It's F# record types, discriminated unions being stored in LiteDB, a single-file, in-process document database that runs under .NET.

Hello. I need to check if a form is dirty before closing of removing the div. I did that in javascript by using .serialize() on the form and comparing it the similar one saved on load. How can this be done using HTMX?

Hi, I have a slider (slick slider https://kenwheeler.github.io/slick/ ) on one of my pages and I noticed that when I go back to that page by means of an HTMX history restore that it no longer works. The slider is stuck and no longer scrolls. Is there a way to fix this?

would likely require javascript

what are you trying to accomplish?

are you initializing it using

htmx.onLoad

? https://htmx.org/docs/#3rd-party

No, I basically just slammed the appropriate script tag in there and did not edit the standard code of the slider apart from the settings script one has to fill to out to configure it (how many slides, use of dots, breakpoints etc...)

My backend in django and I am trying to do the regular get/post calls for creating an object from the frontend. A button sends a get request and gets back my form in html which is modal in nature (I am following the custom modal design given on htmx.org). What I want is that if the user tries to close the form but has already input some data, I need to check if the form is dirty and then show him another modal to give a fair warning that data may be lost if closed without saving.

OK, to be compatible w/ htmx partial loading you'll need to initialize it when it comes out of history, and the

htmx.onLoad

mechanism is the easiest way to do that

ok, sounds like a pure client side feature, so in this case I would listen for a

change

event on the form and mark the form as dirty. This can be done in javascript or if you are brave, maybe hyperscript

Thanks! That's what I suspected. Just wanted to make sure there was no inbuit htmx feature which deals with this.

Nope, but it does seem like a candidate for an extension!

some hyperscript like that would give you something to check

Another quick question. I am a newbie so I have a few doubts. I looked at hyperscript and it seems html fragments can be swapped using just hyperscript too? So how do we differentiate the two? Is it that htmx deals with all kinds of htm elements whereas hypertext is only for links and buttons?

htmx is for hypermedia-driven interactions w/ a server

hyperscript is (mainly) for front end scripting needs

by and large we would encourage you to use htmx for interacting w/ the server and they hyperscript for small UI tweaks on the front end

hm. I think it would become more clearer when I actually build something using both.

yep

From what I can see, there is still no support for any swapping from right? I have seen that the handling of tags was improved in 1.6.0 but it is only for tags contained in the of the response? If so, I just wanted to know if I was the only one to have this use-case of needing to swap in the head especially for CSS and JS (even if I understand that it is more complicated than just swapping the elements in the actual dom, as it touches on how the browser will load those elements). My use-case is for instance if my fragment coming from the server needs external JS/CSS pulled from a CDN. I could put that globally in the initial document, but I suppose it would be more accurate to only load them at the same time as the fragment when needed.

(and the problem is even more visible for boosted pages at the body level, as it is not possible to load new resources in those dynamically fetched and swapped "full pages")

And I see this issue related to that question: https://github.com/bigskysoftware/htmx/issues/247

I like the idea of a general head-management extension, are you interested in looking into it? To do it right would require using a template-parsing technique, which isn't IE11 compatible, but as an extension that would be OK

are there any security implications with swapping out things like or ? I mean I'm no security engineer but I can just imagine some hackers rubbing their hands together if they see that type of code on my website... or trying to break the playground?

I am currently busy on a personal project with an end-of-year deadline, but after that, I’d be glad to give it a try. I just wanted to be sure first that it was still a thing that people wanted by posting here 🙃

I am by no mean a security expert but I don’t see any problem at the top of my head

There's nothing you can do with htmx that any person can't do with f12 and some time. Unless you're suggesting letting a user somehow affect the html that another user loads

That's where you'd want to be careful. You don't want someone to be able to put something on the server that ends up getting passed to other users. Or to construct your responses using GET parameters, as then people can send malicious links.

But HTMX is quite different to Ember, React, etc - normally there the assumption is that you don't want to be changing the DOM, so there are sandboxes and escaping mechanisms, etc. HTMX is quite explicit about the fact that you're modifying the DOM though. You'll want to have your escaping etc being handled by the server instead.