Wrong document URL after parsing some HTML dynamically

Поделиться
HTML-код
  • Опубликовано: 19 окт 2024
  • Continuing the climb up mount WPT... Today we look at some failures under /domparsing where Flow is doing better than we are!
    PR: github.com/Lad...
    WPT dashboard: wpt.fyi/result...
    Grafana: linegoup.lol/
    Ladybird: ladybird.org/
    GitHub: github.com/Lad...
    Discord: / discord
    I'm using the CLion IDE from JetBrains: jb.gg/clion_ak
    My coding font is Berkeley Mono: berkeleygraphi...

Комментарии • 43

  • @pavelshlyak8385
    @pavelshlyak8385 3 дня назад +69

    Could you make some code review videos so we understand the workflow, common mistakes etc?

  • @metroid031993
    @metroid031993 2 дня назад +54

    I think when you made the test, you forgot to pass the mimetype to parseFromString! Also, when checking for correctness, could you not verify that the URLs are the same? (Not super familiar with JS, but I'm pretty sure the original WPT test printed out the correct URL in the test results, which was working correctly)

    • @awesomekling
      @awesomekling  2 дня назад +25

      Yeah we should totally check that the urls are the same! I’ll make that edit when I’m back at the computer, great suggestion :)

  • @avramitra
    @avramitra 2 дня назад +10

    These videos are extremely helpful and motivating for anyone interested to contribute to LB.
    I am seriously considering to *try* to contribute, and it's mainly because of your videos.
    Only 3 blockers are there:
    1. My hectic schedule
    2. My lack of advanced cpp knowledge. I'm only proficient in c.
    3. LB's huge and sophisticated code base.
    I hope I can start soon. This is a project which I deeply believe in, and want to be successful.

    • @Bobbias
      @Bobbias 2 дня назад +1

      Diving into this codebase would be a good way to get a feel for modern C++ in a fairly high quality codebase. Sparse documentation, except for the web standard docs they use though.

  • @DanielMircea
    @DanielMircea 2 дня назад +13

    I compiled lb, the other day. I tried a few websites, and just realized that I may have been their first Ladybird visitor ever. For whatever reason it felt nice 😅

  • @TheMrKeksLp
    @TheMrKeksLp 2 дня назад +9

    I really like these browser fixing videos

  • @fredleckie5880
    @fredleckie5880 3 дня назад +21

    Thank you, sir

  • @zaitsev5990
    @zaitsev5990 2 дня назад +1

    These videos are so addictive, thank you so much Andreas!

  • @UraYukimitsu
    @UraYukimitsu 2 дня назад +19

    I'm not smart enough to do half of this but in the test, you did not pass the mime type to parseFromString

    • @awesomekling
      @awesomekling  2 дня назад +11

      Oh jeez you are absolutely right, thanks for pointing that out! Updated the PR 😊

  • @chairclr
    @chairclr 2 дня назад +5

    I love these videos

  • @DivjotSingh
    @DivjotSingh 2 дня назад +1

    These videos are really inspiring!

  • @msclrhd
    @msclrhd 2 дня назад +3

    It would be more spec compliant to have a run method without any parameters that doesn't set the document URL. This is because the calling code set the URL. It is not a problem here, but if other places calling the parser use a different document URL (e.g. about:blank) then those will fail as implemented here.

  • @coooperrr
    @coooperrr 2 дня назад

    Would love to see some code cleanup and refactoring videos. Something very relaxing about these

  • @mats66
    @mats66 2 дня назад +1

    Hardest part.. constructing the test 🤭

  • @StevenHokins
    @StevenHokins 2 дня назад

    Thank you for video ❤

  • @Gondrux
    @Gondrux 2 дня назад +3

    The mimeType in parseFromString in your test is not set correctly, you forgot to put the variable

  • @evyatarshafran5017
    @evyatarshafran5017 2 дня назад

    I would love to do some work on Ladybird! Do you have some sort of getting started video to explain some basic concepts, workflow, etc? Or should I just jump right into cloning the project, building, and letting it rip?

    • @RubenKelevra
      @RubenKelevra 2 дня назад

      Just jump in and find the smallest change you can make, where you have fun with :)

  • @cferreiramoura
    @cferreiramoura 2 дня назад +2

    You missed change the mimetype in the parser string

  • @cajal6
    @cajal6 2 дня назад

    On line 7 of your test, you accidentally hard coded text/html as the mime type. You need to set it to the mime type variable.

  • @RubenKelevra
    @RubenKelevra 2 дня назад

    What about iFrames? ;)

  • @charlotteathena
    @charlotteathena 2 дня назад

    why is it the associated document of the global object and not just `this->url()` to run the DOM parser? since you already set that URL from the associated document

  • @jpack61108
    @jpack61108 2 дня назад

    Why do you write local tests when the wpt tests already cover it?

    • @michaelennen3432
      @michaelennen3432 2 дня назад

      Andreas did mention this in the previous video I believe - he wants to build up a local test base because the local test suite runs much faster than the full WPT test suite.

  • @raajroy5931
    @raajroy5931 20 часов назад

    7:13 what's the font name?

  • @oren2234
    @oren2234 2 дня назад +3

    are you ok ? are you struggling with bills, why the cringe meta thumbnails lately ?

    • @UAVXP
      @UAVXP 2 дня назад +32

      Rude and weak. Seek for some help.

    • @krolmuch
      @krolmuch 2 дня назад +1

      @@UAVXP Rude and weak. Seek for some help.

    • @niazhimselfangels
      @niazhimselfangels 2 дня назад +6

      I think they look better now. It's easy to know that it's his videos and easy to tell if it's a new video. 🎉

    • @awesomekling
      @awesomekling  2 дня назад +47

      I’m doing better than ever, just having fun with thumbnails 🤓

    • @tauiin
      @tauiin 2 дня назад +1

      well thats silly