From 01d9236c8b98b9901cba1250a93c1be1288b26e4 Mon Sep 17 00:00:00 2001 From: rabuma abraham Date: Sun, 5 Oct 2025 16:14:34 +0200 Subject: [PATCH 1/2] Fix #986: Preserve Wikipedia

headings - Ensure

headings with meaningful content are not removed even if parent div contains edit/talk links - Updated _cleanHeaders, _cleanConditionally, and _prepArticle to handle this scenario - Added test case in test/wikipedia-h2.js to verify fix - Existing functionality and other tests remain unaffected Fixes #986 --- Readability.js | 88 +++++++++++++++++++++++++++++--- test/wikipedia-h2.js | 116 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 196 insertions(+), 8 deletions(-) create mode 100644 test/wikipedia-h2.js diff --git a/Readability.js b/Readability.js index 5cff4540..642a3c76 100644 --- a/Readability.js +++ b/Readability.js @@ -815,10 +815,29 @@ Readability.prototype = { this._forEachNode(articleContent.children, function (topCandidate) { this._cleanMatchedNodes(topCandidate, function (node, matchString) { - return ( + var shouldRemove = this.REGEXPS.shareElements.test(matchString) && - node.textContent.length < shareElementThreshold - ); + node.textContent.length < shareElementThreshold; + + // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, + // even if they match share elements criteria. This prevents Wikipedia h2 headings from + // being removed along with their parent divs. + if (shouldRemove && node.tagName === "DIV") { + var h2Elements = this._getAllNodesWithTag(node, ["h2"]); + for (var i = 0; i < h2Elements.length; i++) { + var h2Text = this._getInnerText(h2Elements[i], false).trim(); + if (h2Text.length) { + this.log( + "Preserving div with h2 heading despite share elements criteria:", + node, + h2Text + ); + return false; + } + } + } + + return shouldRemove; }); }); @@ -2513,6 +2532,23 @@ Readability.prototype = { var contentScore = 0; if (weight + contentScore < 0) { + // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, + // even if the div has negative class weight (e.g., contains "share" in class name). + // This prevents Wikipedia h2 headings from being removed along with their parent divs. + if (tag === "DIV") { + var h2Elements = this._getAllNodesWithTag(node, ["h2"]); + for (var i = 0; i < h2Elements.length; i++) { + var h2Text = this._getInnerText(h2Elements[i], false).trim(); + if (h2Text.length) { + this.log( + "Preserving div with h2 heading despite negative class weight:", + node, + h2Text + ); + return false; + } + } + } return true; } @@ -2540,18 +2576,18 @@ Readability.prototype = { "iframe", ]); - for (var i = 0; i < embeds.length; i++) { + for (var k = 0; k < embeds.length; k++) { // If this embed has attribute that matches video regex, don't delete it. - for (var j = 0; j < embeds[i].attributes.length; j++) { - if (this._allowedVideoRegex.test(embeds[i].attributes[j].value)) { + for (var j = 0; j < embeds[k].attributes.length; j++) { + if (this._allowedVideoRegex.test(embeds[k].attributes[j].value)) { return false; } } // For embed with tag, check inner HTML as well. if ( - embeds[i].tagName === "object" && - this._allowedVideoRegex.test(embeds[i].innerHTML) + embeds[k].tagName === "object" && + this._allowedVideoRegex.test(embeds[k].innerHTML) ) { return false; } @@ -2636,6 +2672,24 @@ Readability.prototype = { var haveToRemove = shouldRemoveNode(); + // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, + // even if they fail other cleaning criteria. This prevents Wikipedia h2 headings from + // being removed along with their parent divs. + if (haveToRemove && tag === "DIV") { + var h2Elements2 = this._getAllNodesWithTag(node, ["h2"]); + for (var m = 0; m < h2Elements2.length; m++) { + var h2Text2 = this._getInnerText(h2Elements2[m], false).trim(); + if (h2Text2.length) { + this.log( + "Preserving div with h2 heading despite cleaning criteria:", + node, + h2Text2 + ); + return false; + } + } + } + // Allow simple lists of images to remain in pages if (isList && haveToRemove) { for (var x = 0; x < node.children.length; x++) { @@ -2686,6 +2740,24 @@ Readability.prototype = { let headingNodes = this._getAllNodesWithTag(e, ["h1", "h2"]); this._removeNodes(headingNodes, function (node) { let shouldRemove = this._getClassWeight(node) < 0; + + // Fix for issue #986: Preserve h2 headings that contain meaningful content, + // even if their parent div contains extra links like "edit" or "talk". + // This prevents Wikipedia h2 headings from being incorrectly removed. + if (shouldRemove && node.tagName === "H2") { + // Check if the h2 has meaningful text content (more than just whitespace) + let headingText = this._getInnerText(node, false).trim(); + if (headingText.length) { + // Preserve h2 headings with actual content, regardless of parent container class weights + shouldRemove = false; + this.log( + "Preserving h2 heading with content despite negative class weight:", + node, + headingText + ); + } + } + if (shouldRemove) { this.log("Removing header with low class weight:", node); } diff --git a/test/wikipedia-h2.js b/test/wikipedia-h2.js new file mode 100644 index 00000000..37b5bc43 --- /dev/null +++ b/test/wikipedia-h2.js @@ -0,0 +1,116 @@ +/* eslint-env node, mocha */ + +/** + * Test case for issue #986: Wikipedia h2 headings being skipped when parent div contains extra links + * + * This test verifies that h2 headings are preserved even when their parent div contains + * links like "edit" or "talk" that might trigger negative class weight filtering. + */ + +const { JSDOM } = require("jsdom"); +const chai = require("chai"); +const expect = chai.expect; +const Readability = require("../index").Readability; + +// Sample HTML that mimics Wikipedia structure with h2 headings in divs containing edit/talk links +const testHTML = ` + + + + Test Wikipedia Article + + +
+

Main Article Title

+ +

This is the introduction paragraph.

+ + +
+

History

+ + [ + edit + ] + +
+ +

This is content under the History section.

+ + +
+

Development

+ + [ + edit + ] + +
+ +

This is content under the Development section.

+ + +
+

Important Section

+ talk + share +
+ +

This is content under the Important Section.

+ + +
+

+ talk +
+ +

This paragraph should remain.

+
+ + +`; + +describe("Wikipedia H2 Heading Preservation", function () { + it("should preserve h2 headings even when parent div contains edit/talk links", function () { + const dom = new JSDOM(testHTML, { + url: "https://example.com", + contentType: "text/html", + }); + + const reader = new Readability(dom.window.document, { + debug: false, + }); + + const article = reader.parse(); + + expect(article).to.not.be.null; + expect(article.content).to.not.be.undefined; + + // Check that h2 headings are preserved + const h2Count = (article.content.match(/

/g) || []).length; + + // We expect 4 h2 headings: "Main Article Title", "History", "Development", and "Important Section" + // The empty h2 should be removed + expect(h2Count).to.equal(4, `Expected 4 h2 headings, but found ${h2Count}`); + + // Check for specific headings + expect(article.content).to.include( + "

History

", + "History h2 heading not found" + ); + expect(article.content).to.include( + "

Development

", + "Development h2 heading not found" + ); + expect(article.content).to.include( + "

Important Section

", + "Important Section h2 heading not found" + ); + + // Check that empty h2 was removed + expect(article.content).to.not.include( + "

", + "Empty h2 heading should have been removed" + ); + }); +}); From 19d050ef54243a8a0ce7e31b3b575a5c1bd4ff8a Mon Sep 17 00:00:00 2001 From: rabuma abraham Date: Sat, 11 Oct 2025 16:11:27 +0200 Subject: [PATCH 2/2] Fix issue #986: Preserve Wikipedia h2 headings by removing mw-editsection spans Remove Wikipedia edit section spans (mw-editsection class) early in _prepDocument() to prevent them from adding negative class weight that would cause h2 headings to be incorrectly removed. This is a minimal, focused fix that only modifies _prepDocument() rather than adding exceptions in multiple cleaning functions. - Updated test cases: wikipedia, wikipedia-3, wikipedia-4, nytimes-1, nytimes-2 - All 1984 tests passing Fixes #986 --- Readability.js | 87 +++------------- test/test-pages/nytimes-1/expected.html | 3 +- test/test-pages/nytimes-2/expected.html | 6 +- test/test-pages/wikipedia-3/expected.html | 36 +++---- test/test-pages/wikipedia-4/expected.html | 4 +- test/test-pages/wikipedia/expected.html | 72 +++++++------- test/wikipedia-h2.js | 116 ---------------------- 7 files changed, 69 insertions(+), 255 deletions(-) delete mode 100644 test/wikipedia-h2.js diff --git a/Readability.js b/Readability.js index 642a3c76..3eab6e0e 100644 --- a/Readability.js +++ b/Readability.js @@ -677,6 +677,15 @@ Readability.prototype = { } this._replaceNodeTags(this._getAllNodesWithTag(doc, ["font"]), "SPAN"); + + // Fix for issue #986: Remove Wikipedia edit section links that cause headings + // to be improperly removed. These spans contain "edit" links and add negative + // weight to otherwise good headings. + this._forEachNode(this._getAllNodesWithTag(doc, ["span"]), function (span) { + if (span.className && span.className.includes("mw-editsection")) { + span.remove(); + } + }); }, /** @@ -815,29 +824,10 @@ Readability.prototype = { this._forEachNode(articleContent.children, function (topCandidate) { this._cleanMatchedNodes(topCandidate, function (node, matchString) { - var shouldRemove = + return ( this.REGEXPS.shareElements.test(matchString) && - node.textContent.length < shareElementThreshold; - - // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, - // even if they match share elements criteria. This prevents Wikipedia h2 headings from - // being removed along with their parent divs. - if (shouldRemove && node.tagName === "DIV") { - var h2Elements = this._getAllNodesWithTag(node, ["h2"]); - for (var i = 0; i < h2Elements.length; i++) { - var h2Text = this._getInnerText(h2Elements[i], false).trim(); - if (h2Text.length) { - this.log( - "Preserving div with h2 heading despite share elements criteria:", - node, - h2Text - ); - return false; - } - } - } - - return shouldRemove; + node.textContent.length < shareElementThreshold + ); }); }); @@ -2532,23 +2522,6 @@ Readability.prototype = { var contentScore = 0; if (weight + contentScore < 0) { - // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, - // even if the div has negative class weight (e.g., contains "share" in class name). - // This prevents Wikipedia h2 headings from being removed along with their parent divs. - if (tag === "DIV") { - var h2Elements = this._getAllNodesWithTag(node, ["h2"]); - for (var i = 0; i < h2Elements.length; i++) { - var h2Text = this._getInnerText(h2Elements[i], false).trim(); - if (h2Text.length) { - this.log( - "Preserving div with h2 heading despite negative class weight:", - node, - h2Text - ); - return false; - } - } - } return true; } @@ -2672,24 +2645,6 @@ Readability.prototype = { var haveToRemove = shouldRemoveNode(); - // Fix for issue #986: Don't remove divs that contain h2 headings with meaningful content, - // even if they fail other cleaning criteria. This prevents Wikipedia h2 headings from - // being removed along with their parent divs. - if (haveToRemove && tag === "DIV") { - var h2Elements2 = this._getAllNodesWithTag(node, ["h2"]); - for (var m = 0; m < h2Elements2.length; m++) { - var h2Text2 = this._getInnerText(h2Elements2[m], false).trim(); - if (h2Text2.length) { - this.log( - "Preserving div with h2 heading despite cleaning criteria:", - node, - h2Text2 - ); - return false; - } - } - } - // Allow simple lists of images to remain in pages if (isList && haveToRemove) { for (var x = 0; x < node.children.length; x++) { @@ -2740,24 +2695,6 @@ Readability.prototype = { let headingNodes = this._getAllNodesWithTag(e, ["h1", "h2"]); this._removeNodes(headingNodes, function (node) { let shouldRemove = this._getClassWeight(node) < 0; - - // Fix for issue #986: Preserve h2 headings that contain meaningful content, - // even if their parent div contains extra links like "edit" or "talk". - // This prevents Wikipedia h2 headings from being incorrectly removed. - if (shouldRemove && node.tagName === "H2") { - // Check if the h2 has meaningful text content (more than just whitespace) - let headingText = this._getInnerText(node, false).trim(); - if (headingText.length) { - // Preserve h2 headings with actual content, regardless of parent container class weights - shouldRemove = false; - this.log( - "Preserving h2 heading with content despite negative class weight:", - node, - headingText - ); - } - } - if (shouldRemove) { this.log("Removing header with low class weight:", node); } diff --git a/test/test-pages/nytimes-1/expected.html b/test/test-pages/nytimes-1/expected.html index 9213f63a..c35e3fe1 100644 --- a/test/test-pages/nytimes-1/expected.html +++ b/test/test-pages/nytimes-1/expected.html @@ -36,8 +36,7 @@

Obama administration officials said that they had briefed President-elect Donald J. Trump’s transition team, but that they did not know if Mr. Trump would stick with a policy of warmer relations with Sudan.

They said that Sudan had a long way to go in terms of respecting human rights, but that better relations could help increase American leverage.

Mr. Reeves said he thought that the American government was being manipulated and that the Obama administration had made a “deal with the devil.”

-

Continue reading the main story -

+

Continue reading the main story

diff --git a/test/test-pages/nytimes-2/expected.html b/test/test-pages/nytimes-2/expected.html index 040fa796..9796ba06 100644 --- a/test/test-pages/nytimes-2/expected.html +++ b/test/test-pages/nytimes-2/expected.html @@ -19,8 +19,7 @@

First, let’s say what the Yahoo sale is not. It is not a sale of the publicly traded company. Instead, it is a sale of the Yahoo subsidiary and some related assets to Verizon.

The sale is being done in two steps. The first step will be the transfer of any assets related to Yahoo business to a singular subsidiary. This includes the stock in the business subsidiaries that make up Yahoo that are not already in the single subsidiary, as well as the odd assets like benefit plan rights. This is what is being sold to Verizon. A license of Yahoo’s oldest patents is being held back in the so-called Excalibur portfolio. This will stay with Yahoo, as will Yahoo’s stakes in Alibaba Group and Yahoo Japan.

It is hard to overestimate how complex an asset sale like this is. Some of the assets are self-contained, but they must be gathered up and transferred. Employees need to be shuffled around and compensation arrangements redone. Many contracts, like the now-infamous one struck with the search engine Mozilla, which may result in a payment of up to a $1 billion, will contain change-of-control provisions that will be set off and have to be addressed. Tax issues always loom large.

-

Continue reading the main story -

+

Continue reading the main story

Continue reading the main story

@@ -42,8 +41,7 @@

Whether this is the most tax-efficient way is unclear to me as a nontax lawyer (email me if you know). Yahoo is likely to have a tax bill on the sale, possibly a substantial one. And I presume there were legal reasons for not using a Morris Trust structure, in which Yahoo would have been spun off and immediately sold to Verizon so that only Yahoo’s shareholders paid tax on the deal. In truth, the Yahoo assets being sold are only about 10 percent of the value of the company, so the time and logistics for such a sale when Yahoo is a melting ice cube may not have been worth it.

Finally, if another bidder still wants to acquire Yahoo, it has time. The agreement with Verizon allows Yahoo to terminate the deal and accept a superior offer by paying a $144 million breakup fee to Verizon. And if Yahoo shareholders change their minds and want to stick with Yahoo’s chief executive, Marissa Mayer, and vote down the deal, there is a so-called naked no-vote termination fee of $15 million payable to Verizon to reimburse expenses.

All in all, this was as hairy a deal as they come. There was the procedural and logistical complications of selling a company when the chief executive wanted to stay. Then there was the fact that this was an asset sale, including all of the challenges that go with it. Throw in all of the tax issues and the fact that this is a public company, and it is likely that the lawyers involved will have nightmares for years to come.

-

Continue reading the main story -

+

Continue reading the main story

diff --git a/test/test-pages/wikipedia-3/expected.html b/test/test-pages/wikipedia-3/expected.html index 6797e37e..4d9911c0 100644 --- a/test/test-pages/wikipedia-3/expected.html +++ b/test/test-pages/wikipedia-3/expected.html @@ -17,37 +17,35 @@

Hermitian matrices are named after Charles Hermite, who demonstrated in 1855 that matrices of this form share a property with real symmetric matrices of always having real eigenvalues. Other, equivalent notations in common use are , although note that in quantum mechanics, typically means the complex conjugate only, and not the conjugate transpose.

- Alternative characterizations[edit] + Alternative characterizations

Hermitian matrices can be characterized in a number of equivalent ways, some of which are listed below:

- Equality with the adjoint[edit] + Equality with the adjoint

A square matrix is Hermitian if and only if it is equal to its adjoint, that is, it satisfies

-

-

+

for any pair of vectors , where denotes the inner product operation.

This is also the way that the more general concept of self-adjoint operator is defined.

- Reality of quadratic forms[edit] + Reality of quadratic forms

A square matrix is Hermitian if and only if it is such that

-

-

+

- Spectral properties[edit] + Spectral properties

A square matrix is Hermitian if and only if it is unitarily diagonalizable with real eigenvalues.

- Applications[edit] + Applications

Hermitian matrices are fundamental to the quantum theory of matrix mechanics created by Werner Heisenberg, Max Born, and Pascual Jordan in 1925.

- Examples[edit] + Examples

In this section, the conjugate transpose of matrix is denoted as , the transpose of matrix is denoted as and conjugate of matrix is denoted as .

See the following example:

@@ -60,18 +58,16 @@

Well-known families of Pauli matrices, Gell-Mann matrices and their generalizations are Hermitian. In theoretical physics such Hermitian matrices are often multiplied by imaginary coefficients,[1][2] which results in skew-Hermitian matrices (see below).

Here, we offer another useful Hermitian matrix using an abstract example. If a square matrix equals the multiplication of a matrix and its conjugate transpose, that is, , then is a Hermitian positive semi-definite matrix. Furthermore, if is row full-rank, then is positive definite.

- Properties[edit] + Properties

@@ -194,7 +190,7 @@

(Alternatively, the determinant is the product of the matrix's eigenvalues, and as mentioned before, the eigenvalues of a Hermitian matrix are real.)

- Decomposition into Hermitian and skew-Hermitian[edit] + Decomposition into Hermitian and skew-Hermitian

Additional facts related to Hermitian matrices include: @@ -219,7 +215,7 @@

- Rayleigh quotient[edit] + Rayleigh quotient

In mathematics, for a given complex Hermitian matrix M and nonzero vector x, the Rayleigh quotient[4] , is defined as:[3]:p. 234[5]

@@ -233,7 +229,7 @@

The Rayleigh quotient is used in the min-max theorem to get exact values of all eigenvalues. It is also used in eigenvalue algorithms to obtain an eigenvalue approximation from an eigenvector approximation. Specifically, this is the basis for Rayleigh quotient iteration.

The range of the Rayleigh quotient (for matrix that is not necessarily Hermitian) is called a numerical range (or spectrum in functional analysis). When the matrix is Hermitian, the numerical range is equal to the spectral norm. Still in functional analysis, is known as the spectral radius. In the context of C*-algebras or algebraic quantum mechanics, the function that to M associates the Rayleigh quotient R(M, x) for a fixed x and M varying through the algebra would be referred to as "vector state" of the algebra.

- See also[edit] + See also

  • @@ -256,7 +252,7 @@

- References[edit] + References

    @@ -278,7 +274,7 @@

- External links[edit] + External links

  • diff --git a/test/test-pages/wikipedia-4/expected.html b/test/test-pages/wikipedia-4/expected.html index 89c51c2e..57cb2b01 100644 --- a/test/test-pages/wikipedia-4/expected.html +++ b/test/test-pages/wikipedia-4/expected.html @@ -518,13 +518,13 @@
-

[icon] -

+

[icon]

 -

This section needs expansion with: Proof of the properties requested. You can help by adding to it. (February 2018) -

+

This section needs expansion with: Proof of the properties requested. You can help by adding to it. (February 2018)
-

See also[edit]

+

See also

-

References[edit]

+

References

  1. ^ diff --git a/test/test-pages/wikipedia/expected.html b/test/test-pages/wikipedia/expected.html index 2b5eeb03..bb0f4672 100644 --- a/test/test-pages/wikipedia/expected.html +++ b/test/test-pages/wikipedia/expected.html @@ -106,7 +106,7 @@

    Contents

  2. 8 External links
-

History[edit] +

History

On January 23, 1998, Netscape made two announcements: first, that Netscape Communicator will be free; second, that the source code will also be free.[3] One day later, Jamie Zawinski from Netscape registered mozilla.org.[4] The project was named Mozilla after the original code name of the Netscape Navigator browser which is a blending of "Mosaic and Godzilla"[5] and used to co-ordinate the development of the Mozilla Application Suite, the open source version of Netscape's internet software, Netscape Communicator.[6][7] Jamie Zawinski says he came up with the name "Mozilla" at a Netscape staff meeting.[8][9] A small group of Netscape employees were tasked with coordination of the new community.

Originally, Mozilla aimed to be a technology provider for companies, such as Netscape, who would commercialize their open source code.[10] When AOL (Netscape's parent company) greatly reduced its involvement with Mozilla in July 2003, the Mozilla Foundation was designated the legal steward of the project.[11] Soon after, Mozilla deprecated the Mozilla Suite in favor of creating independent applications for each function, primarily the Firefox web browser and the Thunderbird email client, and moved to supply them directly to the public.[12]

@@ -114,17 +114,17 @@

History[[16]

At the end of 2013, Mozilla announced a deal with Cisco Systems whereby Firefox would download and use a Cisco-provided binary build of an open source[17] codec to play the proprietary H.264 video format.[18][19] As part of the deal, Cisco would pay any patent licensing fees associated with the binaries that it distributes. Mozilla's CTO, Brendan Eich, acknowledged that this is "not a complete solution" and isn't "perfect".[20] An employee in Mozilla's video formats team, writing in an unofficial capacity, justified[21] it by the need to maintain their large user base, which would be necessary in future battles for truly free video formats.

In December 2013, Mozilla announced funding for the development of non-free games[22] through its Game Creator Challenge. However, even those games that may be released under a non-free software or open source license must be made with open web technologies and Javascript as per the work criteria outlined in the announcement.

-

Eich CEO promotion controversy[edit] +

Eich CEO promotion controversy

On March 24, 2014, Mozilla promoted Brendan Eich to the role of CEO. This led to boycotts and protests from the LGBT community and its supporters, as Eich previously donated US$1,000[23] in 2008 in support of California's Proposition 8, a California ballot proposition and state constitutional amendment in opposition to same-sex marriage.[24] Eich's donation first became public knowledge in 2012, while he was Mozilla’s chief technical officer, leading to angry responses on Twitter—including the use of the hashtag "#wontworkwithbigots".[25]

Protests also emerged in 2014 following the announcement of Eich's appointment as CEO of Mozilla. U.S. companies OkCupid and CREDO Mobile received media coverage for their objections, with the former asking its users to boycott the browser,[26] while Credo amassed 50,000 signatures for a petition that called for Eich's resignation

Due to the controversy, Eich voluntarily stepped down on April 3, 2014[27] and Mitchell Baker, executive chairwoman of Mozilla Corporation, posted a statement on the Mozilla blog: "We didn’t move fast enough to engage with people once the controversy started. Mozilla believes both in equality and freedom of speech. Equality is necessary for meaningful speech. And you need free speech to fight for equality."[28] Eich's resignation promoted a larger backlash from conservatives who felt he had been forced out of the company internally.[citation needed]

OkCupid co-founder and CEO Sam Yagan had also donated $500[29] to Republican candidate Chris Cannon who proceeded to vote for multiple measures viewed as "anti-gay", including the banning of same-sex marriage.[30][31][32][33] Yagan claims he did not know about Cannon's stance on gay rights and that his contribution was due to the candidate being the ranking Republican participating in the House subcommittee that oversaw Internet and Intellectual Property matters.[34][35][36][37][38]

Reader comments on articles that were published close to the events were divided between support for OkCupid's actions and opposition to them. Supporters claimed the boycott was justified and saw OkCupid's actions as a firm statement of opposition to intolerance towards the gay community. Opponents saw OkCupid's actions as hypocritical, since Eich is also the inventor of JavaScript, which is still required to browse OkCupid's website, and felt that users should not be punished for the actions of Mozilla and suspected that OkCupid's actions were a publicity stunt.[36][39]

-

Values[edit] +

Values

According to Mozilla's manifesto,[40] which outlines goals, principles, and a pledge, "The Mozilla project uses a community-based approach to create world-class open source software and to develop new types of collaborative activities". Mozilla's manifesto mentions only its beliefs in regards to the Internet and Internet privacy, and has no mention of any political or social viewpoints.

-

Pledge[edit] +

Pledge

According to the Mozilla Foundation:[41]

@@ -137,7 +137,7 @@

Pledge[Software[edit] +

Software

@@ -145,25 +145,25 @@

Software[Firefox[edit] +

Firefox

Firefox is a web browser, and is Mozilla's flagship software product. It is available in both desktop and mobile versions. Firefox uses the Gecko layout engine to render web pages, which implements current and anticipated web standards.[42] As of late 2015, Firefox has approximately 10-11% of worldwide usage share of web browsers, making it the 4th most-used web browser.[43][44][45]

Firefox began as an experimental branch of the Mozilla codebase by Dave Hyatt, Joe Hewitt and Blake Ross. They believed the commercial requirements of Netscape's sponsorship and developer-driven feature creep compromised the utility of the Mozilla browser.[46] To combat what they saw as the Mozilla Suite's software bloat, they created a stand-alone browser, with which they intended to replace the Mozilla Suite.

Firefox was originally named Phoenix but the name was changed so as to avoid trademark conflicts with Phoenix Technologies. The initially-announced replacement, Firebird, provoked objections from the Firebird project community.[47][48] The current name, Firefox, was chosen on February 9, 2004.[49]

-

Firefox Mobile[edit] +

Firefox Mobile

Firefox Mobile (codenamed Fennec) is the build of the Mozilla Firefox web browser for devices such as smartphones and tablet computers.

Firefox Mobile uses the same Gecko layout engine as Mozilla Firefox. For example, version 1.0 used the same engine as Firefox 3.6, and the following release, 4.0, shared core code with Firefox 4.0. Its features include HTML5 support, Firefox Sync, add-ons support and tabbed browsing.[50]

Firefox Mobile is currently available for Android 2.2 and above devices with an ARMv7 or ARMv6 CPU.[51] The x86 architecture is not officially supported.[52] Tristan Nitot, president of Mozilla Europe, has said that it's unlikely that an iPhone or a BlackBerry version will be released, citing Apple's iTunes Store application approval policies (which forbid applications competing with Apple's own, and forbid engines which run downloaded code) and BlackBerry's limited operating system as the reasons.[53]

-

Firefox OS[edit] +

Firefox OS

Firefox OS (project name: Boot to Gecko also known as B2G) is an open source operating system in development by Mozilla that aims to support HTML5 apps written using "open Web" technologies rather than platform-specific native APIs. The concept behind Firefox OS is that all user-accessible software will be HTML5 applications, that use Open Web APIs to access the phone's hardware directly via JavaScript.[54]

Some devices using this OS include[55] Alcatel One Touch Fire, ZTE Open, LG Fireweb.

-

Thunderbird[edit] +

Thunderbird

Thunderbird is a free, open source, cross-platform email and news client developed by the volunteers of the Mozilla Community.

On July 16, 2012, Mitchell Baker announced that Mozilla's leadership had come to the conclusion that on-going stability was the most important thing for Thunderbird and that innovation in Thunderbird was no longer a priority for Mozilla. In that update Baker also suggested that Mozilla had provided a pathway for community to innovate around Thunderbird if the community chooses.[56]

-

SeaMonkey[edit] +

SeaMonkey

@@ -173,7 +173,7 @@

SeaMonkey[SeaMonkey (formerly the Mozilla Application Suite) is a free and open source cross platform suite of Internet software components including a web browser component, a client for sending and receiving email and USENET newsgroup messages, an HTML editor (Mozilla Composer) and the ChatZilla IRC client.

On March 10, 2005, the Mozilla Foundation announced that it would not release any official versions of Mozilla Application Suite beyond 1.7.x, since it had now focused on the standalone applications Firefox and Thunderbird.[57] SeaMonkey is now maintained by the SeaMonkey Council, which has trademarked the SeaMonkey name with help from the Mozilla Foundation.[58] The Mozilla Foundation provides project hosting for the SeaMonkey developers.

-

Bugzilla[edit] +

Bugzilla

@@ -182,57 +182,57 @@

Bugzilla[Bugzilla is a web-based general-purpose bug tracking system, which was released as open source software by Netscape Communications in 1998 along with the rest of the Mozilla codebase, and is currently stewarded by Mozilla. It has been adopted by a variety of organizations for use as a bug tracking system for both free and open source software and proprietary projects and products, including the Mozilla Foundation, the Linux kernel, GNOME, KDE, Red Hat, Novell, Eclipse and LibreOffice.[59]

-

Components[edit] +

Components

-

NSS[edit] +

NSS

Network Security Services (NSS) comprises a set of libraries designed to support cross-platform development of security-enabled client and server applications. NSS provides a complete open-source implementation of crypto libraries supporting SSL and S/MIME. NSS was previously tri-licensed under the Mozilla Public License 1.1, the GNU General Public License, and the GNU Lesser General Public License, but upgraded to GPL-compatible MPL 2.0.

AOL, Red Hat, Sun Microsystems/Oracle Corporation, Google and other companies and individual contributors have co-developed NSS and it is used in a wide range of non-Mozilla products including Evolution, Pidgin, and Apache OpenOffice.

-

SpiderMonkey[edit] +

SpiderMonkey

SpiderMonkey is the original JavaScript engine developed by Brendan Eich when he invented JavaScript in 1995 as a developer at Netscape. It became part of the Mozilla product family when Mozilla inherited Netscape's code-base in 1998. In 2011, Eich transferred the nominal ownership of the SpiderMonkey code and project to Dave Mandelin.[60]

SpiderMonkey is a cross-platform engine written in C++ which implements ECMAScript, a standard developed from JavaScript.[60][61] It comprises an interpreter, several just-in-time compilers, a decompiler and a garbage collector. Products which embed SpiderMonkey include Firefox, Thunderbird, SeaMonkey, and many non-Mozilla applications.[62]

-

Rhino[edit] +

Rhino

Rhino is an open source JavaScript engine managed by the Mozilla Foundation. It is developed entirely in Java. Rhino converts JavaScript scripts into Java classes. Rhino works in both compiled and interpreted mode.[63]

-

Gecko[edit] +

Gecko

Gecko is a layout engine that supports web pages written using HTML, SVG, and MathML. Gecko is written in C++ and uses NSPR for platform independence. Its source code is licensed under the Mozilla Public License.

Firefox uses Gecko both for rendering web pages and for rendering its user interface. Gecko is also used by Thunderbird, SeaMonkey, and many non-Mozilla applications.

-

Rust[edit] +

Rust

Rust is a compiled programming language being developed by Mozilla Research. It is designed for safety, concurrency, and performance. Rust is intended for creating large and complex software which needs to be both safe against exploits and fast.

Rust is being used in an experimental layout engine, Servo, which is developed by Mozilla and Samsung. Servo is not used in any consumer-oriented browsers yet. However, the Servo project developers plan for parts of the Servo source code to be merged into Gecko, and Firefox, incrementally.[64][65]

-

XULRunner[edit] +

XULRunner

XULRunner is a software platform and technology experiment by Mozilla, that allows applications built with the same technologies used by Firefox extensions (XPCOM, Javascript, HTML, CSS, XUL) to be run natively as desktop applications, without requiring Firefox to be installed on the user's machine. XULRunner binaries are available for the Windows, GNU/Linux and OS X operating systems, allowing such applications to be effectively cross platform.

-

pdf.js[edit] +

pdf.js

Pdf.js is a library developed by Mozilla that allows in-browser rendering of pdf documents using the HTML5 Canvas and Javascript. It is included by default in recent versions of Firefox, allowing the browser to render pdf documents without requiring an external plugin; and it is available separately as an extension named "PDF Viewer" for Firefox for Android, SeaMonkey, and the Firefox versions which don't include it built-in. It can also be included as part of a website's scripts, to allow pdf rendering for any browser that implements the required HTML5 features and can run Javascript.

-

Shumway[edit] +

Shumway

Shumway is an open source replacement for the Adobe Flash Player, developed by Mozilla since 2012, using open web technologies as a replacement for Flash technologies. It uses Javascript and HTML5 Canvas elements to render Flash and execute Actionscript. It is included by default in Firefox Nightly and can be installed as an extension for any recent version of Firefox. The current implementation is limited in its capabilities to render Flash content outside simple projects.

-

Other activities[edit] +

Other activities

-

Mozilla VR[edit] +

Mozilla VR

Mozilla VR is a team focused on bringing Virtual reality tools, specifications, and standards to the open Web.[66] Mozilla VR maintains A-Frame (VR), a web framework for building VR experiences, and works on advancing WebVR support within web browsers.

-

Mozilla Persona[edit] +

Mozilla Persona

Mozilla Persona is a secure, cross-browser website authentication mechanism which allows a user to use a single username and password (or other authentication method) to log in to multiple sites.[67] Mozilla Persona will be shutting down on November 30, 2016.[68]

-

Mozilla Location Service[edit] +

Mozilla Location Service

This open source crowdsourced geolocation service was started by Mozilla in 2013 and offers a free API.

-

Webmaker[edit] +

Webmaker

Mozilla Webmaker is Mozilla's educational initiative, Webmaker's goal is to "help millions of people move from using the web to making the web." As part of Mozilla’s non-profit mission, Webmaker aims "to help the world increase their understanding of the web, take greater control of their online lives, and create a more web literate planet."[69][70][70]

-

Mozilla Developer Network[edit] +

Mozilla Developer Network

Mozilla maintains a comprehensive developer documentation website called the Mozilla Developer Network which contains information about web technologies including HTML, CSS, SVG, JavaScript, as well Mozilla-specific information. In addition, Mozilla publishes a large number of videos about web technologies and the development of Mozilla projects on the Air Mozilla website.[71][72]

-

[edit] +

The Mozilla Community consists of over 40,000 active contributors from across the globe[citation needed]. It includes both paid employees and volunteers who work towards the goals set forth[40] in the Mozilla Manifesto. Many of the sub-communities in Mozilla have formed around localization efforts for Mozilla Firefox, and the Mozilla web properties.

-

Local communities[edit] +

Local communities

@@ -241,7 +241,7 @@

Local communities[

There are a number of sub-communities that exist based on their geographical locations, where contributors near each other work together on particular activities, such as localization, marketing, PR and user support.

-

Mozilla Reps[edit] +

Mozilla Reps

@@ -260,9 +260,9 @@

Mozilla Reps[Conferences and events[edit] +

Conferences and events

-

Mozilla Festival[edit] +

Mozilla Festival

@@ -272,13 +272,13 @@

Mozilla Festival[

The Mozilla Festival is an annual event where hundreds of passionate people explore the Web, learn together and make things that can change the world. With the emphasis on making—the mantra of the Festival is "less yack, more hack." Journalists, coders, filmmakers, designers, educators, gamers, makers, youth and anyone else, from all over the world, are encouraged to attend, with attendees from more than 40 countries, working together at the intersection between freedom, the Web, and that years theme.

The event revolves around design challenges which address key issues based on the chosen theme for that years festival. In previous years the Mozilla Festival has focused on Learning, and Media, with the 2012 festival being based around making. The titles of the festival revolve around the main theme, freedom (as in freedom of speech not free beer), and the Web.

-

MozCamps[edit] +

MozCamps

MozCamps are the critical part of the Grow Mozilla initiative which aims to grow the Mozilla Community. These camps aim to bring core contributors from around the world together. They are intensive multi-day summits that include keynote speeches by Mozilla leadership, workshops and breakout sessions (led by paid and unpaid staff), and fun social outings. All of these activities combine to reward contributors for their hard work, engage them with new products and initiatives, and align all attendees on Mozilla's mission.

-

Mozilla Summit[edit] +

Mozilla Summit

Mozilla Summit are the global event with active contributors and Mozilla employees to develop a shared understanding of Mozilla's mission together. Over 2,000 people representing 90 countries and 114 languages gathered in Santa Clara, Toronto and Brussels in 2013. Mozilla has since its last summit in 2013 replaced summits with all-hands where both employees and volunteers come together to collaborate the event is a scaled down version of Mozilla Summit.

-

See also[edit] +

See also

-

References[edit] +

References

    @@ -504,7 +504,7 @@

    References[Constant downloads failure in firefox

    -

    External links[edit] +

    External links

    diff --git a/test/wikipedia-h2.js b/test/wikipedia-h2.js deleted file mode 100644 index 37b5bc43..00000000 --- a/test/wikipedia-h2.js +++ /dev/null @@ -1,116 +0,0 @@ -/* eslint-env node, mocha */ - -/** - * Test case for issue #986: Wikipedia h2 headings being skipped when parent div contains extra links - * - * This test verifies that h2 headings are preserved even when their parent div contains - * links like "edit" or "talk" that might trigger negative class weight filtering. - */ - -const { JSDOM } = require("jsdom"); -const chai = require("chai"); -const expect = chai.expect; -const Readability = require("../index").Readability; - -// Sample HTML that mimics Wikipedia structure with h2 headings in divs containing edit/talk links -const testHTML = ` - - - - Test Wikipedia Article - - -
    -

    Main Article Title

    - -

    This is the introduction paragraph.

    - - -
    -

    History

    - - [ - edit - ] - -
    - -

    This is content under the History section.

    - - -
    -

    Development

    - - [ - edit - ] - -
    - -

    This is content under the Development section.

    - - -
    -

    Important Section

    - talk - share -
    - -

    This is content under the Important Section.

    - - -
    -

    - talk -
    - -

    This paragraph should remain.

    -
    - - -`; - -describe("Wikipedia H2 Heading Preservation", function () { - it("should preserve h2 headings even when parent div contains edit/talk links", function () { - const dom = new JSDOM(testHTML, { - url: "https://example.com", - contentType: "text/html", - }); - - const reader = new Readability(dom.window.document, { - debug: false, - }); - - const article = reader.parse(); - - expect(article).to.not.be.null; - expect(article.content).to.not.be.undefined; - - // Check that h2 headings are preserved - const h2Count = (article.content.match(/

    /g) || []).length; - - // We expect 4 h2 headings: "Main Article Title", "History", "Development", and "Important Section" - // The empty h2 should be removed - expect(h2Count).to.equal(4, `Expected 4 h2 headings, but found ${h2Count}`); - - // Check for specific headings - expect(article.content).to.include( - "

    History

    ", - "History h2 heading not found" - ); - expect(article.content).to.include( - "

    Development

    ", - "Development h2 heading not found" - ); - expect(article.content).to.include( - "

    Important Section

    ", - "Important Section h2 heading not found" - ); - - // Check that empty h2 was removed - expect(article.content).to.not.include( - "

    ", - "Empty h2 heading should have been removed" - ); - }); -});