May 2018 - webkit-dev - lists.webkit.org

PSA: Bit fields won't be packed on Windows if you mix types
by Ryosuke Niwa 04 Sep '20

04 Sep '20

Hi, Unlike gcc and clang, MSVC pads each consecutive member variables of the same type in bitfields. e.g. if you have: struct AB { unsigned m_1 : 31; bool m_2 : 1; } then *MSVC pads m_1 and allocates sizeof(unsigned) * 2 for AB* whereas gcc and clang only allocate sizeof(unsigned) * 1 for AB. This is *not a compiler bug*. It's a spec. compliant behavior, and may in fact have a better run-time performance in some situations. However, for core objects like RenderObject and InlineBox, allocating extra 4-8 bytes per each object is prohibitory expensive. In such cases, please *use the same POD type for all bitfield member variables*. (Storage class classifiers and variable qualifiers seem to have no effect on how variables are packed; e.g. mutable, const, etc...). For example, MSVC will allocate sizeof(unsigned) * 1 for AB if we rewrite the above code as: struct AB { unsigned m_1 : 31; unsigned m_2 : 1; } When you're making this change, *be sure to audit all code that assigns a non-boolean value to m_2* because implicit type coercion into boolean is no longer in effect. For example, AB ab; ab.m_2 = 2; puts(ab.m_2 ? "true" : "false"); will print "true" before the change and will print "false" after the change. An easy way to ensure you've audited all such code is to add getters and setters for all bitfield member variables or wrap them in a special structure or class as done in http://trac.webkit.org/changeset/103353 and http://trac.webkit.org/changeset/104254. Best regards, Ryosuke Niwa Software Engineer Google Inc.

4 8

Support "registerProtocolHandler" in WebKit2
by Gyuyoung Kim 20 Apr '20

20 Apr '20

Hello, I would like to listen what do you think to support 'registerProtocolHandler' in WebKit2. This feature is to execute web content through registered custom protocol. - For example, web content can register "mailto" custom protocol using this feature, <script> navigator.registerProtocolHandler("mailto", "https://mail.naver.com/", "Web Mail"); </script> <html> <head> <title>Web Protocol Handler Sample - Test</title> </head> <body> <p>Send an email : <a href="mailto://">this !</a></p> </body> </html> Besides this feature has been supported by Mozilla and Chromium (From Mozilla 3.0, Chromium 13). - https://developer.mozilla.org/en-US/docs/Web/API/Navigator/registerProtocol… - https://developer.mozilla.org/en/docs/Web-based_protocol_handlers The feature is included in the W3C recommendation 28 released on Oct. 2014. - W3C spec : http://www.w3.org/TR/html5/webappapis.html#custom-handlers IIRC, some WebKit1 ports supported this feature though, those ports were deprecated. Now WebKit port supports this feature no more. There is a very old bug to support this feature though, it wasn't maintained so far. Recently I updated it based on latest WebKit. - https://bugs.webkit.org/show_bug.cgi?id=92749 Feel free to give me any feedback about this feature. Gyuyoung.

10 26

WebKit project in Coverity
by Carlos Alberto Lopez Perez 09 Jan '20

09 Jan '20

Hi, Coverity is an static analysis tool that allows to find bugs and vulnerabilities on the source code via static analysis. For open source projects, they offer free usage of their platform. The WebKit project is already registered there since a while. [1] To read the reports in detail or run new scans you have to be member of the WebKit project in Coverity. I happen to be one of the admins there, and I will happily grant you access to this platform if you are a WebKit committer (listed in contributors.json). So if you are interested in this, just send me an email requesting access. Regards ------- [1] https://scan.coverity.com/projects/webkit

2 1

C++17 is here. Should we use it?
by JF Bastien 26 Jun '19

26 Jun '19

Hello WebKilters, Our Chrome-y friends are considering the use of C++14 <https://groups.google.com/a/chromium.org/forum/?utm_medium=email&utm_source…>. I have to say that C++14 in WebKit has been quite amazing, and we should consider using C++17: it has many wonderful new things <https://github.com/tvaneerd/cpp17_in_TTs/blob/master/ALL_IN_ONE.md>, some of which we already use through WTF’s re-implementation of library features. By now (table as witness <http://en.cppreference.com/w/cpp/compiler_support>) most C++17 languages features are in clang and GCC, and MSVC isn’t doing too bad either. Language things can just come through WTF if we really want them. So how about it? JF

8 23

For your consideration: Naming scheme for fooIfExists/ensureFoo
by Darin Adler 22 Feb '19

22 Feb '19

Lets bike shed! For some time, functions with names like fooIfExists and ensureFoo have been bothering me. I find both names kind of opaque and unpleasant. Here’s an example: StyleResolver* styleResolverIfExists(); StyleResolver* ensureStyleResolver() What do you think of these names instead? StyleResolver* optionalStyleResolver(); StyleResolver& requiredStyleResolver(); I like them better. Note also that I think the requiredStyleResolver function should return a reference so nobody is tempted to do null checks. It seems like the C++ community likes the name optional for this concept; isn’t there some kind of std::optional template? -- Darin

15 25

CSSOM View behavior for scrollLeft, scrollTop, scrollWidth etc
by Frédéric Wang 24 Sep '18

24 Sep '18

Hi everybody, I'd like to announce that I started to implement the behavior for scrollLeft, scrollTop, scrollWidth, scrollHeight, scrollTo, scrollBy and scrollingElement [1] as specified by the CSSOM View specification [2]. To summarize, the main difference is which element to use in order to set or get the scroll properties of the viewport in standard (non-Quirks) mode: * The CSSOM View spec says it should be document.documentElement and most browsers implement it that way. * Chromium used to use document.body but this has recently been changed to follow the CSSOM View spec [3]. * WebKit still uses document.body and I'm proposing to change this behavior. Note that this is likely to break existing content. The change was attempted in WebKit in the past but broke some non-regression tests [4]. Chromium people also had some issues when trying to ship the change but things went well at the end [5]. For now I'm developing it under a CSSOMViewScrollingAPI developer flag. I was able to run the tests with that flag enabled, modulo a few test adjustments. [1] https://bugs.webkit.org/show_bug.cgi?id=5991 [2] https://drafts.csswg.org/cssom-view [3] https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/X64Sg16RhT4 [4] https://bugs.webkit.org/show_bug.cgi?id=106133, https://bugs.webkit.org/show_bug.cgi?id=121876 [5] https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/2iOz5-fgD8Y/… ; https://bugs.webkit.org/show_bug.cgi?id=5991#c20 -- Frédéric Wang - frederic-wang.fr

1 3

Standard process for exporting new WPT tests?
by Frédéric Wang 31 May '18

31 May '18

Hi all, When implementing new platform features in WebKit, some of us write new WPT tests in order to verify spec conformance. One of the reason to use this format is that the tests can then be easily shared with web engine developers and people involved in standardization, an important feature for web platform developers. Youenn has shared various ideas to export such WPT tests to the W3C GitHub repository in the past [1] [2] [3] but I'm not sure the WebKit community has reached an agreement. There are basically two ways to proceed: 1) Either submit WPT tests to GitHub (or Mozilla / Chromium) repository, get them reviewed and do the import within the patch that implements the feature in WebKit. 2) Or submit the patch with new WPT tests to WebKit, get it reviewed and export it to GitHub (later sync with Mozilla / Chromium). The first option has the advantage to get more people involved in test review, which sometimes gives faster replies or useful advice from people who more familiar with the spec. However, the second option has the advantage of not splitting the coding & testing parts and is better for the WebKit implementers and reviewers. That latter option also aligns with what Mozilla and Chromium developers do. Can we please agree on a way to proceed? And add documentation somewhere for developers/reviewers? Thanks! [1] https://lists.webkit.org/pipermail/webkit-dev/2017-March/028884.html [2] https://lists.webkit.org/pipermail/webkit-dev/2017-April/028932.html [3] https://lists.webkit.org/pipermail/webkit-dev/2017-December/029837.html -- Frédéric Wang - frederic-wang.fr

6 17

why the "WebKitStorageProcess”　process missing when midori running?
by tugouxp 28 May '18

28 May '18

Hi folks: i porting the webkit to my boards, and also the browser midori and minibrowser can be running normally, but from the pstree , i know that the process "WebKitStorageProcess“　are not running, so can you tell me why? so you can see only WebKitNetWork, WebKitWebProcess and midori process here, where is the storage process? init-+-WebKitNetworkPr-+-2*[{ReceiveQueue}] | |-{Storage} | |-{WebKitNetworkPr} | |-{background} | |-{erialBackground} | |-{gmain} | `-{pool} |-WebKitWebProces-+-{EventDispatcher} | |-2*[{ReceiveQueue}] | |-{WebKitWebProces} | |-{gmain} | |-{nnectionManager} | `-{rruptDispatcher} |-adbd-+-adb_shell---sh---pstree | `-4*[{adbd}] |-ash-+-midori-+-{PressureMonitor} | | |-2*[{ReceiveQueue}] | | |-{StorageManager} | | |-{e: IconDatabase} | | |-{ebsiteDataStore} | | |-{gmain} | | |-{midori} | | `-{pool} | `-weston-+-weston-desktop- | `-3*[{weston}] |-ntpd |-udevd |-udhcpc `-wpa_supplicant

2 1

Proposal: Removing ENABLE(INTL)
by Yusuke SUZUKI 24 May '18

24 May '18

Hi WebKittens, I would like to remove ENABLE(INTL) compile time flag and always enable INTL. This is ECMAScript Intl (i18n) feature. This feature depends on ICU. But right now, even JSC (and WTF) always require ICU. Even if we disable ENABLE(INTL), we still require ICU. New INTL features are gurded separately with the compile flags (since some new features require new ICU). So, removing ENABLE(INTL) just enables Intl basic features, which are shipped more than 1 year ago and it is quite mature. Best regards, Yusuke Suzuki -- Regards, Yusuke Suzuki

4 6

Switch to std::variant?
by zan＠falconsigh.net 23 May '18

23 May '18

Hi, with C++17 adoption progressing, are there existing plans to switch over from WTF::Variant to stdlib-provided std::variant implementations? Some GCC-and-libstdc++ configurations would still require a fallback implementation, just like with std::optional. Bug #175583 started work on importing a C++17-compliant implementation, but currently seems stalled. https://bugs.webkit.org/show_bug.cgi?id=175583 Are there any other reservations or obstacles about adopting std::variant? Cheers, Zan

3 3