| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12
Comments:
<0> autorepeat isn't disabled, for one
<1> Note that the extra good keycodes I define are probably on keys your computer doesn't have
<1> antifuchs: no, that's the known problem (and probably an xkb bug)
<0> I checked, the keycodes are the same
<0> also, super-modified keys aren't recognized by any application
<1> what does xev(1) say?
<2> antifuchs pasted "xev says..." at http://paste.lisp.org/display/35354
<1> antifuchs: the key is generating a keyrelease event apparently too early
<0> well, argh.
<1> lord knows why
<0> it's not generating any release events at all! /-:
<1> um, the second Super_L entry is a release
<0> oh, right
<0> it doesn't matter how long I press the button, the release event always happens at the same time as the press event
<0> I believe that this is why I switched to OS X (;
<3> antifuchs: Moin, der Herr Oberhofrath!
<0> Jabberwockey: Mahlzeit, Herr Professor!
<3> Ui, ich wurde befrdert :)
<1> antifuchs: ok, well, you can submit a different bug report to your X11 vendor
<0> that might work
<0> on second thought, I could just sell this wretched laptop and get something decent (:
<4> Something that can run McCLIM at FTL!
<1> nono. Something that runs mcclim slowly, so that antifuchs feels he has to improve its performance
<4> I have a laptop that would be perfect for that.
<4> Of course, he'd probably have to make SBCL fit in 32MiB of main memory first.
<3> antifuchs: What kind of laptop do you have?
<3> antifuchs: And how much would you want for it?
<0> Jabberwockey: this is an IBM T40p, bought in nov 2003; it has 512MB of main memory and a CPU that still outruns desktops in terms of sbcl compile speed. (:
<0> (also, a nice 100GB hard drive)
<0> I think 700 eur would work. if you want, I can throw in a new battery (for a few more eur), as well.
<3> Sounds good to me :)
<0> oooh
<5> Hi. Is anyone here?
<5> I'm having problems doing some basic text-processing in lisp.
<5> when (equalp line "CONTENTS")
<5> is the line I have in my loop statement
<6> and?
<5> (loop for line = (read-line stream nil)
<5> is the loop construct.
<6> lisppaste: help
<2> To use the lisppaste bot, visit http://paste.lisp.org/new/lisp and enter your paste.
<5> (equalp line "contents) never returns t
<5> Even though when I have it (read-line) out that line in the text
<5> it returns "contents" like it should, and I know that equalp is not case-sensative
<0> clhs read-line
<7> http://www.lispworks.com/reference/HyperSpec/Body/f_rd_lin.htm
<6> marco__: use string-upcase?
<6> marco__: or well, do not compare it with upcased version?
<1> actually, I'm not sure I believe marco__'s diagnosis of the symptoms
<2> marco__ pasted "Problems" at http://paste.lisp.org/display/35355
<6> and what the problem is?
<6> other than that loop is endless.
<5> it never prints contents
<5> even though I run it on a file in which contents is alone and on one line
<8> there could be a carriage return stuck at the end of the line
<1> try flushing the output buffer
<8> (if the file comes from a different platform)
<5> The file comes from Gutenberg.
<1> try flushing the output buffer
<5> How do I flush the output buffer?
<1> (finish-output)
<5> (finish-output) did not work.
<5> Also note that I tried including a carriage return on the end of CONTENTS
<5> like so: contents^M
<5> because that is how emacs showed the file
<5> neither string worked
<6> does this file has a line "CONTENTS", and not "contents"?
<6> actually no, this is irrelevant.
<0> somebody kick lisppaste, please.
<5> equalp is case insensative though?
<8> if emacs shows the lines with ^M then like I said it has carriage returns
<0> that is, restore non-proxy-errorness
<6> marco__: yeah, it is.
<5> @urfin: then, would I just write "contents^M" in my string? this didn't work.
<5> when I tried it.
<5> Do I need to do something else to get the carriage returns chopped off?
<0> (string-right-trim '(#\Return) (read-line ...)) should do the trick
<8> maybe use something like #.(concatenate 'string "contents" '(#\Return))
<8> oh, yes, chopping them off might be better
<5> that worked!
<5> Thankyou very much.
<5> Noob question: why didn't it work when I just had it (equalp "Contents^M)?
<1> because at a guess you wrote out ^ and M as two characters
<5> Oh. wor.
<5> *wow
<5> Yeah, I'm dumb. Thanks very much :-)
<0> behold! the telepathic debugging skills.
<4> What should happen if you call `stream-advance-to-column' with a column number that the read position is past?
<1> write a newline then spaces? (Dunno)
<4> SBCL does nothing. I shall copy that behavior.
<4> antifuchs: so, have you picked a holiday for the next release yet? :)
<0> not yet (:
<0> but you're going to tell me it's going to have to be soon? (;
<0> we could release on the 7th of feb and name it "Happy Birthday, Mr. Release Manager!" (;
<4> We should probably aim for a release in a month or two.
<4> McCLIM 0.9.5 "Release Manager Birthday"!
<4> 273 lines of tests for 162 lines of code. It better work.
<0> I've now spliit mcclim.asd into parts; so far, things look good
<0> will move clim-gtkairo to its own top-level asd soon, then people should have an easier time installing it (:
<4> Good!
<4> It would be neat if we could end up making Gtkairo the default soon.
<0> oh yes
<4> I'm almost considering possibly evaluating whether or not to look at the horrors of the C API to help out.
<0> ((-:
<9> what's that top-level asd file good for?
<0> _8work: asdf-install will auto-symlink it
<9> I see. That's seems silly though. Why not fix asdf-install to symlink all systems found in the asd file instead?
<0> because internal systems shouldn't be linked? dunno
<4> Committage.
<0> anyway, I would prefer not to change the 3 or 4 versions of asdf-install
<9> Once you've loaded the asd file they are in the global namespace anyway.
<0> ...and wait for compatibility with stuff
<0> feel free to create a better asdf-install.. but please choose a different name (:
<9> So you wouldn't want that change in the SBCL version of asdf-install?
<9> (This channel is already pretty good at ignoring "the other asdf-install", nothing new about that.)
<0> still, the other asdf-install serves a large percentage of asdf-install users.
<4> If the streams abstraction was perfect, I would be able to copy data from one stream to another by merely fitting them together and letting it flow.
<0> I would prefer a solution that works with all asdf-installs, out-of-the-box, without the need for the user to upgrade anything but mcclim.
<6> Athas: yes, this is very reasonable thing to wish for.
<9> Well, yeah.
<9> Can you post your new asd file on the mailing list for review though?
<0> oh, sure
<0> I was going to just commit it (;
<10> Athas: Is tab-layout moving into McCLIM, by the way?
<0> (ah, and gtkairo isn't in the bunch yet)
<9> antifuchs: It's not that I'm opposed to improvements :-), but I'm kind of used to how things work currently.
<0> as are a lot of people here
<4> rydis: no idea, I don't work on that. But I think it is.
<0> I'm sure this will break a lot of setups
<9> antifuchs: Will I be still able to (pushnew :clim-gtkairo *features*) (operate 'load-op :mcclim) and get gtkairo?
<0> that is indeed the plan (:
<9> rydis, Athas: No, unfortunately the tab-layout cannot move into McCLIM yet because it's not open source.
<4> It's not?
<9> Athas: No. The files say (c) mgr and don't mention a license.
<4> Has mgr commented on it?
<1> some might say that the licence is implicitly LGPL
<1> but that would probably take lawyers to decide on
<4> Xof: because of the LGPL-status of McCLIM?
<1> yes
<4> Doesn't mgr use ACL CLIM for development?
<10> By the way, if it isn't known, clim-gtkairo in particular, and McCLIM in general, only works so-so with cmucl-mp/serve-event stuff. (Sorry I can't be more specific in what the problems are.)
<9> mgr has taken interest in my completely revamped tab-layout, and we've had a good discussion about that.
<9> He has not responded to my inquiries regarding the license though.
<10> It's probably mostly a problem with CMUCL, I guess.
<10> (The reason I can't be more specific is that I've got to run. Perhaps I can investigate more and write it up some other time.)
<9> rydis: I wasn't aware that clim-clx has trouble with CMUCL's mp too, but I was told about the gtkairo issues.
<10> _8work: If you'd like to see the problem, start Hemlock, start a slave lisp, (mp::startup-idle-and-top-level-loops), and, for instance, (clim-listener:run-listener :new-process t); there appears to be something that's not yielding as it should. (Or start a process running Hemlock from the listener; it appears to be something with CMUCL:s handling of CLX as opposed to McCLIMs.)
<10> Hmm. I'm running telent-clx, though. Perhaps it's different in CMUCL CLX. I haven't tested in a while, but I think I saw it there, too.
<10> But, as I said, got to leave for work, now.
<9> That sounds like it should be entered into the non-existing McCLIM bug tracker.
Return to
#lisp
or
Go to some related
logs:
#openzaurus
#suse
www.helenus.com
ubunut linksys wireless
sblive asoundrc
#dns
#linux
#perl
moodle hopto
ecore_x_event_screensaver_notify
|
|