You are right of course!

Yeah, there are some malformed URLs we should tackle for sure.

I have edited the answer. Because with draw.io you can integrate Mermaid out of the box. With R5 you can start using the syntax like instantly.

@valerio.bozzolan Could you try again to install it? There is a readme with new instructions.

In D25577#16640, @mturdus wrote:

Now it's working!

@mturdus right, I don't have the whole picture of the little bit wild codebase 😁 Thanks for the pointer. As far as I can see, this image and the DOMContentLoaded event handler serves no purpose anymore. All is working without. I have renamed the main js and some functions in it for clarity. Would love if you could give it a try.

• Renames behavior and some functions and gets rid of unnecessary image proxy

Fixed by R5:91dd43de7dee: T15781: arc liberate throws exception

@mturdus I have added some console logs in the js code (so patch and purge the caches on <phorge-base>config/cache/!)

• Caches draw.io assets, uses base64 method of diagram and adds some console logs for js debugging

Firefox does work for me. Does Chrome work for you?

@mturdus You don't have to download the diff, you can do an arc patch D25577 (and you should, because I have changed something again). Mh, but I don't know why it should not work.... 🤔

• Dumps base64 image data directly while remarkup rendering and adds js/css files for remarkup object rule

@mturdus Is the phorge_extension.js file loaded?

@mturdus Just test this if you like. I think this is a first step in the right direction: D25577

Well I have worked the entire day on this (and I see your mentioned issues)... main problem is: extension is not working with security.alternate-file-domain. That I have fixed at least for the creation/editing of diagrams. I have changed pretty much and would love to send a diff.

(Working on fixing it...)

I never use them, because there is no real method search. I think, we should be considering generating a documentation with a real php documention tool. But whatever we do: I think we can remove them in their current form.

In D25538#16175, @avivey wrote:

I'd even suggest we can go one step further, and in the Home tab show the Readme above the files (and maybe change the order/highlight some other UI elements too).

• Moves readme to the top in home view

Well, my idea was to use the already implemented logic of the diagram extension and just extend it with another diagram renderer. But @mturdus is not responding 😁

Well, you know what would be next level for a dev? If (s)he clicks on it, the color is copied to the clipboard. We can't use this behavior, because behaviors loose their state in the remarkup cache...

Updated Pygments with pip to 2.17.2 und cut the time in half. So yeah, that's already an improvement.

Well, good to know that it is not something in the phorge codebase. Our server has pygmentize 2.14.0 - but the server itself is really not the best, so that could be an explanation. Maybe we could make some remarks about performance and keeping the versions fresh on the diviner page...

The word solution would solve it from the expecting side, you are right. Here in my company the search is used frequently and only as a plain text search. So we have to escape the special regex signs time and again. Just would be interested if this is seen as an issue by others who uses the feature.

Thanks for the clarification. I still think we should tackle the exception.

In T15728#15768, @valerio.bozzolan wrote:

I tried to understand the situation. Thanks.

In your opinion: on the repo view (with clone buttons) is it really important to link to the other similar "reduced" view (without clone buttons)?

Just asking for a friend: Is there some progress on this?

Maybe this could be integrated in the Diagrams extension...? I mean: the entire infrastructure is there - you just have to decide, what lib you wanne use when creating a new diagram. @mturdus What do you think about this?

In T15673#15587, @valerio.bozzolan wrote:

TRUST ME - YOU DON'T WANT TO LISTEN MY ENGLISH :D :D

Btw now the video is here :D :D AGAIN SORRY PHORGE FOR MY MISTAKES <3 <3

https://video.fosdem.org/2024/h2215/fosdem-2024-3610-unpack-phabricator-welcome-phorge-forking-the-opinionated-open-source-project-manager.av1.webm

• Making @valerio.bozzolan happy

I really have struggled to find any device in our company that can't do await/async. I strongly bevlieve it is a non issue. Let us land this thing now and see if we get any feedback of an error.

• Changes the order of the variables and function calls

• Extends select behavior and only select content on first click

In D25536#15448, @valerio.bozzolan wrote:

Again, non-blocking opinion:

Are we aware that the "async" keyword (that was never used in Phorge) may just cause a crash by syntax error (almost-immediately even without clicking?) by that kind of browsers, so, breaking the page, just because we want to provide a copy function and we want to avoid .then()?

I think we can introduce breaking changes, but it's strange to do that, now, with this minor feature.

If we are aware of this minor risk, indeed let's land.

In D25536#15446, @valerio.bozzolan wrote:

Additional non-blocking thing:

Maybe should be copy(text) with var data defined outside

Bonus point: copy() not defined during every call of JX.Stratcom.listen, but outside (?)

In D25536#15437, @valerio.bozzolan wrote:

Yes, we can avoid async.

So, this is just a tip, to avoid async and also to have everything a little more modular for no reason:

1. remove async from copy()
2. rename copy() to copyWithoutFeedback() and have the function that always return a Promise
   • so copyWithoutFeedback() has return navigator.clipboard.writeText(text); to always return a Promise
   • so copyWithoutFeedback() can use copyWithFeedback() but after that it return Promise.resolve(true) to always return a Promise
   • so copyWithoutFeedback() can fail with just return Promise.reject() to always return a Promise
3. so we can create the function with feedback, copyWithFeedback(), that just calls the previous one, so, copyWithoutFeedback().then(function() { show_success_message() } ); with also a chained .fail( function() { show_error_message(); } ) or something similar

• Updates map

• Ignores async jshint

In D25536#15422, @avivey wrote:

I'll try to see if there's config for the jshint to allow async.

Well, now I have tried to satisfy every wish.

• use async/await for copy promise hell
• make the io button to just a type icon
• select entire text when clicking inside the input field
• patch JsShrink to shrink await/async syntax

(This Quickfix would help: https://github.com/vrana/JsShrink/pull/6/commits/5b2dd500d25b29ace2fbd76fc95b4518ef22abfc 😆 )

await/async won't go through the shrink process without destroying the entire syntax. And if it will go through this jsxmin binary, who can know? Note for the future: Maybe replace the entire shrink process with a new and used php shrinker like JShrink.

@avivey worst case is a syntax error because the operators are not known. Even that could be wrapped inside a try/catch.
@valerio.bozzolan could be done with the current behavior but this kind of implicit actions seems not the phorge way...?

In D25536#15395, @avivey wrote:

Yes, I insist on using the absolute required minimum of Promise-style coding.

I have to wait for the success or failure of the copy action, so I think this ui example makes no sense for my case. The notification should not be triggered by a button click.

Well, I see a tremendous critism and scepticism concerning modern client side techniques here. The world of frontend tooling has changed tremendously in the last 5 years. And I understand concerns, but nobody has to write JavaScript for certain browsers anymore. Javelin for example tries to solve so many problems that Babel solves with one config entry. But that's okay and I like the architecture here and know we can't change everything at once.

The overwriting of variables with media queries could be configured inside the theming class. So possible breakpoints can be part of a theming API.

And my general approach is: when the client can do something (and that even better), why should the server do it?

@avivey CSS vars are a vital part of modern modularized theming. They are versatile, can be overwritten with media queries or per class/element basis. They are editable in browser dev tools. If you would ask me: this step is pretty important.

In T15736#15726, @avivey wrote:

to have a working dev server when developing js - at the moment I have to ./bin/celerity map after every change to js and css to see my changes...

Are you sure about that? I thought celerity map was only needed when adding a file (in dev mode).

In T15039#1358, @bekay wrote:

The clone repo modal could really use some copy buttons.

Preferably replacing the Read / Write Button.