searching for reviewer(smaug)
ca92748672f1: Bug 1530519 - Fix error handling in TransferableToIPCTransferable. r=smaug a=lizzard
Tom Schuster <evilpies@gmail.com> - Sat, 02 Mar 2019 12:42:56 +0000 - rev 516272
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1530519 - Fix error handling in TransferableToIPCTransferable. r=smaug a=lizzard Differential Revision: https://phabricator.services.mozilla.com/D21836
aab4fc980d3c: Bug 1524640 - Fix font loading during mozPrintCallback. r=smaug a=lizzard
Brendan Dahl <bdahl@mozilla.com> - Wed, 06 Feb 2019 23:46:58 +0000 - rev 516128
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1524640 - Fix font loading during mozPrintCallback. r=smaug a=lizzard Copy fonts loaded during a mozPrintCallback into the cloned document, so they are available during printing. Differential Revision: https://phabricator.services.mozilla.com/D18613
753ea9ad88e0: Bug 1524212 - Make nsContentUtils::DispatchInputEvent() update HTMLInputElement.validationMessage before dispatching "input" event r=smaug a=lizzard
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 08 Feb 2019 22:17:10 +0000 - rev 515914
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1524212 - Make nsContentUtils::DispatchInputEvent() update HTMLInputElement.validationMessage before dispatching "input" event r=smaug a=lizzard "input" event listener may want to check HTMLInputElement.validationMessage. However, due to moving "input" event dispatcher from HTMLInputElement::SetUserInput() to editor, HTMLInputElement::SetValueInternal() updates it **after** dispatching "input" event. This patch makes nsContentUtils::DispatchInputEvent() guarantees to update validationMessage value before dispatching every event. On the other hand, SetValueInternal() may be called without "input" event dispatchers. Therefore, it needs to keep updating validationMessage value in such cases. Differential Revision: https://phabricator.services.mozilla.com/D19126
59b0981f244c: Bug 1514940 - part 1: Forcibly disable new keyCode/charCode value of keypress events if the document is Confluence r=smaug,Ehsan,kmag a=lizzard
Masayuki Nakano <masayuki@d-toybox.com> - Tue, 05 Feb 2019 11:35:43 +0000 - rev 515840
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514940 - part 1: Forcibly disable new keyCode/charCode value of keypress events if the document is Confluence r=smaug,Ehsan,kmag a=lizzard Old Confluence does not aware of conflated model keypress event (see UI Events spec, https://w3c.github.io/uievents/#determine-keypress-keyCode). Additionally, Confluence can be hosted with any domains. Therefore, we cannot use blacklist to disable the conflated model keypress event only on it. This patch checks whether current or parent document is Confluence with JS module, called KeyPressEventModelCheckerChild. For kicking this module, nsHTMLDocument dispatches an custom event, CheckKeyPressEventModel, when it becomes editable only first time. Finally, if it's a Confluence instance, the module let PresShell know that we need to use split model keypress event in it. Differential Revision: https://phabricator.services.mozilla.com/D17907
5a11b523f8e3: Bug 1504157 - Make onsize event triggered once. r=smaug, a=test-only
Liang-Heng Chen <xeonchen@gmail.com> - Thu, 07 Feb 2019 20:34:06 +0000 - rev 515831
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1504157 - Make onsize event triggered once. r=smaug, a=test-only Avoid the condition been checked twice. Differential Revision: https://phabricator.services.mozilla.com/D18961
0d2e173a01fd: Bug 1521082 - Don't dispatch pointer event if the touch in touchstart event is already registered; r=smaug a=lizzard
Edgar Chen <echen@mozilla.com> - Tue, 29 Jan 2019 09:08:05 +0000 - rev 515779
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1521082 - Don't dispatch pointer event if the touch in touchstart event is already registered; r=smaug a=lizzard Differential Revision: https://phabricator.services.mozilla.com/D16976
c5a3a2a563c9: Bug 1524214 - Grab caching capturing content with local variable (for beta, release and esr60). r=smaug a=abillings
Masayuki Nakano <masayuki@d-toybox.com> - Mon, 04 Feb 2019 11:58:53 +0100 - rev 515692
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1524214 - Grab caching capturing content with local variable (for beta, release and esr60). r=smaug a=abillings
7f8594d21c51: Bug 1270059: Defer setTimeout/Intervals()s in content during pageload r=smaug
Randell Jesup <rjesup@jesup.org> - Sat, 26 Jan 2019 12:18:06 -0500 - rev 515567
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1270059: Defer setTimeout/Intervals()s in content during pageload r=smaug They will run off the new DeferredTimers idle queue if idle time is available
9ae7ae0acee4: Bug 1518252 - Block layout on Fluent. r=smaug
Zibi Braniecki <zbraniecki@mozilla.com> - Fri, 25 Jan 2019 23:14:03 +0000 - rev 515537
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1518252 - Block layout on Fluent. r=smaug Differential Revision: https://phabricator.services.mozilla.com/D17334
bf7da92f820b: Bug 1514007 - Annotate <script> evaluation profiler labels with async/defer. r=smaug
Markus Stange <mstange@themasta.com> - Fri, 25 Jan 2019 20:38:39 +0000 - rev 515515
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514007 - Annotate <script> evaluation profiler labels with async/defer. r=smaug This also moves the label from ScriptLoader::ProcessScriptElement to ScriptLoader::EvaluateScript so that it also kicks in for scripts that are run from NotifyOffThreadScriptLoadCompletedRunnable::Run. Differential Revision: https://phabricator.services.mozilla.com/D14924
ab592ce5d5ae: Bug 1514364 - Add appendRawFilter to nsIFilePicker to expose actual accept filters to GV for onFilePrompt. r=snorp, smaug
Dylan Roeh <droeh@mozilla.com> - Thu, 24 Jan 2019 10:21:05 -0600 - rev 515375
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514364 - Add appendRawFilter to nsIFilePicker to expose actual accept filters to GV for onFilePrompt. r=snorp, smaug
910e176da1f9: Bug 1514547 - Timing token to allow external protocol URLs are blocked in iframes without user-interaction, r=smaug
Andrea Marchesini <amarchesini@mozilla.com> - Thu, 24 Jan 2019 20:05:03 +0100 - rev 515368
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514547 - Timing token to allow external protocol URLs are blocked in iframes without user-interaction, r=smaug
eac3071c276b: Bug 1514547 - No error should be returned when external protocol URLs are blocked in iframes, r=smaug
Andrea Marchesini <amarchesini@mozilla.com> - Thu, 24 Jan 2019 20:05:03 +0100 - rev 515367
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514547 - No error should be returned when external protocol URLs are blocked in iframes, r=smaug
a68a2da104f3: Bug 1513014 - Defer reporting errors while preloading until the request is actually used r=smaug
Jon Coppeard <jcoppeard@mozilla.com> - Thu, 24 Jan 2019 14:57:30 +0000 - rev 515306
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1513014 - Defer reporting errors while preloading until the request is actually used r=smaug
24ad7d6e7775: Bug 1522181 - multiple external protocol URL blocker behind pref, r=smaug
Andrea Marchesini <amarchesini@mozilla.com> - Thu, 24 Jan 2019 06:31:41 +0100 - rev 515295
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1522181 - multiple external protocol URL blocker behind pref, r=smaug
b4dbd3d276f6: Bug 1521884 - Make test_input_time_key an xhtml file to get test coverage for this. r=smaug
Emilio Cobos Álvarez <emilio@crisal.io> - Wed, 23 Jan 2019 21:27:41 +0000 - rev 515270
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1521884 - Make test_input_time_key an xhtml file to get test coverage for this. r=smaug Also remove an assertion that doesn't really hold at all, and doesn't seem relevant to the function anyway. Differential Revision: https://phabricator.services.mozilla.com/D17359
b6f80451898e: Bug 1518252 - Block layout on Fluent. r=smaug
Zibi Braniecki <zbraniecki@mozilla.com> - Wed, 23 Jan 2019 21:29:15 +0000 - rev 515195
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1518252 - Block layout on Fluent. r=smaug Differential Revision: https://phabricator.services.mozilla.com/D17334
89875cd4d942: Bug 1520361 - Add Telemetry to know how people add site to the exception list. r=smaug,cpearce,janerik
Alastor Wu <alwu@mozilla.com> - Tue, 22 Jan 2019 18:22:44 +0000 - rev 514853
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1520361 - Add Telemetry to know how people add site to the exception list. r=smaug,cpearce,janerik Add Telemetry to know the number of times user adds site to allow list, remove site from allow list, add site to block list and remove site from block list. Differential Revision: https://phabricator.services.mozilla.com/D16642
4e227c741b4b: Bug 1519090 - Part 3: nsFocusManager::GetNextTabbableContentInScope should also check slot if aSkipOwner is false; r=smaug
Edgar Chen <echen@mozilla.com> - Tue, 22 Jan 2019 08:15:02 +0000 - rev 514809
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1519090 - Part 3: nsFocusManager::GetNextTabbableContentInScope should also check slot if aSkipOwner is false; r=smaug Slot element could be also focusable. Differential Revision: https://phabricator.services.mozilla.com/D16863
48f01f2f5573: Bug 1519090 - Part 2: Skip the node under slot scope in frame traversal; r=smaug
Edgar Chen <echen@mozilla.com> - Tue, 22 Jan 2019 08:11:56 +0000 - rev 514808
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1519090 - Part 2: Skip the node under slot scope in frame traversal; r=smaug The nodes under slot scope is already being traversed in GetNextTabbableContentIn*Scope. Differential Revision: https://phabricator.services.mozilla.com/D16860
12ffc4e92ff6: Bug 1519090 - Part 1: Some minor nsFocusManager cleanup; r=smaug
Edgar Chen <echen@mozilla.com> - Tue, 22 Jan 2019 07:41:36 +0000 - rev 514807
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1519090 - Part 1: Some minor nsFocusManager cleanup; r=smaug Differential Revision: https://phabricator.services.mozilla.com/D16848
81c98263341c: Bug 1521396 - Make ClickHandlerChild prevent multiple action of middle click on link element for preventing middle click paste r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Tue, 22 Jan 2019 07:28:55 +0000 - rev 514799
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1521396 - Make ClickHandlerChild prevent multiple action of middle click on link element for preventing middle click paste r=smaug When user middle clicks a link, most users must not expect to expose clipboard content to the web application. Therefore, we should stop firing paste event when user click a link with middle button. This patch makes ClickHandlerChild.handleEvent() prevent multiple action when it posts middle click event on a link. Note that even if middle click event is consumed, default event handler will dispatch paste event. Unfortunately, this is compatible behavior with the other browsers. Therefore, we cannot change this behavior with calling preventDefault() and this is the reason why this patch adds Event.preventMultipleActions(). Out of scope of this bug though, if there is an element which looks like a link but implemented with JS, web apps can steal clipboard content if user enables middle click event and user just wants to open the link in new tab. It might be better to stop dispatching paste event in any browsers and request to change each web apps. Differential Revision: https://phabricator.services.mozilla.com/D17209
0d9a334f6ae8: Bug 1489440 - Fix nsFormFillController in presence of Shadow DOM. r=smaug
Emilio Cobos Álvarez <emilio@crisal.io> - Mon, 21 Jan 2019 14:29:16 +0000 - rev 514757
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1489440 - Fix nsFormFillController in presence of Shadow DOM. r=smaug Differential Revision: https://phabricator.services.mozilla.com/D17121
d323b050b0ce: Bug 1518075 - Add another check for null script because compilation can 'succeed' if scripting is disabled r=smaug
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 21 Jan 2019 12:40:55 +0000 - rev 514710
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1518075 - Add another check for null script because compilation can 'succeed' if scripting is disabled r=smaug
cbac4cd15f5f: Bug 1521211 - Remove unused property nsContentUtils::sIsCustomElementsEnabled r=smaug
Timothy Guan-tin Chien <timdream@gmail.com> - Sat, 19 Jan 2019 00:11:26 +0000 - rev 514488
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1521211 - Remove unused property nsContentUtils::sIsCustomElementsEnabled r=smaug Differential Revision: https://phabricator.services.mozilla.com/D17036
4145884732ca: Bug 1517611 - Cycle collect WebAuthnManager and U2F more. r=smaug
Andrew McCreight <continuation@gmail.com> - Fri, 18 Jan 2019 23:21:46 +0000 - rev 514477
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1517611 - Cycle collect WebAuthnManager and U2F more. r=smaug Differential Revision: https://phabricator.services.mozilla.com/D17026
1d6b361c337a: Bug 1495363 - Abort the previous request, if a form is submitted twice - tests, r=smaug
Andrea Marchesini <amarchesini@mozilla.com> - Wed, 16 Jan 2019 22:51:17 +0100 - rev 514202
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1495363 - Abort the previous request, if a form is submitted twice - tests, r=smaug
675a97d54d12: Bug 1347070 - add qpSum to RTC outbound RTP stats r=jib,smaug
Nico Grunbaum <na-g@nostrum.com> - Tue, 15 Jan 2019 21:26:05 +0000 - rev 513974
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1347070 - add qpSum to RTC outbound RTP stats r=jib,smaug This adds qpSum to RTC outbound RTP stats Differential Revision: https://phabricator.services.mozilla.com/D16016
ac94b0e3484b: Bug 1518664 - Make special casting properties available to UA Widget r=smaug
Timothy Guan-tin Chien <timdream@gmail.com> - Tue, 15 Jan 2019 09:20:55 +0000 - rev 513936
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1518664 - Make special casting properties available to UA Widget r=smaug They are needed so the chrome script (CastingApp.js) can read back the state of the video element. Differential Revision: https://phabricator.services.mozilla.com/D16489
346851347c90: Bug 1514687 - Allow accessibility code to focus XUL radio buttons without selecting them. r=bgrins,smaug,MarcoZ,paolo
James Teh <jteh@mozilla.com> - Fri, 11 Jan 2019 04:52:50 +0000 - rev 513666
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514687 - Allow accessibility code to focus XUL radio buttons without selecting them. r=bgrins,smaug,MarcoZ,paolo 1. This requires exposing radiogroup's focusedItem property to C++. Unfortunately, there's no existing equivalent in nsIDOMXULSelectControlItemElement. radiogroup is the only element that needs this, so a new interface has been created for it. 2. Accessibility uses focusedItem instead of selectedItem when setting focus. 3. When an item is focused, accessibility needs to be notified. This is done using a DOMMenuItemActive event. Differential Revision: https://phabricator.services.mozilla.com/D15295
897a97a4931d: Bug 181137 - part 9: Move kNSURIs in Document.cpp and FragmentOrElement.cpp into their users r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:53:10 +0000 - rev 513425
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 9: Move kNSURIs in Document.cpp and FragmentOrElement.cpp into their users r=smaug Due to renaming nsContentIterator.cpp to ContentIterator.cpp, Document.cpp and FragmentOrElement.cpp are compiled in a unified cpp file now. However, both of them have same name constant, kNSURIs and some build systems claim that it in FragmentOrElement.cpp is never used. Fortunately, each of them is used only by one method. Therefore, this patch moves the each declaration into each user method. Differential Revision: https://phabricator.services.mozilla.com/D16186
dbf919528d97: Bug 181137 - part 8: Make ContentIteratorBase and its subclasses non-refcountable r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:52:26 +0000 - rev 513424
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 8: Make ContentIteratorBase and its subclasses non-refcountable r=smaug This patch makes ContentIteratorBase, PostContentIterator, PreContentIterator and ContentSubtreeIterator classes non-refcountable because most users can create their instances in stack and such users may be in a hot path. So, we can save a lot of cost of instantiation. Unfortunately, only ScriptableContentIterator creates one of the concrete classes and needs to destroy it properly. Therefore, its EnsureContentIterator(), destructor, traverse and unlink code becomes messy. However, ScriptableContentIterator was designed for automated tests and we need to maintain it not so many times. Therefore, improvement of other users must be worthwhiler than this demerit. Differential Revision: https://phabricator.services.mozilla.com/D15928
2a320da37ca0: Bug 181137 - part 7: Get rid of nsIContentIterator interface r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:51:42 +0000 - rev 513423
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 7: Get rid of nsIContentIterator interface r=smaug Now, nobody requires nsIContentIterator interface. So, we can get rid of it. Unfortunately, there is no macro to keep the inherited class, ContentSubtreeIterator, in the cycle collection to make it keep managing ContentSubtreeIterator::mRange without nsISupports interface. Therefore, this patch moves it into ContentIteratorBase temporarily. Anyway, the following patch makes those classes not refcountable. At that time, this issue will be fixed. Differential Revision: https://phabricator.services.mozilla.com/D15927
61f2c3b1b368: Bug 181137 - part 6: Make nsFilteredContentIterator not derived from nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:51:35 +0000 - rev 513422
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 6: Make nsFilteredContentIterator not derived from nsIContentIterator r=smaug nsFilteredContentIterator is used only by TextServicesDocument and there is no reason that it should be derived from nsIContentIterator except consistency. Additionally, it's now only class which is derived from nsIContentIterator except ContentIteratorBase. So, after this change, we can get rid of nsIContentIterator completely. This patch moves nsFilteredContentIterator into mozilla namespace and makes TextServicesDocument treat FilteredContentIterator directly instead of nsIContentIterator interface. Differential Revision: https://phabricator.services.mozilla.com/D15925
0d3c97c78d7c: Bug 181137 - part 5: Make all users of PostContentIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:50:19 +0000 - rev 513421
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 5: Make all users of PostContentIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of PostContentIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15923
7b087e0123bd: Bug 181137 - part 4: Make all users of PreContentIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:49:36 +0000 - rev 513420
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 4: Make all users of PreContentIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of PreContentIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15922
5fb4b3194d8e: Bug 181137 - part 3: Make all users of ContentSutreeIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:49:00 +0000 - rev 513419
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 3: Make all users of ContentSutreeIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of ContentSubtreeIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15920
a29d33c6fca1: Bug 181137 - part 2: Make nsContentIterator class is a base class of other concrete classes r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:48:11 +0000 - rev 513418
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 2: Make nsContentIterator class is a base class of other concrete classes r=smaug Currently, ContentIterator is created with a bool flag to decide whether the instance lists up post-order or pre-order. However, this is not clear. For example: nsCOMPtr<nsIContentIterator> preOrderIter = new ContentIterator(false); This is not clear whether this does right thing or not. This patch makes any users can create PostContentIterator for post-order iterator, and creates PreContentIterator for pre-order iterator. So, now, each creator needs to writhe above as: nsCOMPtr<nsIContentIterator> preOrderIter = new PreContentIterator(); or nsCOMPtr<nsIContentIterator> postOrderIter = new PostContentIterator(); Additionally, with this change, if each user starts to use concrete classes directly, compiler can stop using virtual calls because of all concrete classes are now marked as "final". Differential Revision: https://phabricator.services.mozilla.com/D15918
2eaf630d9925: Bug 181137 - part 1: Move nsContentIterator and nsContentSubtreeIterator into mozilla namespace r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Fri, 11 Jan 2019 01:47:15 +0000 - rev 513417
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 1: Move nsContentIterator and nsContentSubtreeIterator into mozilla namespace r=smaug First, we should move nsContentIterator and nsContentSubtreeIterator into mozilla namespace and then, remove "ns" prefix. Additionally, this patch separates the definition of the classes into ContentIterator.h and exposes it as "mozilla/ContentIterator.h". This allows everybody access those concrete classes. Differential Revision: https://phabricator.services.mozilla.com/D15917
58fad84cc31a: Bug 1515277 - Separate ClearData flag for Reporting data, r=jkt, r=smaug
Andrea Marchesini <amarchesini@mozilla.com> - Fri, 11 Jan 2019 09:43:58 +0100 - rev 513395
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1515277 - Separate ClearData flag for Reporting data, r=jkt, r=smaug
519e21226224: Bug 256180 parser part - Insert elements as siblings instead of children at the Blink-defined magic depth for compatibility. r=smaug.
Henri Sivonen <hsivonen@hsivonen.fi> - Tue, 12 Sep 2017 16:57:05 +0300 - rev 513390
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 256180 parser part - Insert elements as siblings instead of children at the Blink-defined magic depth for compatibility. r=smaug. MozReview-Commit-ID: K8fgv3rgklt
3bd1a8ffe4a9: Bug 1514975 - Synthesize mousemove event before contextmenu event caused by long tap r=smaug,kats
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 09:09:42 +0000 - rev 513206
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 1514975 - Synthesize mousemove event before contextmenu event caused by long tap r=smaug,kats Chrome synthesizes mousemove event and related events (mouseover, mouseenter, etc) when context menu event is fired by long tap. This allows users to open submenu which is opened by moving mouse cursor over a link. So, this fix improves accessibility of our users on some websites which are designed for desktop. Differential Revision: https://phabricator.services.mozilla.com/D14857
99a977d519a0: Bug 181137 - part 8: Make ContentIteratorBase and its subclasses non-refcountable r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:50:41 +0000 - rev 513204
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 8: Make ContentIteratorBase and its subclasses non-refcountable r=smaug This patch makes ContentIteratorBase, PostContentIterator, PreContentIterator and ContentSubtreeIterator classes non-refcountable because most users can create their instances in stack and such users may be in a hot path. So, we can save a lot of cost of instantiation. Unfortunately, only ScriptableContentIterator creates one of the concrete classes and needs to destroy it properly. Therefore, its EnsureContentIterator(), destructor, traverse and unlink code becomes messy. However, ScriptableContentIterator was designed for automated tests and we need to maintain it not so many times. Therefore, improvement of other users must be worthwhiler than this demerit. Differential Revision: https://phabricator.services.mozilla.com/D15928
65a4b245e851: Bug 181137 - part 7: Get rid of nsIContentIterator interface r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:47:54 +0000 - rev 513203
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 7: Get rid of nsIContentIterator interface r=smaug Now, nobody requires nsIContentIterator interface. So, we can get rid of it. Unfortunately, there is no macro to keep the inherited class, ContentSubtreeIterator, in the cycle collection to make it keep managing ContentSubtreeIterator::mRange without nsISupports interface. Therefore, this patch moves it into ContentIteratorBase temporarily. Anyway, the following patch makes those classes not refcountable. At that time, this issue will be fixed. Differential Revision: https://phabricator.services.mozilla.com/D15927
5385d5fd9b8b: Bug 181137 - part 6: Make nsFilteredContentIterator not derived from nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:47:11 +0000 - rev 513202
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 6: Make nsFilteredContentIterator not derived from nsIContentIterator r=smaug nsFilteredContentIterator is used only by TextServicesDocument and there is no reason that it should be derived from nsIContentIterator except consistency. Additionally, it's now only class which is derived from nsIContentIterator except ContentIteratorBase. So, after this change, we can get rid of nsIContentIterator completely. This patch moves nsFilteredContentIterator into mozilla namespace and makes TextServicesDocument treat FilteredContentIterator directly instead of nsIContentIterator interface. Differential Revision: https://phabricator.services.mozilla.com/D15925
83bec02c21d9: Bug 181137 - part 5: Make all users of PostContentIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:46:32 +0000 - rev 513201
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 5: Make all users of PostContentIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of PostContentIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15923
b7ab59bf545e: Bug 181137 - part 4: Make all users of PreContentIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:46:25 +0000 - rev 513200
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 4: Make all users of PreContentIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of PreContentIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15922
b6fc7a332db7: Bug 181137 - part 3: Make all users of ContentSutreeIterator treat it directly rather than via nsIContentIterator r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:45:07 +0000 - rev 513199
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 3: Make all users of ContentSutreeIterator treat it directly rather than via nsIContentIterator r=smaug Now, all users of ContentSubtreeIterator can access it directly. This patch makes them use the concrete class directly. Differential Revision: https://phabricator.services.mozilla.com/D15920
654fdbad67db: Bug 181137 - part 2: Make nsContentIterator class is a base class of other concrete classes r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:44:24 +0000 - rev 513198
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 2: Make nsContentIterator class is a base class of other concrete classes r=smaug Currently, ContentIterator is created with a bool flag to decide whether the instance lists up post-order or pre-order. However, this is not clear. For example: nsCOMPtr<nsIContentIterator> preOrderIter = new ContentIterator(false); This is not clear whether this does right thing or not. This patch makes any users can create PostContentIterator for post-order iterator, and creates PreContentIterator for pre-order iterator. So, now, each creator needs to writhe above as: nsCOMPtr<nsIContentIterator> preOrderIter = new PreContentIterator(); or nsCOMPtr<nsIContentIterator> postOrderIter = new PostContentIterator(); Additionally, with this change, if each user starts to use concrete classes directly, compiler can stop using virtual calls because of all concrete classes are now marked as "final". Differential Revision: https://phabricator.services.mozilla.com/D15918
90a1ff49b6b1: Bug 181137 - part 1: Move nsContentIterator and nsContentSubtreeIterator into mozilla namespace r=smaug
Masayuki Nakano <masayuki@d-toybox.com> - Thu, 10 Jan 2019 08:43:50 +0000 - rev 513197
Push 1953 by ffxbld-merge at Mon, 11 Mar 2019 12:10:20 +0000
Bug 181137 - part 1: Move nsContentIterator and nsContentSubtreeIterator into mozilla namespace r=smaug First, we should move nsContentIterator and nsContentSubtreeIterator into mozilla namespace and then, remove "ns" prefix. Additionally, this patch separates the definition of the classes into ContentIterator.h and exposes it as "mozilla/ContentIterator.h". This allows everybody access those concrete classes. Differential Revision: https://phabricator.services.mozilla.com/D15917