh present - could be a noframe relict...
// (is already processed or timer is longer...)
return wpdCommon.removeNodeFromParent(aNode);
}
break;
case "base":
//<BASE HREF="http://www.amin.org/look/amin/">
// we need to set the base url to currenturl
if (aNode.hasAttribute("href") && (aNode.getAttribute("href") != "")) this.currentURL = aNode.getAttribute("href");
return wpdCommon.removeNodeFromParent(aNode);
break;
case "frame":
case "iframe":
// normal and embedded frames (iframe) -> call "saveDocumentEx" for saving the frame document
try {
// we don't have to worry about the currentURL - saveDocumentEx will set the
// currentURL to the URL of the frame document and afterwards back to the baseURL
if (this.frameNumber < this.frameList.length) {
var newFileName = this.saveDocumentEx(this.frameList[this.frameNumber++].document, this.name + "_" + this.frameNumber);
aNode.setAttribute("src", this.relativeLinkFix(newFileName));
}
} catch (ex) {
wpdCommon.addError("[wpdCommon.processDOMNode]:\n -> aNode.nodeName: " + aNode.nodeName, ex);
}
break;
case "xmp":
// TO DO
var pre = aNode.ownerDocument.createElement("pre");
pre.appendChild(aNode.firstChild);
aNode.parentNode.replaceChild(pre, aNode);
break;
}
if (!this.option["format"]) {
aNode.removeAttribute("style");
} else if (aNode.style && aNode.style.cssText) {
var newCSStext = this.processCSSText(aNode.style.cssText, this.currentURL, true);
if (newCSStext) aNode.setAttribute("style", newCSStext);
}
if (!this.option["script"]) {
aNode.removeAttribute("onmouseover");
aNode.removeAttribute("onmouseout");
aNode.removeAttribute("onload");
}
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.processDOMNode]:\n -> aNode.nodeName: " + aNode.nodeName, ex);
}
return aNode;
},
// get through the DOM tree (recursiv function)
processDOMRecursively: function (rootNode) {
if (rootNode == null) return;
for (var curNode = rootNode.firstChild; curNode != null; curNode = curNode.nextSibling) {
if (curNode.nodeName != "#text" && curNode.nodeName != "#comment") {
curNode = this.processDOMNode(curNode);
this.processDOMRecursively(curNode);
} else if ((curNode.nodeName == "#text") && (wpdCommon.trim(curNode.nodeValue) != "")) {
// we need to replace special chars with HTML Entities
if (WPD_ENTITYBUG) curNode.nodeValue = this.convertEntity(curNode.nodeValue);
// if we have CRLFs before or after the text "innerhtml" will remove them,
// so we have to make sure that we preserve this CRLFs for the PRE Tag
if (WPD_CRLFBUG) curNode.nodeValue = wpdCommon.checkCRLF(curNode);
}
}
},
// Do a correction directly inside the final HTML text.
// This is necessary because setting the css text for the
// style attribute does not work - innerHTML will finally
// generate e.g "repeat scroll 0%;" regardless of the style setting
// (e.g. "repeat;")
repairInlineCSS: function (aHTMLText) {
if ((WPD_CSSSCROLLBUG) && (aHTMLText.match(/background:/i))) {
// Regex fixed by Dan for Zotero
//var re = new RegExp(/style=\"(.*)background:(.*)(repeat scroll 0(?:pt|px|%);)/);
var re = new RegExp(/style=\"([^\"]*)background:([^;\"]*)(repeat scroll 0(?:pt|px|%);?)/);
while (re.exec(aHTMLText)) {
var firstPart = RegExp.$1;
var secondPart = RegExp.$2;
// '?' added by Dan for Zotero
//var thirdPart = RegExp.$3.replace(/scroll 0(pt|px|%);/g, ';');
var thirdPart = RegExp.$3.replace(/scroll 0(pt|px|%);?/g, ';');
aHTMLText = aHTMLText.replace(re, "style=\"" + firstPart + "background:" + secondPart + thirdPart);
}
}
if ((WPD_CSSBACKGROUNDPOSITIONBUG) && (aHTMLText.match(/background-position: /i))) {
// Regex fixed by Dan for Zotero
//var re = new RegExp(/style=\"(.*)background-position: 0(?:pt|px|%);/);
var re = new RegExp(/style=\"([^\"]*)background-position: 0(?:pt|px|%);/);
while (re.exec(aHTMLText)) {
aHTMLText = aHTMLText.replace(re, "style=\"" + RegExp.$1 + "background-position: ;");
}
}
return aHTMLText;
},
// While we're replacing references with local file paths,
// we don't want to have the browser try and fetch them
// We prefix them with 'about:blank?' and remove later via repairRelativeLinks
relativeLinkFix: function (aFileName) {
return "about:blank?" + aFileName;
},
// Added by Dan S. for Zotero to restore relative links,
// which are prepended with "about:blank?" to fix a bug in Scrapbook/WPD
// that sending an invalid request to the server when the img src
// is a relative link to a file in a different directory
repairRelativeLinks: function (aHTMLText) {
return aHTMLText.replace(/(src|background|data|href)="about:blank\?([^"]*)"/g, '$1="$2"');
},
// process the CSS text of one stylesheet element
processCSSText: function (aCSStext, aCSShref, inline) {
if (!aCSStext) return "";
// search for "url" entries inside the css
// Double-quotes in regexp added by Dan S. for Zotero
var re = new RegExp(/ url\("?([^'")]+)"?\)/);
var i = 0;
while (aCSStext.match(re)) {
if (++i > 20) break; // safer (we try it maximal 20 times for one stylesheet element)
var imgFile = this.download(wpdCommon.resolveURL(aCSShref, RegExp.$1), true);
aCSStext = aCSStext.replace(re, " url('" + imgFile + "')");
}
// search for "content" entries inside the css and clean "attr"
re = new RegExp(/ content: \"(.*?)\"; /);
if (aCSStext.match(re)) {
var innerQuote = RegExp.$1;
innerQuote = innerQuote.replace(/\"/g, '\\"');
innerQuote = innerQuote.replace(/\\\" attr\(([^\)]+)\) \\\"/g, '" attr($1) "');
aCSStext = aCSStext.replace(re, ' content: "' + innerQuote + '"; ');
}
//
if ((WPD_CSSSCROLLBUG) && (aCSStext.match(/background: /i))) aCSStext = aCSStext.replace(/ scroll 0(pt|px|%);/g, ";");
if ((WPD_CSSBACKGROUNDPOSITIONBUG) && (aCSStext.match(/background-position: /i))) aCSStext = aCSStext.replace(/ background-position: 0(pt|px|%);/g, ";");
return aCSStext;
},
// process the CSS stylesheets (recursively)
// CSS Types:
// UNKNOWN_RULE = 0,
// STYLE_RULE = 1,
// CHARSET_RULE = 2,
// IMPORT_RULE = 3,
// MEDIA_RULE = 4,
// FONT_FACE_RULE = 5,
// PAGE_RULE = 6
processCSSRecursively: function (aCSS) {
if (!aCSS || aCSS.disabled) return "";
var content = "";
var medium = aCSS.media.mediaText;
if (medium != "" && medium.indexOf("screen") < 0 && medium.indexOf("all") < 0) {
return "";
}
// Disabled by Dan S. to fix CSS on snapshots of Reader View
//if (aCSS.href != null && aCSS.href.indexOf("chrome") == 0) return "";
var flag = "";
// Added by Dan S. for Zotero
//
// Make sure cssRules is accessible -- it might not be if a <link>
// element appears within <body> instead of <head>
try {
aCSS.cssRules
} catch (e) {
var msg = "Unable to access cssRules property of " + aCSS.href + " in wpdDOMSaver.processCSSRecursively()";
Zotero.debug("WebPageDump: " + msg, 2);
Components.utils.reportError(msg);
return "";
}
for (var i = 0; i < aCSS.cssRules.length; i++) {
if (aCSS.cssRules[i].type == 1 || aCSS.cssRules[i].type == 4) {
if (flag == "") {
content += "\n/* ::::: " + aCSS.href + " ::::: */\n\n"; // write original css filename
flag = aCSS.href;
}
var ref = aCSS.href;
if (flag == null || flag.indexOf(".css") == -1) ref = this.currentURL;
content += this.processCSSText(aCSS.cssRules[i].cssText, ref, false) + "\n";
} else if (aCSS.cssRules[i].type == 3) {
content += this.processCSSRecursively(aCSS.cssRules[i].styleSheet);
}
}
return content;
},
//given a file name and source URL (optional) with content (optional)
//returns a unique file name and registers it
getUniqueFileNameAndRegister: function(fileName, sourceURL, content) {
fileName = this.checkForEqualFilenames(
wpdCommon.getValidFileName(fileName),
sourceURL);
this.registerFile(fileName, sourceURL, content);
return fileName;
},
//register filename, so we don't overwrite them later
registerFile: function (newFileName, sourceURL, content) {
this.fileInfo[newFileName.toLowerCase()] = {
url: sourceURL,
downloaded: content
}
},
// is the file registered (e.g. downloaded)?
isFileRegistered: function (newFileName) {
if (this.fileInfo[newFileName.toLowerCase()] != undefined) return true;
return false;
},
isDownloaded: function(fileName) {
fileName = fileName.toLowerCase();
if(!this.fileInfo[fileName]) return;
return this.fileInfo[fileName].downloaded;
},
// check for equal Filenames with different locations
// if this is the case, we generate a new name
// if no aURLSpec is passed, this generates a unique file name
checkForEqualFilenames: function (newFileName, aURLSpec) {
if (this.isFileRegistered(newFileName)) {
if (!aURLSpec || this.fileInfo[newFileName.toLowerCase()]["url"] != aURLSpec) {
// the file is already registered but from a different location
// => probably not the same file, so we have to find a different name it (e.g. filename_001.ext)
var seq = 1;
var fileLR = wpdCommon.splitFileName(newFileName);
if (!fileLR[1]) fileLR[1] = "dat";
newFileName = fileLR[0] + "_" + wpdCommon.addLeftZeros(seq++, 3) + "." + fileLR[1];
while (this.fileInfo[newFileName.toLowerCase()] != undefined) {
// is the file already registered with the new name?
if (aURLSpec && this.fileInfo[newFileName.toLowerCase()]["url"] == aURLSpec) return newFileName; // Yes -> so it's already downloaded and we are finished
newFileName = fileLR[0] + "_" + wpdCommon.addLeftZeros(seq++, 3) + "." + fileLR[1]; // No -> "increment" filename
}
}
}
return newFileName;
},
// Download the specified URL to "this.currentDir". Takes
// care about equal filenames from different locations
download: function (aURLSpec, aDownload) {
if (!aURLSpec) return "";
// is this a relative URL (no protocol present) which needs to be resolved?
if (aURLSpec.indexOf("://") < 0) aURLSpec = wpdCommon.resolveURL(this.currentURL, aURLSpec);
try {
var aURL = wpdCommon.convertURLToObject(aURLSpec);
// generate a filename
var newFileName = aURL.fileName;
if (!newFileName) newFileName = "untitled";
// same name but different location?
newFileName = this.getUniqueFileNameAndRegister(newFileName, aURLSpec);
// is the file already registered (processed) ?
if (!this.isDownloaded(newFileName)) {
if (aDownload) {
aDownload = wpdCommon.downloadFile(aURLSpec, this.currentDir + newFileName);
} else {
aDownload = true;
}
this.registerFile(newFileName, aURLSpec, aDownload);
}
return newFileName;
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.download]\n -> aURLSpec: " + aURLSpec, ex);
return "";
}
},
// Get a CSS filename node for inserting in the DOM Tree
createCSSFileNode: function (aDocument, rootNode, aFileName) {
var newLinkNode = aDocument.createElement("link");
rootNode.firstChild.appendChild(aDocument.createTextNode("\n"));
newLinkNode.setAttribute("media", "all");
newLinkNode.setAttribute("href", aFileName);
newLinkNode.setAttribute("type", "text/css");
newLinkNode.setAttribute("rel", "stylesheet");
rootNode.firstChild.appendChild(newLinkNode);
rootNode.firstChild.appendChild(aDocument.createTextNode("\n"));
//return newLinkNode;
},
// Creates a placeholder node for inserting the DOCTYPE after the html tag
createPseudeDocTypeNode: function (aDocument, rootNode) {
var aDoctype = aDocument.doctype;
if (!aDoctype) return;
try {
rootNode.insertBefore(aDocument.createTextNode("\n"), rootNode.firstChild);
var metaNode = aDocument.createElement("wpd_doctype");
rootNode.insertBefore(metaNode, rootNode.firstChild);
rootNode.insertBefore(aDocument.createTextNode("\n"), rootNode.firstChild);
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.createDocTypeNode]", ex);
}
},
// replaces the placeholder node generated by createPseudeDocTypeNode with the DOCTYPE
replaceDocType: function (aDocument, aHTMLText) {
var aDoctype = aDocument.doctype;
if (!aDoctype) return aHTMLText;
try {
return aHTMLText.replace("<wpd_doctype></wpd_doctype>", this.getDocType(aDocument));
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.replaceDocType]", ex);
}
return aHTMLText;
},
// Returns the HTML Text generated from rootNode and does
// some processing (WPD_DOCTYPEBUG, WPD_ENTITYBUG, cleaning,...)
generateHTMLString: function (aDocument, rootNode) {
if (WPD_DOCTYPEBUG) this.createPseudeDocTypeNode(aDocument, rootNode);
var HTMLText = wpdCommon.nodeToHTMLString(rootNode);
if (WPD_DOCTYPEBUG) HTMLText = this.replaceDocType(aDocument, HTMLText);
// adding the doctype entry at the top
HTMLText = this.getDocType(aDocument) + HTMLText;
HTMLText = HTMLText.replace(/\x00/g, " ");
// replace the &amp; added by the innerHTML method
// because we have already generated all entities
if (WPD_ENTITYBUG) HTMLText = HTMLText.replace(/&amp;/g, "&");
// Added by Dan S. for Zotero
HTMLText = this.repairRelativeLinks(HTMLText);
return this.repairInlineCSS(HTMLText);
},
// Returns a DOCTYPE definition string based on aDocument.doctype
getDocType: function (aDocument) {
var aDoctype = aDocument.doctype;
if (!aDoctype) return "";
var dt = "<!DOCTYPE " + aDoctype.name;
if (aDoctype.publicId) dt += ' PUBLIC "' + aDoctype.publicId + '"';
if (aDoctype.systemId) dt += ' "' + aDoctype.systemId + '"';
dt += ">\n";
return dt;
},
// Get the meta charset information from the document
getMetaCharset: function (aDocument) {
var metas = aDocument.getElementsByTagName("meta");
for (var i = metas.length; --i >= 0;) {
var meta = metas[i];
if (/content-type/i.test(meta.httpEquiv)) {
r = /^text\/html; *charset=(.*)$/i.exec(meta.content);
return r[1];
}
}
return "";
},
// Create and return a meta charset node for the DOM Tree
createMetaCharsetNode: function (aDocument, rootNode, aContentType, aCharSet) {
try {
var metaNode = aDocument.createElement("meta");
rootNode.firstChild.insertBefore(aDocument.createTextNode("\n"), rootNode.firstChild.firstChild);
metaNode.setAttribute("content", aContentType + "; charset=" + aCharSet);
metaNode.setAttribute("http-equiv", "Content-Type");
rootNode.firstChild.insertBefore(metaNode, rootNode.firstChild.firstChild);
rootNode.firstChild.insertBefore(aDocument.createTextNode("\n"), rootNode.firstChild.firstChild);
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.createMetaCharsetNode]", ex);
}
},
// get a meta node for the DOM Tree
createMetaNameNode: function (aDocument, rootNode, name, content) {
try {
var metaNode = aDocument.createElement("meta");
metaNode.setAttribute("content", content);
metaNode.setAttribute("name", name);
rootNode.firstChild.insertBefore(aDocument.createTextNode("\n"), rootNode.firstChild.firstChild);
rootNode.firstChild.insertBefore(metaNode, rootNode.firstChild.firstChild);
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.createMetaNameNode]", ex);
}
},
/*existMetaCharsetNode : function(aDocument);
{
var metaNodes = aDocument.getElementsByTagName("meta");
for (var i=0; i<metaNodes.length; i++ ) {
if ( (metaNodes[i].hasAttribute("http-equiv") && metaNodes[i].hasAttribute("content")) &&
(metaNodes[i].getAttribute("http-equiv").toLowerCase() == "content-type") &&
(metaNodes[i].getAttribute("content").match(/charset\=/i)) )
return true;
}
return false;
}*/
// Return the WPD Meta Base URL Information from aFile
getMetaBaseURL: function (aFile) {
if (wpdCommon.pathExists(aFile)) {
str = new String(wpdCommon.readFile(aFile, false, true));
re = new RegExp(/<meta name=\"wpd_baseurl\" content=\"(.*?)\">/);
if (str.match(re)) {
return RegExp.$1;
}
}
return "";
},
// Return the WPD Meta Date Information from aFile
getMetaDate: function (aFile) {
if (wpdCommon.pathExists(aFile)) {
str = new String(wpdCommon.readFile(aFile, false, true));
re = new RegExp(/<meta name=\"wpd_date\" content=\"(.*?)\">/);
if (str.match(re)) {
return RegExp.$1;
}
}
return "";
},
// creates the meta nodes for the wpd meta tags (version, baseurl, url, date/time)
createMetaInformation: function (aDocument, rootNode) {
// insert url/date/time meta information
//
var d = this.dateObj.getUTCFullYear() + "-" + wpdCommon.addLeftZeros(this.dateObj.getUTCMonth(), 2) + "-" + wpdCommon.addLeftZeros(this.dateObj.getUTCDate(), 2);
d = d + "T" + wpdCommon.addLeftZeros(this.dateObj.getUTCHours(), 2) + ":" + wpdCommon.addLeftZeros(this.dateObj.getUTCMinutes(), 2) + "Z";
this.createMetaNameNode(aDocument, rootNode, "wpd_date", d);
this.createMetaNameNode(aDocument, rootNode, "wpd_url", this.currentURL);
this.createMetaNameNode(aDocument, rootNode, "wpd_baseurl", this.baseURL);
this.createMetaNameNode(aDocument, rootNode, "wpd_version", WPD_VERSION);
rootNode.firstChild.insertBefore(aDocument.createTextNode("\n\n"), rootNode.firstChild.firstChild);
},
// save a non HTML "aDocument" as "aFileName" and generate a
// wrapper HTML File which references "aDocument"
// ("aFileName" is the filename without(!) extension)
saveDocumentFile: function (aDocument, aFileName) {
Zotero.debug("[wpdDOMSaver.saveDocumentFile]: Saving file from " + this.currentURL);
aFileName = this.download(this.currentURL, true)
Zotero.debug("[wpdDOMSaver.saveDocumentFile]: Saved to " + aFileName);
return aFileName;
/* Wrapper file disabled by Dan S. for Zotero
var aFileURL = aDocument.location.href;
if ( !aFileName ) aFileName = "file" + Math.random().toString();
// this.download will generate a unique filename
var newFileName = this.download(this.currentURL,true);
if ( aDocument.contentType.substring(0,5) == "image" ) {
var HTMLText = '<html><body><img src="' + newFileName + '"></body></html>';
} else {
var HTMLText = '<html><head><meta http-equiv="refresh" content="0;URL=' + newFileName + '"></head><body></body></html>';
}
var HTMLFile = this.currentDir + aFileName + ".html";
if (!wpdCommon.writeFile(HTMLText,HTMLFile))
wpdCommon.addError("[wpdDOMSaver.saveDocumentFile]: could not write HTML wrapper for "+aFileName+"\n");
return aFileName + ".html";
*/
},
// save the CSS Stylesheets of "aDocument" as "aFileName" and
// process the CSS Text
// "aFileName" is the filename without(!) extension
// (".css" will be added)
saveDocumentCSS: function (aDocument, aFileName) {
var CSSText = ""; //"body {display: block;margin: 8px;}; ";
if (this.option["format"]) {
var myStyleSheets = aDocument.styleSheets;
// get all style sheets to "CSSText"
for (var i = 0; i < myStyleSheets.length; i++) {
CSSText += this.processCSSRecursively(myStyleSheets[i]);
}
if (CSSText) {
// don't forget to convert the CSS String to the document charset..
// (necessary for e.g. font-family)
if (this.option["encodeUTF8"]) {
CSSText = wpdCommon.ConvertFromUnicode16(CSSText, "UTF-8");
} else {
CSSText = wpdCommon.ConvertFromUnicode16(CSSText, this.curCharacterSet);
}
aFileName = this.getUniqueFileNameAndRegister(aFileName + ".css");
Zotero.debug("[wpdDOMSaver.saveDocumentCSS]: " + this.currentDir + aFileName);
// write css file
var CSSFile = this.currentDir + aFileName;
if (!wpdCommon.writeFile(CSSText, CSSFile)) wpdCommon.addError("[wpdDOMSaver.saveDocumentCSS]: could not write CSS File");
return aFileName;
}
}
return false;
},
// save the HTML "aDocument" as "aFileName" and process the
// DOM Tree (see processDOMNode) - calls also saveDocumentCSS
// "aFileName" is the filename without(!) extension
// (".html" will be added)
saveDocumentHTML: function (aDocument, aFileName) {
aFileName = this.getUniqueFileNameAndRegister(aFileName + ".html");
var aFileNameNoExt = wpdCommon.splitFileName(aFileName)[0];
Zotero.debug("[wpdDOMSaver.saveDocumentHTML]: " + this.currentDir + aFileName);
this.curDocument = aDocument;
this.curCharacterSet = aDocument.characterSet;
var charset = this.curCharacterSet;
// we get the html node without childs and add the head and body trees
// manually so we are sure that we have a correct html file
var rootNode = aDocument.getElementsByTagName("html")[0].cloneNode(false);
try {
var headNode = aDocument.getElementsByTagName("head")[0].cloneNode(true);
rootNode.appendChild(headNode);
rootNode.appendChild(aDocument.createTextNode("\n"));
} catch (ex) {}
try {
this.curBody = aDocument.body.cloneNode(true);
} catch (ex) {
this.curBody = aDocument.getElementsByTagName("body")[0].cloneNode(true);
}
rootNode.appendChild(this.curBody);
rootNode.appendChild(aDocument.createTextNode("\n"));
// now the processing of the dom nodes (changing hrefs, downloading...)
this.processDOMRecursively(rootNode);
// write css file and add css node with the new css filename in the DOM Tree
var cssFileName = this.saveDocumentCSS(aDocument, aFileNameNoExt);
if (cssFileName) this.createCSSFileNode(aDocument, rootNode, cssFileName);
// create meta information (version, base_url, url, date/time)
if (this.option["metainfo"]) this.createMetaInformation(aDocument, rootNode);
// add the charset defintions previously removed by processDOMNode
if (this.option["encodeUTF8"]) {
this.createMetaCharsetNode(aDocument, rootNode, aDocument.contentType, "UTF-8");
} else {
// charset probably sent by web server only -> add the charset meta header for local viewing
this.createMetaCharsetNode(aDocument, rootNode, aDocument.contentType, charset);
}
// convert the nodes to a html string (including some processing)
// "var " added by Dan S. for Zotero
var HTMLText = this.generateHTMLString(aDocument, rootNode);
// convert the DOM String to the desired Charset
if (this.option["encodeUTF8"]) {
HTMLText = wpdCommon.ConvertFromUnicode16(HTMLText, "UTF-8");
} else {
HTMLText = wpdCommon.ConvertFromUnicode16(HTMLText, charset);
}
this.curCharacterSet = charset;
// and write the file...
var HTMLFile = this.currentDir + aFileName;
if (!wpdCommon.writeFile(HTMLText, HTMLFile)) wpdCommon.addError("[wpdDOMSaver.saveDocumentHTML]: could not write HTML File");
return aFileName;
},
// Decides the calling of SaveDocumentFile or saveDocumentHTML
saveDocumentEx: function (aDocument, aFileName) {
// we have to set a new current url which is the
// base reference url (necessary for frame processing)
this.currentURL = aDocument.location.href;
// distinguish between HTML Documents and other
// embedded files like flash, video or images...
if ((aDocument.getElementsByTagName("head").length == 0) || !aDocument.contentType.match(/htm|html|xml/i)) {
aFileName = this.saveDocumentFile(aDocument, aFileName);
} else {
aFileName = this.saveDocumentHTML(aDocument, aFileName)
}
// set the current URL back to the original base URL
this.currentURL = this.baseURL;
return aFileName;
},
// Main Routine: call it for saving the actual active top window
// (be sure to call the init function at the top of this file before)
saveHTMLDocument: function () {
try {
return this.saveDocumentEx(this.document, this.name);
} catch (ex) {
wpdCommon.addError("[wpdDOMSaver.saveHTMLDocument]", ex);
}
}
};
Reference in a new issue View git blame Copy permalink