freenode/#clim - IRC Chatlog
Search
13:11:43
scymtym
and i couldn't find a way to clip the function names such that the output-record size wouldn't grow beyond the clipping region
15:05:31
scymtym
oleo: thanks, but passing any of those to the application pane doesn't seem to change anything
16:00:05
scymtym
oleo: i tried something similar, but aren't the pane classes and the whole setup totally different?
16:01:12
oleo
scymtym, that's the class for instantiating instances, if you instantiate an instance and pass along other options those will also take effect
16:02:28
oleo
so something unspecified at instantiating time will be taken from the class, if you pass a new thing onto the instance it will take effect/override the class version
16:02:42
scymtym
oleo: yes, i'm convinced that i successfully passed the options you suggested to the extended stream instance, but they didn't seem to affect the scrolling or output record size problems. sorry if i'm not understanding what you are suggesting
16:05:15
oleo
that one implements scrollbars, but the growing of the buffer horizontally is independent of that
16:06:28
oleo
i tried only a vertical scroll bar and the buffer would still grow horizontally on input exceeding the pane size, however after adding the :wrap and :text-margin stuff there was no growing afterwards
16:08:35
oleo
and if you don't have a horizontal scrollbar you can't even get back to the beginning of the buffer in some cases so i always include it
16:11:46
jackdaniel
so it seems that even basic draw-rectangle* operation doesn't resize sheet region
16:13:32
oleo
have a look at my version and do a diff, maybe there are some changes i made which made it work......
16:14:35
jackdaniel
I find it too tedious to parse like that. when you load this flame-graph example and comment out this princ 1, does it resize for you correctly?
16:14:47
oleo
i surely made some changes just not sure where and if it was necessary/appropriate....
17:21:55
jackdaniel
if you add (change-space-requirements pane) it will grow as desired without printc etc
17:37:45
jackdaniel
beach: do you remember anything in the spec what mandates pane to resize its viewport on drawing? or the programmer must call change-space-requirements himself?
17:38:17
jackdaniel
we can't call change-space-requirements from inside draw-* functions, because that will lead to infinite recursion
17:43:43
jackdaniel
so this is arguably not a bug, but I'd love to hear opinions if there are some reasons why we shouldn't automate updating space requirements on each iteration of display loop
17:45:30
jackdaniel
http://hellsgate.pl/files/b104d9af (screenshot, I've turned off scrollbars and let it resize the frame itself)
18:38:59
jackdaniel
I have also figured out, how it happens, that when we write string to stream scrolls are extended
18:39:59
jackdaniel
it's thanks to %note-stream-end-of-page similar mechanism called from seos-write-string (and seos-write-newline)
18:41:06
jackdaniel
oleo: the difference you have observed was because you were working on string sequences written to pane, while scymtym used draw-* functions, which doesn't deal with end-of-page and end-of-line