freenode/#clim - IRC Chatlog
Search
8:40:14
loke___
jackdaniel: I'm working on implementing a new initialisation flag for :APPLICATION panes: :RESDISPLAY-ON-RESIZE-P
8:44:56
jackdaniel
if it doesn't change the default behavior and is cleanly implemented, I see no problem with adding it. My 2 cents: flag during creation should be :redisplay-on-resize t , without -p, because it's creation flag, not a query
8:45:17
jackdaniel
beach: do you think such extension would be OK for application pane if it doesn't change default behavior?
8:46:44
beach
I personally think it's the wrong solution to the problem (as I have said before), but as long as it does not influence the default behavior, I have no objections.
8:46:46
jackdaniel
loke___: it's not only my call, you could raise an issue on github with proposal tagged [rfc], so we could discuss it there in a thread (i.e not irc) - api, what does it change in the current implementation etc
8:48:20
beach
I guess what I need to do, once I have more time on my hands for this kind of stuff, is to provide a solution to the problem based on REPLAY and the manipulation of output records. Hopefully, such a solution would then supersede this one.
8:48:22
jackdaniel
since it is outside of the spec scope, I'd like you to document it as part of the PR
8:48:54
jackdaniel
yes, that's why I'm concentrated more on API, implementation may change with time
8:49:22
flip214
loke___: was that a re-type error, or copy&paste and you mistyped in your sources? shouldn't that have been :REDISPLAY-...
8:49:43
beach
Well, if I can implement a solution according to my preferences, the API will change, because it will not involve REDISPLAY.
8:51:32
beach
Like I said, and like I have said before, I don't think it is a good idea to implement reflow by calling REDISPLAY. I understand that this particular solution to the reflow problem involves REDISPLAY, but I am saying that I am planning (when I have time) to implement a solution to reflow that does NOT involve redisplay.
8:52:21
loke___
The worst part (except that it's horribly inefficitent when called repeatedly) is that it resets the scroll position
8:52:46
beach
Very much so. And, while incremental redisplay could be used, invoking REDISPLAY basically involves re-creating all the output records, and not only the ones that happen to be on display.
8:53:48
jackdaniel
loke___: the main point is that the api doesn't root into redisplay, so we may switch the implementation without bothering the user to use new machinery
8:54:56
loke___
beach: My initial version of the implementation worked fine until I tried it with incremental redisplay
9:01:35
loke___
I try to detect when the resize happens by implementing SHEET-REGION-CHANGED. However, it's called all the time, not only when a resize has happened.
14:49:17
slyrus_
oh, it's ok. but I keep being amazed at 1) how many things are broken and 2) how many things work by sketchy mechanisms :)
14:50:44
slyrus_
I'm coming to the conclusion that there aren't a lot of programs that exercise these code paths
14:50:56
jackdaniel
if someone needs me, I'm going at festival and I won't be around over the weekend (just saying)
14:51:54
jackdaniel
thank you :) well, it is recreation of traditional summer event from the region, "Kupalas Night"
14:52:11
jackdaniel
it goes all over the night, and we are going to spend it in the "Fantasy Village"
16:52:25
oleo
welp Robert Strandhs Second-Climacs also depends on SICL it seems.... the documentation is a bit lacking
17:46:49
beach
oleo: It is going to get worse. The Common Lisp editing mode will depend on the Cleavir compiler framework.
17:47:46
beach
Anyway, time for me to call it a day and go spend time with my (admittedly small) family. I'll be back tomorrow morning (UTC+2).
18:41:55
oleo
a host of libs/frameworks/systems/packages until it gets totally lost in complexity to just run it ?