[epiphany] Update Readability.js
- From: Jan-Michael Brummer <jbrummer src gnome org>
- To: commits-list gnome org
- Cc:
- Subject: [epiphany] Update Readability.js
- Date: Thu, 4 Jun 2020 18:06:25 +0000 (UTC)
commit 2683bba39a3033cf90a82c558588fa08a00801d4
Author: Jan-Michael Brummer <jan brummer tabos org>
Date: Thu Jun 4 19:06:56 2020 +0200
Update Readability.js
Update Readability.js to latest revision and move it to third_party with
an own README.epiphany.
embed/ephy-web-view.c | 4 +-
embed/meson.build | 2 +-
src/meson.build | 3 +-
src/resources/epiphany.gresource.xml | 2 -
tests/meson.build | 1 +
third-party/meson.build | 7 +
third-party/readability/README.epiphany | 18 ++
.../readability}/Readability.js | 322 ++++++++++++++++++---
third-party/readability/readability.gresource.xml | 7 +
.../readability}/reader.css | 0
10 files changed, 324 insertions(+), 42 deletions(-)
---
diff --git a/embed/ephy-web-view.c b/embed/ephy-web-view.c
index 8132ac011..6ebcad0b5 100644
--- a/embed/ephy-web-view.c
+++ b/embed/ephy-web-view.c
@@ -731,7 +731,7 @@ run_readability_js_if_needed (gpointer data)
/* Internal pages should never receive reader mode. */
if (!ephy_embed_utils_is_no_show_address (web_view->address)) {
webkit_web_view_run_javascript_from_gresource (WEBKIT_WEB_VIEW (web_view),
- "/org/gnome/epiphany/Readability.js",
+ "/org/gnome/epiphany/readability/Readability.js",
web_view->cancellable,
readability_js_finish_cb,
web_view);
@@ -3454,7 +3454,7 @@ ephy_web_view_toggle_reader_mode (EphyWebView *view,
view->reader_url = g_strdup (ephy_web_view_get_address (view));
html = g_string_new ("");
- style_css = g_resources_lookup_data ("/org/gnome/epiphany/reader.css", G_RESOURCE_LOOKUP_FLAGS_NONE, NULL);
+ style_css = g_resources_lookup_data ("/org/gnome/epiphany/readability/reader.css",
G_RESOURCE_LOOKUP_FLAGS_NONE, NULL);
title = webkit_web_view_get_title (web_view);
font_style = enum_nick (EPHY_TYPE_PREFS_READER_FONT_STYLE,
g_settings_get_enum (EPHY_SETTINGS_READER,
diff --git a/embed/meson.build b/embed/meson.build
index 917334c05..d78c261c4 100644
--- a/embed/meson.build
+++ b/embed/meson.build
@@ -43,7 +43,7 @@ libephyembed_deps = [
libsecret_dep,
libsoup_dep,
m_dep,
- webkit2gtk_dep,
+ webkit2gtk_dep
]
libephyembed_includes = include_directories(
diff --git a/src/meson.build b/src/meson.build
index 50234bf6f..03999ca59 100644
--- a/src/meson.build
+++ b/src/meson.build
@@ -136,7 +136,8 @@ epiphany_sources = [
'ephy-main.c',
resources,
pdfjs_resources,
- highlightjs_resources
+ highlightjs_resources,
+ readability_resources,
]
executable('epiphany',
diff --git a/src/resources/epiphany.gresource.xml b/src/resources/epiphany.gresource.xml
index de5e2b5dc..055015b80 100644
--- a/src/resources/epiphany.gresource.xml
+++ b/src/resources/epiphany.gresource.xml
@@ -42,8 +42,6 @@
<file preprocess="xml-stripblanks" compressed="true">gtk/shortcuts-dialog.ui</file>
<file preprocess="xml-stripblanks" compressed="true">gtk/tab-label.ui</file>
<file preprocess="xml-stripblanks" compressed="true">gtk/webapp-additional-urls-dialog.ui</file>
- <file compressed="true">Readability.js</file>
- <file compressed="true">reader.css</file>
</gresource>
<gresource prefix="/org/gnome/Epiphany/icons">
<file compressed="true" alias="scalable/actions/ephy-download-symbolic.svg"
preprocess="xml-stripblanks">ephy-download-symbolic.svg</file>
diff --git a/tests/meson.build b/tests/meson.build
index d879d6aea..00b8f5d14 100644
--- a/tests/meson.build
+++ b/tests/meson.build
@@ -178,6 +178,7 @@ if get_option('unit_tests').enabled()
web_view_test = executable('test-ephy-web-view',
'ephy-web-view-test.c',
resources,
+ readability_resources,
dependencies: ephymain_dep
)
test('Web view test',
diff --git a/third-party/meson.build b/third-party/meson.build
index 72027a07f..43bd89311 100644
--- a/third-party/meson.build
+++ b/third-party/meson.build
@@ -12,6 +12,13 @@ highlightjs_resources = gnome.compile_resources('highlightjs-resources',
source_dir: 'highlightjs'
)
+readability_resource_files = files('readability/readability.gresource.xml')
+readability_resources = gnome.compile_resources('readability-resources',
+ readability_resource_files,
+ c_name: 'readability',
+ source_dir: 'readability'
+)
+
libgvdb_sources = [
'gvdb/gvdb-builder.c',
'gvdb/gvdb-reader.c'
diff --git a/third-party/readability/README.epiphany b/third-party/readability/README.epiphany
new file mode 100644
index 000000000..731426a81
--- /dev/null
+++ b/third-party/readability/README.epiphany
@@ -0,0 +1,18 @@
+# Embedded readability mode based on Readability.js
+
+This directory contains an official readability.js release version, distributed at:
https://github.com/mozilla/readability
+
+## Update process
+
+$ wget https://raw.githubusercontent.com/mozilla/readability/master/Readability.js
+
+Copy Readability.js to <epiphany-source>/third-party/readability/
+
+# Added the following to the bottom of the js file:
+
+// Added for Epiphany
+var documentClone = document.cloneNode(true);
+reader = new Readability(documentClone);
+reader.parse();
+
+# Documentation created by Jan-Michael Brummer <jan brummer tabos org>
\ No newline at end of file
diff --git a/src/resources/Readability.js b/third-party/readability/Readability.js
similarity index 84%
rename from src/resources/Readability.js
rename to third-party/readability/Readability.js
index 2c716b164..8af4f072e 100644
--- a/src/resources/Readability.js
+++ b/third-party/readability/Readability.js
@@ -1,8 +1,3 @@
-//////////////////////////////////////////////////////////////////////////
-// Warning: Epiphany changes at the bottom of the file. //
-// https://github.com/mozilla/readability distributed under Apache V2.0 //
-//////////////////////////////////////////////////////////////////////////
-
/*eslint-env es6:false*/
/*
* Copyright (c) 2010 Arc90 Inc
@@ -41,6 +36,7 @@ function Readability(doc, options) {
options = options || {};
this._doc = doc;
+ this._docJSDOMParser = this._doc.firstChild.__JSDOMParser__;
this._articleTitle = null;
this._articleByline = null;
this._articleDir = null;
@@ -53,6 +49,7 @@ function Readability(doc, options) {
this._nbTopCandidates = options.nbTopCandidates || this.DEFAULT_N_TOP_CANDIDATES;
this._charThreshold = options.charThreshold || this.DEFAULT_CHAR_THRESHOLD;
this._classesToPreserve = this.CLASSES_TO_PRESERVE.concat(options.classesToPreserve || []);
+ this._keepClasses = !!options.keepClasses;
// Start with all flags set
this._flags = this.FLAG_STRIP_UNLIKELYS |
@@ -119,8 +116,8 @@ Readability.prototype = {
REGEXPS: {
// NOTE: These two regular expressions are duplicated in
// Readability-readerable.js. Please keep both copies in sync.
- unlikelyCandidates:
/-ad-|ai2html|banner|breadcrumbs|combx|comment|community|cover-wrap|disqus|extra|foot|gdpr|header|legends|menu|related|remark|replies|rss|shoutbox|sidebar|skyscraper|social|sponsor|supplemental|ad-break|agegate|pagination|pager|popup|yom-remote/i,
- okMaybeItsACandidate: /and|article|body|column|main|shadow/i,
+ unlikelyCandidates:
/-ad-|ai2html|banner|breadcrumbs|combx|comment|community|cover-wrap|disqus|extra|footer|gdpr|header|legends|menu|related|remark|replies|rss|shoutbox|sidebar|skyscraper|social|sponsor|supplemental|ad-break|agegate|pagination|pager|popup|yom-remote/i,
+ okMaybeItsACandidate: /and|article|body|column|content|main|shadow/i,
positive: /article|body|content|entry|hentry|h-entry|main|page|pagination|post|text|blog|story/i,
negative: /hidden|^hid$| hid$| hid |^hid
|banner|combx|comment|com-|contact|foot|footer|footnote|gdpr|masthead|media|meta|outbrain|promo|related|scroll|share|shoutbox|sidebar|skyscraper|sponsor|shopping|tags|tool|widget/i,
@@ -129,10 +126,13 @@ Readability.prototype = {
replaceFonts: /<(\/?)font[^>]*>/gi,
normalize: /\s{2,}/g,
videos:
/\/\/(www\.)?((dailymotion|youtube|youtube-nocookie|player\.vimeo|v\.qq)\.com|(archive|upload\.wikimedia)\.org|player\.twitch\.tv)/i,
+ shareElements: /(\b|_)(share|sharedaddy)(\b|_)/i,
nextLink: /(next|weiter|continue|>([^\|]|$)|»([^\|]|$))/i,
prevLink: /(prev|earl|old|new|<|«)/i,
whitespace: /^\s*$/,
hasContent: /\S$/,
+ srcsetUrl: /(\S+)(\s+[\d.]+[xw])?(\s*(?:,|$))/g,
+ b64DataUrl: /^data:\s*([^\s;,]+)\s*;\s*base64\s*,/i
},
DIV_TO_P_ELEMS: [ "A", "BLOCKQUOTE", "DL", "DIV", "IMG", "OL", "P", "PRE", "TABLE", "UL", "SELECT" ],
@@ -157,6 +157,15 @@ Readability.prototype = {
// These are the classes that readability sets itself.
CLASSES_TO_PRESERVE: [ "page" ],
+ // These are the list of HTML entities that need to be escaped.
+ HTML_ESCAPE_MAP: {
+ "lt": "<",
+ "gt": ">",
+ "amp": "&",
+ "quot": '"',
+ "apos": "'",
+ },
+
/**
* Run any post-process modifications to article content as necessary.
*
@@ -167,8 +176,10 @@ Readability.prototype = {
// Readability cannot open relative uris so we convert them to absolute uris.
this._fixRelativeUris(articleContent);
- // Remove classes.
- this._cleanClasses(articleContent);
+ if (!this._keepClasses) {
+ // Remove classes.
+ this._cleanClasses(articleContent);
+ }
},
/**
@@ -182,6 +193,10 @@ Readability.prototype = {
* @return void
*/
_removeNodes: function(nodeList, filterFn) {
+ // Avoid ever operating on live node lists.
+ if (this._docJSDOMParser && nodeList._isLiveNodeList) {
+ throw new Error("Do not pass live node lists to _removeNodes");
+ }
for (var i = nodeList.length - 1; i >= 0; i--) {
var node = nodeList[i];
var parentNode = node.parentNode;
@@ -201,6 +216,10 @@ Readability.prototype = {
* @return void
*/
_replaceNodeTags: function(nodeList, newTagName) {
+ // Avoid ever operating on live node lists.
+ if (this._docJSDOMParser && nodeList._isLiveNodeList) {
+ throw new Error("Do not pass live node lists to _replaceNodeTags");
+ }
for (var i = nodeList.length - 1; i >= 0; i--) {
var node = nodeList[i];
this._setNodeTag(node, newTagName);
@@ -320,6 +339,7 @@ Readability.prototype = {
if (baseURI == documentURI && uri.charAt(0) == "#") {
return uri;
}
+
// Otherwise, resolve against base URI:
try {
return new URL(uri, baseURI).href;
@@ -333,22 +353,50 @@ Readability.prototype = {
this._forEachNode(links, function(link) {
var href = link.getAttribute("href");
if (href) {
- // Replace links with javascript: URIs with text content, since
+ // Remove links with javascript: URIs, since
// they won't work after scripts have been removed from the page.
if (href.indexOf("javascript:") === 0) {
- var text = this._doc.createTextNode(link.textContent);
- link.parentNode.replaceChild(text, link);
+ // if the link only contains simple text content, it can be converted to a text node
+ if (link.childNodes.length === 1 && link.childNodes[0].nodeType === this.TEXT_NODE) {
+ var text = this._doc.createTextNode(link.textContent);
+ link.parentNode.replaceChild(text, link);
+ } else {
+ // if the link has multiple children, they should all be preserved
+ var container = this._doc.createElement("span");
+ while (link.childNodes.length > 0) {
+ container.appendChild(link.childNodes[0]);
+ }
+ link.parentNode.replaceChild(container, link);
+ }
} else {
link.setAttribute("href", toAbsoluteURI(href));
}
}
});
- var imgs = this._getAllNodesWithTag(articleContent, ["img"]);
- this._forEachNode(imgs, function(img) {
- var src = img.getAttribute("src");
+ var medias = this._getAllNodesWithTag(articleContent, [
+ "img", "picture", "figure", "video", "audio", "source"
+ ]);
+
+ this._forEachNode(medias, function(media) {
+ var src = media.getAttribute("src");
+ var poster = media.getAttribute("poster");
+ var srcset = media.getAttribute("srcset");
+
if (src) {
- img.setAttribute("src", toAbsoluteURI(src));
+ media.setAttribute("src", toAbsoluteURI(src));
+ }
+
+ if (poster) {
+ media.setAttribute("poster", toAbsoluteURI(poster));
+ }
+
+ if (srcset) {
+ var newSrcset = srcset.replace(this.REGEXPS.srcsetUrl, function(_, p1, p2, p3) {
+ return toAbsoluteURI(p1) + (p2 || "") + p3;
+ });
+
+ media.setAttribute("srcset", newSrcset);
}
});
},
@@ -442,13 +490,13 @@ Readability.prototype = {
var doc = this._doc;
// Remove all style tags in head
- this._removeNodes(doc.getElementsByTagName("style"));
+ this._removeNodes(this._getAllNodesWithTag(doc, ["style"]));
if (doc.body) {
this._replaceBrs(doc.body);
}
- this._replaceNodeTags(doc.getElementsByTagName("font"), "SPAN");
+ this._replaceNodeTags(this._getAllNodesWithTag(doc, ["font"]), "SPAN");
},
/**
@@ -528,7 +576,7 @@ Readability.prototype = {
_setNodeTag: function (node, tag) {
this.log("_setNodeTag", node, tag);
- if (node.__JSDOMParser__) {
+ if (this._docJSDOMParser) {
node.localName = tag.toLowerCase();
node.tagName = tag.toUpperCase();
return node;
@@ -572,6 +620,8 @@ Readability.prototype = {
// visually linked to other content-ful elements (text, images, etc.).
this._markDataTables(articleContent);
+ this._fixLazyImages(articleContent);
+
// Clean out junk from the article content
this._cleanConditionally(articleContent, "form");
this._cleanConditionally(articleContent, "fieldset");
@@ -589,7 +639,7 @@ Readability.prototype = {
this._forEachNode(articleContent.children, function (topCandidate) {
this._cleanMatchedNodes(topCandidate, function (node, matchString) {
- return /share/.test(matchString) && node.textContent.length < shareElementThreshold;
+ return this.REGEXPS.shareElements.test(matchString) && node.textContent.length <
shareElementThreshold;
});
});
@@ -626,7 +676,7 @@ Readability.prototype = {
this._cleanConditionally(articleContent, "div");
// Remove extra paragraphs
- this._removeNodes(articleContent.getElementsByTagName("p"), function (paragraph) {
+ this._removeNodes(this._getAllNodesWithTag(articleContent, ["p"]), function (paragraph) {
var imgCount = paragraph.getElementsByTagName("img").length;
var embedCount = paragraph.getElementsByTagName("embed").length;
var objectCount = paragraph.getElementsByTagName("object").length;
@@ -820,6 +870,12 @@ Readability.prototype = {
node = this._removeAndGetNext(node);
continue;
}
+
+ if (node.getAttribute("role") == "complementary") {
+ this.log("Removing complementary content - " + matchString);
+ node = this._removeAndGetNext(node);
+ continue;
+ }
}
// Remove DIV, SECTION, and HEADER nodes without any content(e.g. text, image, video, or iframe).
@@ -1213,6 +1269,26 @@ Readability.prototype = {
return false;
},
+ /**
+ * Converts some of the common HTML entities in string to their corresponding characters.
+ *
+ * @param str {string} - a string to unescape.
+ * @return string without HTML entity.
+ */
+ _unescapeHtmlEntities: function(str) {
+ if (!str) {
+ return str;
+ }
+
+ var htmlEscapeMap = this.HTML_ESCAPE_MAP;
+ return str.replace(/&(quot|amp|apos|lt|gt);/g, function(_, tag) {
+ return htmlEscapeMap[tag];
+ }).replace(/&#(?:x([0-9a-z]{1,4})|([0-9]{1,4}));/gi, function(_, hex, numStr) {
+ var num = parseInt(hex || numStr, hex ? 16 : 10);
+ return String.fromCharCode(num);
+ });
+ },
+
/**
* Attempts to get excerpt and byline metadata for the article.
*
@@ -1293,21 +1369,123 @@ Readability.prototype = {
// get site name
metadata.siteName = values["og:site_name"];
+ // in many sites the meta value is escaped with HTML entities,
+ // so here we need to unescape it
+ metadata.title = this._unescapeHtmlEntities(metadata.title);
+ metadata.byline = this._unescapeHtmlEntities(metadata.byline);
+ metadata.excerpt = this._unescapeHtmlEntities(metadata.excerpt);
+ metadata.siteName = this._unescapeHtmlEntities(metadata.siteName);
+
return metadata;
},
+ /**
+ * Check if node is image, or if node contains exactly only one image
+ * whether as a direct child or as its descendants.
+ *
+ * @param Element
+ **/
+ _isSingleImage: function(node) {
+ if (node.tagName === "IMG") {
+ return true;
+ }
+
+ if (node.children.length !== 1 || node.textContent.trim() !== "") {
+ return false;
+ }
+
+ return this._isSingleImage(node.children[0]);
+ },
+
+ /**
+ * Find all <noscript> that are located after <img> nodes, and which contain only one
+ * <img> element. Replace the first image with the image from inside the <noscript> tag,
+ * and remove the <noscript> tag. This improves the quality of the images we use on
+ * some sites (e.g. Medium).
+ *
+ * @param Element
+ **/
+ _unwrapNoscriptImages: function(doc) {
+ // Find img without source or attributes that might contains image, and remove it.
+ // This is done to prevent a placeholder img is replaced by img from noscript in next step.
+ var imgs = Array.from(doc.getElementsByTagName("img"));
+ this._forEachNode(imgs, function(img) {
+ for (var i = 0; i < img.attributes.length; i++) {
+ var attr = img.attributes[i];
+ switch (attr.name) {
+ case "src":
+ case "srcset":
+ case "data-src":
+ case "data-srcset":
+ return;
+ }
+
+ if (/\.(jpg|jpeg|png|webp)/i.test(attr.value)) {
+ return;
+ }
+ }
+
+ img.parentNode.removeChild(img);
+ });
+
+ // Next find noscript and try to extract its image
+ var noscripts = Array.from(doc.getElementsByTagName("noscript"));
+ this._forEachNode(noscripts, function(noscript) {
+ // Parse content of noscript and make sure it only contains image
+ var tmp = doc.createElement("div");
+ tmp.innerHTML = noscript.innerHTML;
+ if (!this._isSingleImage(tmp)) {
+ return;
+ }
+
+ // If noscript has previous sibling and it only contains image,
+ // replace it with noscript content. However we also keep old
+ // attributes that might contains image.
+ var prevElement = noscript.previousElementSibling;
+ if (prevElement && this._isSingleImage(prevElement)) {
+ var prevImg = prevElement;
+ if (prevImg.tagName !== "IMG") {
+ prevImg = prevElement.getElementsByTagName("img")[0];
+ }
+
+ var newImg = tmp.getElementsByTagName("img")[0];
+ for (var i = 0; i < prevImg.attributes.length; i++) {
+ var attr = prevImg.attributes[i];
+ if (attr.value === "") {
+ continue;
+ }
+
+ if (attr.name === "src" || attr.name === "srcset" || /\.(jpg|jpeg|png|webp)/i.test(attr.value)) {
+ if (newImg.getAttribute(attr.name) === attr.value) {
+ continue;
+ }
+
+ var attrName = attr.name;
+ if (newImg.hasAttribute(attrName)) {
+ attrName = "data-old-" + attrName;
+ }
+
+ newImg.setAttribute(attrName, attr.value);
+ }
+ }
+
+ noscript.parentNode.replaceChild(tmp.firstElementChild, prevElement);
+ }
+ });
+ },
+
/**
* Removes script tags from the document.
*
* @param Element
**/
_removeScripts: function(doc) {
- this._removeNodes(doc.getElementsByTagName("script"), function(scriptNode) {
+ this._removeNodes(this._getAllNodesWithTag(doc, ["script"]), function(scriptNode) {
scriptNode.nodeValue = "";
scriptNode.removeAttribute("src");
return true;
});
- this._removeNodes(doc.getElementsByTagName("noscript"));
+ this._removeNodes(this._getAllNodesWithTag(doc, ["noscript"]));
},
/**
@@ -1490,7 +1668,7 @@ Readability.prototype = {
_clean: function(e, tag) {
var isEmbed = ["object", "embed", "iframe"].indexOf(tag) !== -1;
- this._removeNodes(e.getElementsByTagName(tag), function(element) {
+ this._removeNodes(this._getAllNodesWithTag(e, [tag]), function(element) {
// Allow youtube and vimeo videos through as people usually want to see those.
if (isEmbed) {
// First, check the elements attributes to see if any of them contain youtube or vimeo
@@ -1621,6 +1799,76 @@ Readability.prototype = {
}
},
+ /* convert images and figures that have properties like data-src into images that can be loaded without JS
*/
+ _fixLazyImages: function (root) {
+ this._forEachNode(this._getAllNodesWithTag(root, ["img", "picture", "figure"]), function (elem) {
+ // In some sites (e.g. Kotaku), they put 1px square image as base64 data uri in the src attribute.
+ // So, here we check if the data uri is too short, just might as well remove it.
+ if (elem.src && this.REGEXPS.b64DataUrl.test(elem.src)) {
+ // Make sure it's not SVG, because SVG can have a meaningful image in under 133 bytes.
+ var parts = this.REGEXPS.b64DataUrl.exec(elem.src);
+ if (parts[1] === "image/svg+xml") {
+ return;
+ }
+
+ // Make sure this element has other attributes which contains image.
+ // If it doesn't, then this src is important and shouldn't be removed.
+ var srcCouldBeRemoved = false;
+ for (var i = 0; i < elem.attributes.length; i++) {
+ var attr = elem.attributes[i];
+ if (attr.name === "src") {
+ continue;
+ }
+
+ if (/\.(jpg|jpeg|png|webp)/i.test(attr.value)) {
+ srcCouldBeRemoved = true;
+ break;
+ }
+ }
+
+ // Here we assume if image is less than 100 bytes (or 133B after encoded to base64)
+ // it will be too small, therefore it might be placeholder image.
+ if (srcCouldBeRemoved) {
+ var b64starts = elem.src.search(/base64\s*/i) + 7;
+ var b64length = elem.src.length - b64starts;
+ if (b64length < 133) {
+ elem.removeAttribute("src");
+ }
+ }
+ }
+
+ // also check for "null" to work around https://github.com/jsdom/jsdom/issues/2580
+ if ((elem.src || (elem.srcset && elem.srcset != "null")) &&
elem.className.toLowerCase().indexOf("lazy") === -1) {
+ return;
+ }
+
+ for (var j = 0; j < elem.attributes.length; j++) {
+ attr = elem.attributes[j];
+ if (attr.name === "src" || attr.name === "srcset") {
+ continue;
+ }
+ var copyTo = null;
+ if (/\.(jpg|jpeg|png|webp)\s+\d/.test(attr.value)) {
+ copyTo = "srcset";
+ } else if (/^\s*\S+\.(jpg|jpeg|png|webp)\S*\s*$/.test(attr.value)) {
+ copyTo = "src";
+ }
+ if (copyTo) {
+ //if this is an img or picture, set the attribute directly
+ if (elem.tagName === "IMG" || elem.tagName === "PICTURE") {
+ elem.setAttribute(copyTo, attr.value);
+ } else if (elem.tagName === "FIGURE" && !this._getAllNodesWithTag(elem, ["img",
"picture"]).length) {
+ //if the item is a <figure> that does not contain an image or picture, create one and place it
inside the figure
+ //see the nytimes-3 testcase for an example
+ var img = this._doc.createElement("img");
+ img.setAttribute(copyTo, attr.value);
+ elem.appendChild(img);
+ }
+ }
+ }
+ });
+ },
+
/**
* Clean an element of all tags of type "tag" if they look fishy.
* "Fishy" is an algorithm based on content length, classnames, link density, number of images & embeds,
etc.
@@ -1638,7 +1886,7 @@ Readability.prototype = {
// without effecting the traversal.
//
// TODO: Consider taking into account original contentScore here.
- this._removeNodes(e.getElementsByTagName(tag), function(node) {
+ this._removeNodes(this._getAllNodesWithTag(e, [tag]), function(node) {
// First check if this node IS data table, in which case don't remove it.
var isDataTable = function(t) {
return t._readabilityDataTable;
@@ -1672,10 +1920,7 @@ Readability.prototype = {
var input = node.getElementsByTagName("input").length;
var embedCount = 0;
- var embeds = this._concatNodeLists(
- node.getElementsByTagName("object"),
- node.getElementsByTagName("embed"),
- node.getElementsByTagName("iframe"));
+ var embeds = this._getAllNodesWithTag(node, ["object", "embed", "iframe"]);
for (var i = 0; i < embeds.length; i++) {
// If this embed has attribute that matches video regex, don't delete it.
@@ -1721,7 +1966,7 @@ Readability.prototype = {
var endOfSearchMarkerNode = this._getNextNode(e, true);
var next = this._getNextNode(e);
while (next && next != endOfSearchMarkerNode) {
- if (filter(next, next.className + " " + next.id)) {
+ if (filter.call(this, next, next.className + " " + next.id)) {
next = this._removeAndGetNext(next);
} else {
next = this._getNextNode(next);
@@ -1736,11 +1981,9 @@ Readability.prototype = {
* @return void
**/
_cleanHeaders: function(e) {
- for (var headerIndex = 1; headerIndex < 3; headerIndex += 1) {
- this._removeNodes(e.getElementsByTagName("h" + headerIndex), function (header) {
- return this._getClassWeight(header) < 0;
- });
- }
+ this._removeNodes(this._getAllNodesWithTag(e, ["h1", "h2"]), function (header) {
+ return this._getClassWeight(header) < 0;
+ });
},
_flagIsActive: function(flag) {
@@ -1752,7 +1995,11 @@ Readability.prototype = {
},
_isProbablyVisible: function(node) {
- return (!node.style || node.style.display != "none") && !node.hasAttribute("hidden");
+ // Have to null-check node.style and node.className.indexOf to deal with SVG and MathML nodes.
+ return (!node.style || node.style.display != "none")
+ && !node.hasAttribute("hidden")
+ //check for "fallback-image" so that wikimedia math images are displayed
+ && (!node.hasAttribute("aria-hidden") || node.getAttribute("aria-hidden") != "true" || (node.className
&& node.className.indexOf && node.className.indexOf("fallback-image") !== -1));
},
/**
@@ -1776,6 +2023,9 @@ Readability.prototype = {
}
}
+ // Unwrap image from noscript
+ this._unwrapNoscriptImages(this._doc);
+
// Remove script tags from the document.
this._removeScripts(this._doc);
diff --git a/third-party/readability/readability.gresource.xml
b/third-party/readability/readability.gresource.xml
new file mode 100644
index 000000000..b3216fd5f
--- /dev/null
+++ b/third-party/readability/readability.gresource.xml
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<gresources>
+ <gresource prefix="/org/gnome/epiphany/readability">
+ <file compressed="true">Readability.js</file>
+ <file compressed="true">reader.css</file>
+ </gresource>
+</gresources>
diff --git a/src/resources/reader.css b/third-party/readability/reader.css
similarity index 100%
rename from src/resources/reader.css
rename to third-party/readability/reader.css
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
