https://htmx.org logo
Powered by
# htmx-general
  • g
    Is that event ever triggered? I tested doing a htmx.logAll() and I never see that event. beforeProcess on the other hand I see
  • g
    Ah... Found this function ignoreEventForLogging(eventName) { return eventName === "htmx:afterProcessNode" }
  • m
    🤔
  • m
    there must have been a reason...
  • g
    And it triggers when I do a process... It was the logging that tripped me up
  • m
    why did I do that...
  • m
    there must have been a reason...
  • m
    🤔
  • g
    If only there was a way to attach a comment to every change... 😉
  • g
    We rolled out htmx in production this week. Very pleased with it so far 🙂
  • g
    Combined with a small thing we wrote to replace content over websockets (using Pusher, so not the built in websocket thing in htmx) it's super powerful. Nice work! We looked at hotwire but htmx is less complex and actually more powerful imho
  • m
    Well that’s the nicest thing someone is going to say to me today!
  • g
    you should check this out: https://www.phoenixframework.org/blog/build-a-real-time-twitter-clone-in-15-minutes-with-live-view-and-phoenix-1-5 this is, for me, one of the most inspirational tech demos I've ever seen. I desperately want to think of a good project to try this out on.
  • s
    I actually have a htmx question, can I assume that any event listeners and other things i start on a html fragmet is removed when i use htmx to replace it?
  • s
    imagine i return a html fragmen with htmx and this has a script that does settimeout() .. if i replace this with antoehr htmx call, what happens?
  • g
    Sounds like time for an experiment 🙂
  • g
    But my guess is that the code in the 
    setTimeout()
    lives in its own scope, so it will probably execute separately from htmx. When it does run, however, it seems likely to break because the data it's expecting won't be there -- or worse, it will fail in silent or strange ways, because it's expecting the old DOM structure, but executing on new DOM nodes.
  • m
  • m
    we should clean up everything on removal (by htmx)
  • m
    I actually want to experiment with the observer API and see if we can be more general
  • s
    If I use chrome with Playwright my code works (if I add "time.sleep(0.3)" at some points). But if I use webkit, it fails. See screenshot. How can I debug this?
  • m
    I'd switch to the unminimzed version and then start setting break points
  • m
    a lot of javascript libraries use the load event to wire in functionality, are there any docs for that situation in PLaywright?
  • s
    There are some hints about debugging: https://playwright.dev/docs/next/debug/ How do you test htmx for webkit/safari?
  • s
    I created a minimal example, and tested it with https://www.lambdatest.com/ and now i know that only the Playwright webkit engine does not handle the hx-swap-oob. Safari on MacOS works fine. I created an issue. https://github.com/microsoft/playwright/issues/5894
  • s
    I believ the website has a mistake on the sortable example
  • s
    it should be name="item[]" or else it will not submit as an array of items but instead only as the last value.
  • m
    there are no mistakes, only happy little pull request opportunities
  • m
  • m
    In the following example the parent vals always override the child val/input so 
    large
    and 
    blue
    are always sent with the request.. Bug or feature, or human error?
    Copy code
    html
<form hx-get="/test" hx-vals='{"size": "large", "color": "blue"}'>    
    <button hx-vals='{"size": "small"}'>Button</button>
    <input type="text" name="color" value="red">
</form>
1...196197198...1146Latest