Geth Terminal-Widgets/main: 12 commits pushed by (Geoffrey Broadwell)++
review: github.com/japhb/Terminal-Widgets/...4ac1252432		 04:22
japhb Wheeeeee
Geth Terminal-Widgets/main: 9035d85d34 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/TextContent.rakumod
Add more flat-spans and flat-string-spans multis
 05:37
Terminal-Widgets/main: 69de09b0a8 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/I18N/Locale.rakumod
Add flat-spans/flat-string-spans forwarders to Locale
Terminal-Widgets/main: 5c6be7f666 | (Geoffrey Broadwell)++ | 6 files
Move various Input/ files to TextContent

  * Labeled can now safely make the label TextContent:D
  * SimpleClickable's decendents handle full TextContent in labels
  * SimpleClickable itself passes full TextContent to its descendents,
   but still smashes to plain-text before writing to the grid
Terminal-Widgets/main: 34c7d0f500 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/Input/SimpleClickable.rakumod
Stop flattening to plain-text in SimpleClickable
 05:46
Terminal-Widgets/main: 1b21579b16 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/TextContent.rakumod
For convenience, handle Str:D children in span-tree()
 06:07
Terminal-Widgets/main: 67fb62c191 | (Geoffrey Broadwell)++ | 5 files
Take advantage of span-tree(Str:D) affordance
Terminal-Widgets/main: cd0d6b8190 | (Geoffrey Broadwell)++ | 2 files
Take advantage of span-tree(Str:D) in examples
Terminal-Widgets/main: cc640392a9 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/Input/Menu.rakumod
Switch Menu Input to TextContent model
 06:30
Terminal-Widgets/main: e896e6ade2 | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/Input/SimpleClickable.rakumod
Fix SimpleClickable color lost in 34c7d0f5
Terminal-Widgets/main: 83146bf38a | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/Input/Text.rakumod
Switch Text Input to TextContent/draw-line-spans

This also clears up the last uses of $.grid.set-span-* outside of where they belong.		 07:11
Terminal-Widgets/main: 717c56fe88 | (Geoffrey Broadwell)++ | 2 files
Remove unneeded 'use's from RichText/TreeView
 07:51
Terminal-Widgets/main: 2f1267565e | (Geoffrey Broadwell)++ | lib/Terminal/Widgets/Input/Text.rakumod
Fix substr mathing errors in Text Input
Terminal-Widgets/main: 49cff50cbc | (Geoffrey Broadwell)++ | 3 files
Remove T::MU::L from a few places

  ... in favor of locale.width where possible
Terminal-Widgets/main: 0f456a73d1 | (Geoffrey Broadwell)++ | Changes
Update Changes for next release
 08:20
Terminal-Widgets/main: 87870b6f0e | (Geoffrey Broadwell)++ | 3 files
0.2.3
japhb patrickb: There's a reasonable chance that 0.2.3 will be the last 0.2.x (and if it's not, I don't expect more than one more). I'll bump to 0.3.x when I start deleting stuff. 08:22
patrickb: Do you use SpanStyle::Span.words anywhere? It doesn't appear to be used anywhere in T::W at the moment. 08:27
patrickb Pretty sure I don't.
Pretty impressive progress! 08:28
The debugger at the moment doesn't use RichText or TreeView anywhere yet. 08:29
So there is no need to hold back nuking them.
I'd prefer directly going for the rework instead of doing it with the legacy impls only to rework it right after. 08:30
I think I can resume work on the debugger pretty soon as I believe I have all external dependencies in place (pty, var-args, DirsOfInterest). I promoted getting rakudo/rakudo#5060 merged to the top of my list. But once that's in, I'll go for the debugger. 08:33
Geth MUGS-UI-TUI/main: 4 commits pushed by (Geoffrey Broadwell)++ 09:22
japhb patrickb: OK, roger that. I'm only holding back nuking them because you wanted me to get to feature parity first, and I'm still a little shy of that. :-) 09:24
I'm going to have to do a bit more to clean up some visual artifacts in MUGS-UI-TUI, but the above commits at least get it starting up and not immediately exploding when you select a main menu item. :-P 09:25
patrickb That feature parity request was out of fear of the new impl not reaching feature parity ever. As long as we get there at some point - go ahead! 09:26
japhb :-D
patrickb: If you haven't already, try the dir-tree example, and let me know if you want me to tweak anything there *other* than line wrapping 09:27
patrickb tweaking as in more features or as in bugs? 09:28
japhb Huh. Either I guess? I don't know of any bugs in there at the moment.
(But the implementor is always blind. ;-) ) 09:29
patrickb šŸ‘šŸ¼
japhb Dang it, I need to put my IRC client under Ghostty. I'm tired of skin tones not working right. >:-|
You know, of all the fixes I've done in the last couple weeks, having keyboard-accessible scrollbars is still my favorite. 09:32
patrickb the example is pretty nice! 09:35
japhb Excellent. :-)
patrickb Is it possible to differentiate clicking on the arrows and the rest of the line?
japhb Differentiate in what way?
(Scrollbar arrowhead clicks are actually handled specially, so I wonder what you mean) 09:36
patrickb I wouldn't necessarily want this in this case. But clicking on the arrow expands / collapses, clicking on the text does the contextual action (in this case print to the log)
So I only want the log, but don't want to collapse the folder. That's currently not possible. 09:37
japhb OH! I see what you mean, I was still stuck on scrollbars.
Yeah, that seems reasonable.
Also might be worth checking standard key bindings against a real-world file manager 09:38
patrickb Is it technically possible to get the scroll wheel going?
japhb Hmmm. I don't actually know what that looks like as raw input. I'll add that to the list to check it out.Ā  09:39
patrickb That's the intuitive first thing I tried - scrolling. :-P
japhb Damn your reasonable expectations! ;-) 09:40
patrickb Persistence of expanded state is also missing.
I.e. expand a few things, collapse the top level node and expand again. Should be as before. 09:41
Ideally this should work even with the underlying tree changing it's shape. 09:42
japhb Good point. I'd held off on that for performance reasons, but at this point I've probably got things good enough to hold that state around and not slow down.
patrickb (That's the case for the locals view in the debugger for which I want to use the tree view.)
japhb Yeah, decent point 09:43
That's part of the functionality I wanted to work towards with the whole Volatile/ directory, but that's mostly just barely enough to support the current demo. 09:44
It'll get there. 09:45
patrickb Don't take my comments as critique. I'm pretty impressed how things are turning out and I fully understand that the process is not at it's end yet. :-) 09:46
japhb I hadn't! I was taking it as constructive feedback. :-)
patrickb the legacy tree view had this duality thing. 09:49
japhb ?
patrickb so either provide a callback that returns the children for a node on demand - that's how the new tree view works, right? 09:50
japhb yep
patrickb Alternatively one could pass a static tree of nodes instead. The TreeView would put a wrapper around to make it adhere to the dynamic interface. 09:51
japhb I might just provide a wrapper in Volatile/Tree for that, so Viewer/Tree always sees the dynamic interface. But yeah, same idea. 09:53
patrickb That's not a hard requirement (actually I think I don't even need that for the debugger), but it's easier to wrap one's head around, so might be a nice addition to get people going.
japhb Yeah, agreed. 09:54
patrickb I recall during my studies how I really struggled with that dynamic tree thing in Qt and was really glad I could just stick together a tree structure and pass that in. 09:55
japhb Yeah, totally fair. I've been trying to put back some affordances (like span-tree() being able to accept plain Str children) that I had initially lost in all the refactoring.
A static tree is good for that. 09:56
16:29 librasteve_ joined 21:54 lizmat_ joined 21:58 lizmat left 23:03 lizmat_ left 23:04 lizmat joined 23:10 Geth left, Geth joined

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!