[E-devel] bugfix for e16 crashing with sounds...
OK, so I started to wonder why e16 would segv every time I tried to open the 'About Enlightenment' to check the build date I was actually running... Turns out that sometimes, the 'sound' resource isn't set. And this eventually lead to a crash trying to strcpy from a null pointer down in SclassFind. I'm going to bet that a lot of people never see this because they get saved by the !Conf_sound.enable check just before (This patch *also* fixed a number of other mysterious crashes when e16 tried to put up a dialog box..) Index: src/sound.c === RCS file: /var/cvs/e/e16/e/src/sound.c,v retrieving revision 1.50 diff -u -r1.50 sound.c --- src/sound.c 9 Dec 2006 21:32:03 - 1.50 +++ src/sound.c 13 Jan 2007 10:02:42 - @@ -271,6 +271,9 @@ if (!Conf_sound.enable) return; + if (!name || ! *name) + return; + sclass = SclassFind(name); SclassApply(sclass); } pgpPluP5GYkF2.pgp Description: PGP signature - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] Copyright blurbs (was Re: enlightenment-cvs Digest, Vol 9, Issue 59
On Sat, 13 Jan 2007 05:29:22 PST, [EMAIL PROTECTED] said: > - * Copyright (C) 2000-2006 Carsten Haitzler, Geoff Harrison and various > contributors > - * Copyright (C) 2004-2006 Kim Woelders > + * Copyright (C) 2000-2007 Carsten Haitzler, Geoff Harrison and various > contributors > + * Copyright (C) 2004-2007 Kim Woelders Just as an aside, such legal advice as I've seen on this recommends that you do *not* do a mass rev of all the copyright dates, but to update each one when you actually do something in 2007. This is why you often see stuff that's "2000-2003,2005" - no copyrightable changes happened to it in 2004. (As a practical matter, it doesn't *really* matter since the SCOTUS upheld the Sonny Bono Copyright Mangle Law. But at one time, it *did* matter because portions of the copyright would *expire* in different years. And since most countries are Bern signatories, stuff is automatically copyrighted when you write it, so you can probably batch up the copyright updates for later, once you've got several things that actually got changed this year. It isn't true anymore that anything that goes out without a copyright becomes public domain - these days, you have to *try* to get rid of the copyright. :) pgpF67LGDPH4l.pgp Description: PGP signature - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Evas on Mac OS X
On Thu, 09 Jan 2003 21:35:12 EST, Michael Jennings <[EMAIL PROTECTED]> said: > -O9 is ridiculous. Nothing over -O3 actually means anything, and even > -O3 tends to be a bad idea. He referred to -O3 as "turning on all the optimizations that probably shouldn't even be part of the optimizer." > > It's also worth noting that omitting the frame pointer does give you a > boost but makes debugging essentially impossible, at least with gdb. This topic got beaten to death on the linux-kernel list a while ago. Apparently there's some non-intuitive results here: 1) -Os seems to produce better code than either -O2 or -O3 on some CPUs. On many boxes, especially those with pathetically small L1 instruction caches or prefetch buffers, loop unrolling is actually a Bad Thing in many cases. (This effect has been known about ever since the PDP-10, where the low 16 memory addresses were actually registers - so if you had a tight loop that used 3 registers as variables, getting the code to fit in 13 words and copying it to low memory before execution was a HUGE win). 2) omit-frame-pointer can cause increases in both code size and execution speed - there's apparently a trade-off between the number of cycles and code bytes spend doing frame pointer setup, and the cycles/bytes spent doing stack references. Apparently, if you make more than 5-6 'mov' references to the stack, you're better off *with* a frame pointer - which is why omit-leaf-frame-pointer exists, so non-terminal functions that will beat on the stack will have one, but all those little 3-4 line functions that don't hit the stack much won't... -- Valdis Kletnieks Computer Systems Senior Engineer Virginia Tech msg00028/pgp0.pgp Description: PGP signature
Re: [E-devel] Anybody there?
On Fri, 07 Feb 2003 13:00:08 GMT, Richard Martin <[EMAIL PROTECTED]> said: > Does anyone actually use this list? No. You're hallucinating us. Take your meds... ;) msg00272/pgp0.pgp Description: PGP signature
Re: [E-devel] Re: E CVS: enotes fletch3k
On Mon, 10 Feb 2003 13:08:32 PST, "Kirby Kuehl (kkuehl)" <[EMAIL PROTECTED]> said:
> And from a security perspective, you shouldn't :)
And maintainability too - ever had a script that sometimes worked and
sometimes failed, just because a user had a file in a directory that
shadowed a system command ('test' is a popular whoops) so it matters where
they are cd'ed to? ;)
msg00284/pgp0.pgp
Description: PGP signature
Re: [E-devel] For code owners and contributers
On Fri, 14 Feb 2003 06:37:12 EST, Kevin Brosius <[EMAIL PROTECTED]> said: > - I'd suggest not using the ChangeLog files on pre 1.0 apps/libs and > sticking to it. For packages that are stable, then a ChangeLog should This implies that we force something to go to 1.0 when there's enough usage that people will care about changelogs. openssl 0.9.whatever pam 0.75 enlightenment 0.16.5 Seeing the problem? ;) msg00299/pgp0.pgp Description: PGP signature
Re: [E-devel] Keybindings
On Fri, 07 Mar 2003 14:39:27 GMT, Richard Martin <[EMAIL PROTECTED]> said: > Can anyone tell me how to set-up keybindings in E that use the "Windoze" > keys, the ALT-GR key or that funny little key with the picture of a > context-menu on it? First, just run the 'xev' utility, and see what keycode those keys generate. If they don't generate a keycode that 'xev' can see, you're basically screwed and need to fix/hack the X server so they do get generated. Just as a data point, on my keyboard, I get: for 'left m$ flag': KeyPress event, serial 20, synthetic NO, window 0x261, root 0x8c, subw 0x0, time 5940290, (89,72), root:(771,1089), state 0x0, keycode 115 (keysym 0x0, NoSymbol), same_screen YES, XLookupString gives 0 bytes: "" for 'right m$ flag': KeyPress event, serial 25, synthetic NO, window 0x261, root 0x8c, subw 0x0, time 5944298, (89,72), root:(771,1089), state 0x0, keycode 116 (keysym 0x0, NoSymbol), same_screen YES, XLookupString gives 0 bytes: "" for 'funky-context-menu': KeyPress event, serial 25, synthetic NO, window 0x261, root 0x8c, subw 0x0, time 5944958, (89,72), root:(771,1089), state 0x0, keycode 117 (keysym 0x0, NoSymbol), same_screen YES, XLookupString gives 0 bytes: "" Your keycodes may differ. Second, you need to have a keysym other than 'NoSymbol'. This may be doable by beating up on /usr/X11R6/lib/X11/XKeysymDB but may require more - I haven't actually tried it. Once you get them reporting a keysym, then you can just go ahead and use them like any other keysym Incidentally, the same technique would apply to get (for instance) the row of cute little buttons on a Microsoft Natural keyboard working - the ones labelled back/forward/stop/.../next track/prev trak/media... -- Valdis Kletnieks Computer Systems Senior Engineer Virginia Tech pgp0.pgp Description: PGP signature
Re: [E-devel] Background displays Weather
On Wed, 12 Mar 2003 09:12:13 GMT, Richard Martin said:
> Security is the browser's responsibility. You cannot blame a language
> for insecure implementations of it. (You *can* blame for trying to be
> too much like Java whilst discarding type-safety, using a different
> inheritance model etc. -- but that is another story.)
And does this mean that security is the desktop's responsibility? Or
that you can't blame E for allowing leaky Flash plugins?
Note that I don't have a problem with the *concept* of plugins. It was
the choice of a particularly egregious offender that pushed my button,
although the concept of *scriptable* plugins has lots of security uglies.
I don't think you *can* make Javascript secure - its basic security model
is just so fucked - so yeah, I'm blaming JS for discarding type safety
and having a b0rken inheritance model, etc etc. ;)
--
Valdis Kletnieks
Computer Systems Senior Engineer
Virginia Tech
pgp0.pgp
Description: PGP signature
Re: [E-devel] Background displays Weather
On Thu, 13 Mar 2003 10:28:41 GMT, Richard Martin <[EMAIL PROTECTED]> said: > Could this whole discussion be made irrelevant if people could > "full-screen" (borderless) an application and push it to the bottom > layer? Then anyone could effectivly make *any* application their > background. I am struggling to think how this would differ from the > conventional notion of a background. Well, there's the pesky notion of 'left/middle/right mouse over the root window pops up WM menus while they do program-dependent things over application windows" However, if the "at the bottom" window sets itself to NoInput and the WM is grabbing the events, that can be finessed. pgp0.pgp Description: PGP signature
Re: [E-devel] Re: E17 future
On Thu, 13 Mar 2003 11:41:51 PST, Xun Cheng <[EMAIL PROTECTED]> said: > I actually hope E17 could get rid of dependencies > on gtk, freetype etc. On the other hand, if a package like gtk or freetype fulfills the need, you get several benefits: 1) You save programmer time not re-creating the library. 2) You get to share the shared library instead of dragging along your own subtly different library. Take your 'etc' to the logical conclusion. Get rid of dependencies on libX11? libc? How much longer will E17 take if those have to be re-done because we're trying to "get rid of a dependency"? ;) pgp0.pgp Description: PGP signature
Re: [E-devel] While we await E17...
On Sun, 16 Mar 2003 12:56:23 +0100, Kim Woelders <[EMAIL PROTECTED]> said: > My primary focus has been on making things work (for me) on RH-8.0 > with a Gnome desktop. I have tried not to break backward compatibility > (although it probably would be cleaner an more efficient just to take > out the old Gnome/KDE stuff). Things also seem to work fairly OK on a > RH-8.0 KDE desktop. > I have not tested with Gnome>2.0 or KDE>3.0. While we're on the topic, has anybody whomped on the e_gen_gnome_menu script to deal with the Gnome 2.0+ 'desktop' files? (Probably the KDE ones need whomping too, but I don't have any KDE-ish apps, and I *do* use some of the Gnome apps under E)... pgp0.pgp Description: PGP signature
Re: [E-devel] Blocking name resolution...
On Tue, 18 Mar 2003 10:34:27 CST, Ibukun Olumuyiwa <[EMAIL PROTECTED]> said: > Non-blocking name resolution would require > - At least one fork() Or good thread support. Or (for some applications) careful design of the event loop around a select(). pgp0.pgp Description: PGP signature
Re: [E-devel] mouse wheel z-value
On Sun, 01 Jun 2003 07:48:28 +0600, Yuri <[EMAIL PROTECTED]> said: > > Better still: > > > > h=(ee->z<0)?((h++ == 24)?0:h):((h-- == -1)?23:h); > > h=(ee->z<0)?((++h == 24)?0:h):((--h == -1)?23:h); actually ;) http://www.ioccc.org ;) pgp0.pgp Description: PGP signature
Re: [E-devel] e16.6 configure script
On Mon, 14 Jul 2003 16:00:52 EDT, Michael Jennings <[EMAIL PROTECTED]> said: > On Monday, 14 July 2003, at 21:50:46 (+0200), > > As I see it, the people who want enlightenment into a separate > > folder, still can do that with --prefix... > > No, they can't. FSSTD uses things like /bin, whereas the > default way uses things like /enlightenment/bin. what's wrong with ./configure --prefix=/usr/whatever/enlightenment and letting it tack the /bin on that? pgp0.pgp Description: PGP signature
Re: [E-devel] e16.6 BUGS
On Mon, 11 Aug 2003 21:14:58 +0200, Kim Woelders <[EMAIL PROTECTED]> said: > Dhruba Bandopadhyay wrote: > > > (1) Sticky settings: I think this one has already been dealt with but > > just in case it hasn't here it is. Sometimes, when a window is made > > sticky and in 'Remember...' the sticky checkbox is ticked, on next > > login it is not remembered. You have to tick and untick options in > > Remember... dialogue box repeatedly to achieve this effect. Please > > let me know if this has been disposed of. > > > I don't recall ever having experienced or heard of this problem. > It is quite a lot easier to prove the presence than the absence of a > bug, so I have to assume that if a bug isn't reported, it doesn't exist. > This one hasn't been reported, therefore, it doesn't exist :-) I've seen this one a few times as well, but brushed them off as collateral damage from the fact I often exit from E16 via 'shutdown -h' or other similar take-no-prisoners exit (telinit 3 is good too.. ;) Does the e16 codebase write changes to configuration details like the 'remember' list when the change is made, or only on exit? If the latter, it's a non-bug (unless the failure to save changes when made is considered a bug/misfeature). pgp0.pgp Description: PGP signature
Re: [E-devel] e16.6 BUGS
On Mon, 11 Aug 2003 20:14:14 PDT, Ben Ford said: > XMMS doesn't remember the sticky state. Other apps are fine. Well, xmms tries real hard to do its own window placement, which often earns it a "Does Not Play Well With Other Window Managers" sticker. pgp0.pgp Description: PGP signature
Re: [E-devel] Moderator approval on e-users?
On Sat, 16 Aug 2003 13:30:24 BST, Dhruba Bandopadhyay <[EMAIL PROTECTED]> said: > To: [EMAIL PROTECTED] First guess is that the above address isn't *identical* to the address you're subscribed with. The most common reason is confusion between "foo.com" and "mail.foo.com"... > Posting to a restricted list by sender requires approval Look through the posting you replied to (if you still have it) for Received: lines that contain 'for [EMAIL PROTECTED]' - that will be the address that the list thinks you are. You'll probably find at least one, since you seem to be using Postfix: Received: from localhost (81-86-141-215.dsl.pipex.com [81.86.141.215]) by shockwave.systems.pipex.net (Postfix) with SMTP id 67C3516000166 for <[EMAIL PROTECTED]>; Sat, 16 Aug 2003 13:28:49 +0100 (BST) Hope that helps pgp0.pgp Description: PGP signature
Re: [E-devel] Edjified Eterm
On Wed, 24 Sep 2003 03:22:17 PDT, Ben Rockwood <[EMAIL PROTECTED]> said: > it'd undoubtly be a resource hog, but damn you could do some awsome themes > for Eterm with Edje for the interface. Kool theme ideas would be text fx > (fading, etc), theming _above_ the rendered text (glass effects), full > motion pullouts for the buttonbars, scrollbars and Escreen tabs, full > control over the text apperience of your terminal, etc. > Thoughts? Cautionary words that seem to be needing to be said once in a while: Eterm is a terminal emulator. 'Kool' often translates into 'unreadable on a long-term basis' - the fact that something is doable isn't always sufficient reason for actually doing it. pgp0.pgp Description: PGP signature
Re: [E-devel] Edjified Eterm
On Wed, 24 Sep 2003 15:41:07 +0200, Cristalle Azundris Sabon <[EMAIL PROTECTED]> said: > That was my first reaction too, but I think as long as you play > fancy with the buttons etc. only (and leave the actual text > widget alone to a large extent) it might work. Time will tell > whether anybody finds it worth their time to hack it in. My real concern is that there's a lot more copies of Photoshop out there than there are actual qualified graphic artists... You know *exactly* what I mean. :) I'm not opposed to gratuitous eye candy, but only when it's not getting in the way. If I hated eye candy per se, I'd not use the Ganymede theme ;) pgp0.pgp Description: PGP signature
Re: [E-devel] Problem in 0.16.6-pre8
On Wed, 15 Oct 2003 22:52:15 +0300, Onur Kucuk <[EMAIL PROTECTED]> said: > When I have the "Focus Follows Mouse Click" set, for example I start an > Eterm window and execute gqview inside it, gqview pops up. While gqview has the > focus, I close it and the focus does not return back to Eterm. At this point > the keyboard becomes useless, things like pressing "alt-tab" (to change to an > open window) or "alt-shift-cursor_key" (to move around the virtual screen) does > nothing. If you mouse-click on a window (the Eterm, or anyplace else), does the keyboard start working again? It *sounds* like -pre8 isn't DTRT if *nobody* has focus at the moment (perhaps it should have grabbed the focus itself when the gqview window evaporates, or something?) pgp0.pgp Description: PGP signature
Re: [E-devel] sideways terminal
On Wed, 22 Oct 2003 22:05:06 EDT, Eddie Anzalone <[EMAIL PROTECTED]> said: > Hi. I haven't found that eterm or another x terminal application is cabable of being displayed sideways. So instead of the terminal being oriented left->r ight (like everything on the screen), it would be oriented top->bottom and it w ould scroll towards the center of the screen. OK.. What *problem* are you trying to solve here? Display of languages that actually render top-to-bottom, or dealing with a monitor that will physically rotate for portrait/landscape, or what? I ask because the "best solution" may be different depending on what you're trying to accomplish with a sideways terminal pgp0.pgp Description: PGP signature
Re: [E-devel] window placement setting ?
On Wed, 05 Nov 2003 21:52:37 EST, Alan Schmitt <[EMAIL PROTECTED]> said: > Thanks, but I'm afraid that would not work. The typical usage I have in > mind is: I open an Eterm (which I configured to be half a screen wide), > which gets on the left. Then I open another one, and I'd like it to be > on the right hand side of the screen, where there is nothing yet. Odd, at least *my* copy of 16.5 goes to great lengths to place a new window where it will overlap the smallest amount of other windows (so if there's enough clear real estate it will place it there, ISTR Raster giving a clear description of the algorithm once... pgp0.pgp Description: PGP signature
Sound events (was Re: [E-devel] Shrinking DR16's Dist Size
On Tue, 11 Nov 2003 15:28:23 +0100, Cristalle Azundris Sabon <[EMAIL PROTECTED]> said: > I think they're very much like real alpha that way -- good fun for > about half an hour, and then you turn it off and go back to work. Yes, a lot of the sound events get old pretty quick. On the other hand, I've grown attached to the SOUND_FOCUS_SET to remind me when I've managed to knock the mouse pointer into another window (the one drawback of using focus-follows-mouse), and SOUND_WINDOW_SLIDE is good for telling me that something on another desktop just popped something up that needs attention pgp0.pgp Description: PGP signature
Re: [E-devel] Re: Sound events
On Tue, 11 Nov 2003 19:43:45 +0100, Cristalle Azundris Sabon <[EMAIL PROTECTED]> said: > I activated the self-same feature after my above post, > and got sick of it after almost exactly half an hour -- > specifically because of the focus change, didn't hear > much of the others. You probably have a more subtle > sound for it than I do. I'm using the 'SOUND_FOCUS_SET.wav' from brushedmetal-tigert. Just goes 'tick'. Which is pretty subtle, considering > something happens. Less annoying than sound, especially if you > share an office. I'm a cubicle dweller, so there was only one real solution: http://www.kenwoodusa.com/product/product.jsp?productTypeId=32&productId=31 The denizens of neighboring cubicles don't appreciate my taste in 80's metal... ;) pgp0.pgp Description: PGP signature
Re: [E-devel] Re: Sound events
On Tue, 11 Nov 2003 15:04:10 EST, Michael Jennings <[EMAIL PROTECTED]> said: > http://www.sonystyle.com/is-bin/INTERSHOP.enfinity/eCS/Store/en/-/USD/SY_Disp layProductInformation-Start?CategoryName=pa_Headphones_NoiseCanceling&ProductSK U=MDRNC20&Dept=pa No, although I'll keep those in mind if I have to spend a lot of time across the hall in the machine room. Supercomputers tend to be NOISY. ;) (Yes, I know there's nicer headphones out there - but they tend to be a waste after a certain point - when the sound card is only so good, even perfect headphones don't make things any better.. ;) pgp0.pgp Description: PGP signature
Re: [E-devel] Re: [enlightenment-devel] osxdocker
On Tue, 25 Nov 2003 14:17:38 GMT, Gen Zhang said: > actually, on a side note to rasterman (and somewhat off-topic) has anyone play > around with the freedesktop.org xserver? the one that supposed _true_ alpha > blending? i've got the thing compiled, but it fails somewhat miserably to > work. http://www.freedesktop.org/Software/xserver says: The X server project holds sources to build an X server separately from a full X distribution. The only drivers supplied are based on the kdrive framework. Does the kdrive stuff know about your graphics card? pgp0.pgp Description: PGP signature
Re: [E-devel] expocity
On Tue, 25 Nov 2003 17:23:46 +0100, Cristalle Azundris Sabon <[EMAIL PROTECTED]> said: > No offence, but the layout looks terribly messy to me. > Is that the best they can do? The wonder is not how well the dancing bear dances, but that he does so at all... pgp0.pgp Description: PGP signature
Re: [E-devel] expocity
On Tue, 25 Nov 2003 23:08:24 +0100, Cristalle Azundris Sabon <[EMAIL PROTECTED]> said: > [EMAIL PROTECTED] writes: > > The wonder is not how well the dancing bear dances, but that he does > > so at all... > > I'm Jack's total failure to be impressed. I suspect you're dis-impressed for the exact same reasons I was dis-impressed by the idea of an Edjified Eterm about 2 months ago :) pgp0.pgp Description: PGP signature
Re: [E-devel] Priority (bis)
On Wed, 18 Feb 2004 11:59:23 +0900, Carsten Haitzler said: > form CVS. CVS is a DEVELOPER tool. as a developer you arrange to be in sync with > the autotools needed to build your project - you learn what is needed. it comes I've found the following trick to be helpful when dealing with projects that have squirrely requirements that either lead or lag what all the rest of the code on your machine wants: mkdir ~/bin/project-name cd ~/bin/project-name ln -s /usr/bin/autofoo-squirrelly-version autofoo # make a link for each tool.. export PATH=~/bin/project-name:$PATH So if your system wants 'automake' to point to automake-1.8, but FooBar insists that it be 1.13: mkdir ~/bin/FooBar ln -s /usr/bin/automake-1.13 ~/bin/FooBar/automake and you should be set for most stuff. Remember to install automake 1.13 where it doesn't conflict file-wise (that's what 'configure --prefix=' is for ;) pgp0.pgp Description: PGP signature
Re: [E-devel] Microsoft patenting Enlightenment virtual desktop manager?
On Thu, 26 Feb 2004 17:13:07 +0900, Carsten Haitzler said: > as far as i'm concerned. *shrug*. the day Ms sues ANYONE (including me) i will > get up and fight this tooth & nail with everything i can find. i don't like Your best bet is *right* *now* do the following: 1) Note the *filing* date on their patent. 2) Decide if your code was written before or after that date. 3) If before, drop the USPTO and Microsoft a note that says you have prior art for the patent in question. 4) If your code was written after, run like hell to someplace that has sane patent laws. :) pgp0.pgp Description: PGP signature
Re: [E-devel] xrandr + epplet for DR 16 ?
On Thu, 04 Mar 2004 14:31:26 +0100, Felix Maibaum <[EMAIL PROTECTED]> said: > Hi all. > > I just searched the archives for xrandr, and up came nothing. Which is very > unfortunate, because nothing beats using my two monitors with enlightenment, > but whenever I want to stop using the second monitor, I have to log out and > use a different WM, so I don't screw up my nicely arranged Iconboxes and > desktop pagers. This sounds like 'xinerama' extension that allows 2 monitors to act as one large piece of real estate. 'xrandr' is the 'rotate and render' extension, which shouldn't have anything to do with number of monitors. pgp0.pgp Description: PGP signature
Re: [E-devel] xrandr + epplet for DR 16 ?
On Thu, 04 Mar 2004 20:33:44 +0100, Felix Maibaum said: > aside from the usefulness for on-the-fly resolution switching, seperate setups > for different resolutions also have the advantage, that it would be possible > to regularly use the same home directory on several machines, e.g. laptop, > desktop, and thin X-client, which typically all have different resolutions. There's some tricky corner problems to deal with - for instance, what do you do with a config that's designed for 2 1920x1400 monitors that includes 3 or 4 1200x1100 windows, and it's asked to squeeze that all into a 800x600 window? ;) (Yes, it's an extreme case, but one you have to deal with.) pgp0.pgp Description: PGP signature
Re: [E-devel] evas mmx issues
On Sun, 21 Mar 2004 15:55:05 CST, Daniel Hansen <[EMAIL PROTECTED]> said: > For a while I have been having problems when compiling mmx support into > evas. If I compile without any cpu optimizations, I have no problems, but > once I compile in mmx, (since I have a p2) any efl app run at the same > Also, my p2 doesnt support mmx2, only mmx, and so I think the problem > could be related to that. Then again, I'm not a programmer, so I really > don't know. Anyway, I hope that helps and I hope the bug is fixed :) I don't see any bug here. All I see is somebody who compiled something, asked it to generate instruction codes that aren't supported on their CPU, and then get surprised when the resulting program rolls over and dies: > (gdb) backtrace > #0 evas_common_cpu_mmx2_test () at evas_cpu.c:39 "Doctor, it hurts when I do this" "Don't do that then..." pgp0.pgp Description: PGP signature
Re: [E-devel] Brainstorming: SWF,Perl and Evas
On Sun, 11 Apr 2004 04:26:25 PDT, Ogla Sungutay <[EMAIL PROTECTED]> said: > Here is the funny part: How does an Mozilla-Evas > plug-in sound? This would buy you what, realistically, as far as buy-in goes? It's one thing to do it as a proof-of-concept, but there's a large bootstrap issue here: What (possibly only percieved) benefit do people get from installing the plug-in, and what reason do web content creators have to use the format? For whatever level of "evil" you wish to assign to .SWF and RealMedia formats, the point still remains that people install the plugins in order to display *types* of content that would otherwise not be available. Unless your plugin addresses an identifiable need, it's basically dead-on-arrival. SWF fills a need for bandwidth-friendly animation, RealMedia fills a need for audio/video, and the various VRML plugins fill the needs of those users. What would an Evas plugin actually do besides "be cool"? (Anybody else remember the days when Netscape's website had a plugin page that listed 3 zillion plugins, at least half of which had blurbs that totally failed to explain why it was worth the effort of downloading and installing said plugin?) pgp0.pgp Description: PGP signature
Re: [E-devel] translucent window moving
On Thu, 13 May 2004 19:09:02 EDT, Michael Jennings <[EMAIL PROTECTED]> said: > On Friday, 14 May 2004, at 00:55:40 (+0200), > Andreas Volz wrote: > > > I noticed a very bad effect with translucent window moving. This is what > > I did: > > > > 1. activate translucent window moving > > 2. activate magic borders > > 3. open some windows on one virtual desktop (e.g. 1) > > 4. open some windows on neighbour virtual desktop (e.g. 2) > > 5. drag one window from virtual desktop 1 to 2 and back *without* > > dropping the window. > > > > Can you see the problem? After leaving the first virtual desktop the > > windows (the dragged and all others) aren't painted until the dragged > > window is dropped. > > That's how translucent moves work. The server is grabbed, and it's > not released until the window is placed. There are similar issues with "opaque" moves as well. Grab a window, and if you drag it off the edge of the screen to the next virtual desktop, it's at the BOTTOM of the stacking order - most annoying if you're dragging a smallish eterm onto the next desktop and it ends up behind a large mozilla window. Grab the opaque window, and try to take it to another desktop entirely, and it goes invisible - borders and all. All you get is the X cursor, and the little box that has the size and coordinate in it I suspect that window drags are scrozzling the stacking order, and pushing the window being moved to the bottom - and when moving to another desktop it's ending up behind the other desktop's background vroot. Then the stacking order gets fixed correctly when you finally drop the window at its destination. I however can't point at a cvs commit or a line of code I can prove is doing it, but this looks like a Changelog entry to start at: Fri Apr 9 00:33:46 CEST 2004 (Kim) Introduce debug print function. Fix stacking after layer change. Fix segv when closing windows. Fix stacking after window move. Cleanups. Hope that helps pgp0.pgp Description: PGP signature
Re: [E-devel] where is the current theme?
On Tue, 27 Apr 2004 10:09:44 +0800, Didier Casse <[EMAIL PROTECTED]> said: > Hi all, > I just wanted to know where is the information about the current > theme stored? i.e let's say I using the theme "axios" and I want to make > a simple Perl program tell me which theme I'm using right now, where do I > look for it?. Thanks. #/bin/bash CURRENT=`/usr/local/bin/eesh -ewait current_theme` echo $CURRENT Translating to Perl or what have you should be obvious. Fix the path to eesh to suit your install. pgp0.pgp Description: PGP signature
Re: [E-devel] [OT] pass by reference problems
On Fri, 07 May 2004 01:48:29 +0200, Morten Nilsen <[EMAIL PROTECTED]> said:
> **foo = "Spin the big yellow thing";
> results Spin the big yellow thing
> Segmentation fault
>
> If anyone can give me a tip about why this segfaults when using strtok,
> I'd be very happy :)
Subtle bug. ;)
First, we point **foo at a copy of the string. Then we call strtok() on that
string. Two things to note:
1) strtok *modified* the string to keep track of its state between calls (so it
knows where to pick up parsing on the next call).
2) gcc by default will put string constants in a non-writable memory segment
so they can't be accidentally splatted.
Solutions:
1) **foo = strdup("Spin the big yellow thing");
This will malloc() some space, copy the string into it. Remember that for good
programming practice, you will need to free() that string if/when you're done with
it.
2) If using gcc, compile with -fwriteable-strings. This doesn't fix your real
problem, but at least you won't crash.
pgp0.pgp
Description: PGP signature
Re: [E-devel] X.org
On Tue, 08 Jun 2004 14:06:40 MDT, [EMAIL PROTECTED] said: > E17 have support to X.org(all features) ? and what's the diferent between > Evas+Canvas and Cario+Glitz? Umm.. which "all features" do you mean, exactly? There's quite a few features in there that E17 doesn't care about one way or another. For instance, there's support for setting the gamma value for the monitor. which E shouldn't care about, that's what xgamma and xvidtune are for. On the other hand, E may care about *reading* the gamma values (or more likely, E will call libraries like libpng that will care about reading the gamma values, in order to adjust an image, at which point the question becomes "does libpng support it?") pgppEvjycn84u.pgp Description: PGP signature
Re: [E-devel] X.org
On Wed, 09 Jun 2004 11:03:45 +0900, Carsten Haitzler said: > well actually why would X read the gamma? the video card can adjust the DAC to > compensate for the entire screen without a single cpu cycle being needed :) i > see gamma correction as an "SEP" (somone elses problem). sure we could provide a > small "set your gamma correction" tool that simply asks x to adjust its gamma > correction... but thats as far as i see our involvement being needed. also with > the increasing spread of lcd screens - gamma is looking much nicer :) That was actually my point - E doesn't need to support the X gamma features :) Note however that it isn't always that easy - note the PNG issue (what do you do if you have one picture that has a nominal gamma of 1.25, another that has a gamma of 2.1. Oh, and this laptop of mine - the LCD reports a gamma of 3.55, but the external Gateway monitor has 2.05... Fortunately, once again E doesn't have to support it itself - it can push the PNG problem off on libpng.so and let *it* worry about it. ;) And as you noted, any remaining issues can be pushed off on the user's .xsession file :) (And gamma correction was *one* example - there's plenty of other things in the "all features" that E doesn't have any business "supporting" - even though I may like eye candy, I shudder to think of what *possible* support (other than "stay the f--k out of the way") E could give to the XVideo-MotionCompensation extension. ;) pgpjALZMO5AFJ.pgp Description: PGP signature
Re: [E-devel] X.org
On Thu, 10 Jun 2004 13:12:11 +0900, Carsten Haitzler said: > hey! dont rule out XVideo... u havent seen emotion yet have you? > > ./emotion dvd:// > > woop dvd in a canvas! :) (and divx, avi, wmv, mpeg, .mov) :) Yes, but why would the *WINDOW MANAGER* be doing that? ;) pgpUIuTXoNmZM.pgp Description: PGP signature
Re: [E-devel] How amateurish!
On Tue, 03 Aug 2004 00:42:43 PDT, Ogla Sungutay <[EMAIL PROTECTED]> said: > Also, as an editor she should have been > clear,specific, and precise when flaming the WM,I am > guessing the E16. The term 'visual usuability' is > rather vague; what is the 'eye candy' she is > referring? Obviously she saw a screen shot of the Ganymede theme, or one of the even more outrageous themes pgpXo9TIjJVky.pgp Description: PGP signature
Re: [E-devel] Time to get started.
On Fri, 06 Aug 2004 10:17:19 +1000, Nick Blievers said: > I dont disagree that a simple GUI editor should be the default but lets do it > the Unix way, and > have an EDITOR variable set. Else we end up with yet another way of setting the > default editor. This loses big time for those people who have EDITOR set to something that will DTRT when invoked in any sort of vt-100 emulator (everything from a Linux console to a telnet/ssh to an xterm). "The Unix Way" has long assumed that using your shell's version of 'export EDITOR=/usr/bin/vi' will do something reasonable. And in fact, that's quite sane until some application decides that $EDITOR is presumed to be smart enough to open an X window all by itself, without a 'FooTerm -e $EDITOR $*' wrapper to do the windowing stuff for it. Using $XEDITOR instead, to denote a GUI-aware editor, might be a good idea pgpoY2CtbluFt.pgp Description: PGP signature
Re: [E-devel] Edje: Scroll text with wrap-around
On Sat, 09 Oct 2004 22:16:25 +0200, Tilman Sauerbeck said: > Hi, > once again, I need a nice way to show a large string in a small text > part, i.e. the text usually won't fit, so I have to scroll it. > > Basically, this can be achieved by using two states: > One which aligns the text left and another which aligns it right. > Then you'd use a program to animate between these two states. Consider the case of a 15-char text and a 6-char window. Then you can display the left 6, or the right 6, but you can't get the middle 3 displayed. Scrolling is a bit harder than that... :) pgpBuWGKhD2S3.pgp Description: PGP signature
Re: [E-devel] Conference.
On Wed, 27 Oct 2004 13:02:56 +0200, Cristalle Azundris Sabon said: > Am 4074. September 1993 schrieb Carsten Haitzler: > > all the stuff we used (software-wise) is all in CVS. in fact it is > > more advanced/better now than it was then. the only other thing is > > the pre5 tarball on rasterman.com/files.html thats not in cvs. > > evidence as well, for the time being. Only if it gets subpoenaed. ;) (Sorry, I couldn't resist ;) pgpuns9EWHKcv.pgp Description: PGP signature
Re: [E-devel] DNS issues?
On Wed, 10 Nov 2004 07:12:46 +1100, Dan said: > Maybe it's just me ( I'm no networking expert ), but I'm pretty sure > there is a DNS issue. I can't reach the website. > > dig is giving me: > > proto ~ # dig enlightenment.org > ;; ->>HEADER<<- opcode: QUERY, status: NXDOMAIN, id: 23201 > Am I supposed to get an 'answer section'? There's not much of an answer to be had if the status is NXDOMAIN... ;) (Yes, it tosses NXDOMAIN here too, tested both of the .ORG NS entries...) pgpLvJAVcfi5D.pgp Description: PGP signature
Re: [E-devel] Im new here!
On Fri, 26 Nov 2004 09:00:14 -0300, Douglas De Toni said: > To start with it I think I could translate the incoming E17 to portuguese, the > programs tools and documentations its a lot of work but I can find some > help over here!! Feel free - translation is something that very rarely gets done by the core coders, simply because they don't speak Portuguese or Chinese or Swedish or whatever. So volunteers who read/write the language are always appreciated. (I'd do a Latvian translation, except that although my Latvian is good enough to deal with a newspaper or a TV broadcast, my technical Latvian is non-existent. You rarely see "iconify" in the newspaper. :) > After that if my C skills grow faster and faster maybe I can get my hands in > some code ?? : ) A good way to get started on this is to find a bug, quirk, or irritating misfeature, and fix it - quite often a bug is (a) caused by an error that only affects a few lines of code (forget to initialize something, or do 2 things in the wrong order, etc), and (b) not something that the original programmer can easily duplicate for debugging. Don't worry if you find the problem and don't know how to fix it - quite often, being able to narrow it down from "The window manager suddenly dies" to one function and say "This routine does wrong thing, but I'm not sure what's right here" is enough to let somebody else fix it. Just this week, I learned more than I ever wanted to about the Linux kernel's handling of ioperm() while tracking down a 2-line error in Ingo Molnar's latency patches that caused a nasty memory leak - and I suspect I'm literally the only user who runs both Ingo's patches *and* the buggy gkrellm plug-in that made the leak noticeable. So I ended up fixing 2 separate small bugs that combined to cause major issues on my laptop.. And the importance of having a small army of people finding and fixing these little 2 and 3 line errors cannot be underestimated - I suspect that there isn't any truly successful open-source project that doesn't have such an army pgpZn0KHN8PQT.pgp Description: PGP signature
Re: [E-devel] Engage / strcmp v. strncmp
On Fri, 17 Dec 2004 15:11:43 EST, Frederick Heckel said:
> Anyway, a null check needs to be added to Engage's icon.c, in
> od_icon_mapping_get() to fix the problem. app->winclass was null. Engage
> starts up fine for me with this quick change:
>
> 358c358,359
> < if (strcmp(winclass, app->winclass) == 0) {
> ---
> >
> > if (winclass&&app->winclass&&strcmp(winclass, app->winclass) == 0)
This is just begging the question however - is this:
1) A proper fix, because od_icon_mapping_get() could legitimately be called
with a NULL and it should Do Something Reasonable
2) Papering over the *real* bug, which is that somebody else is failing to
set winclass or app->winclass before calling the function?
If it's case (2), then we're just looking at a matter of time before the *next*
symptom shows up.. ;)
pgpWUu0AdGx0j.pgp
Description: PGP signature
Re: [E-devel] Engage / strcmp v. strncmp
On Fri, 17 Dec 2004 16:15:16 EST, Frederick Heckel said: > What happened here is that there was an empty directory sitting in my > mapping/ dir. It looks like e_app_new is fine with directories, but it > doesn't actually fill out the app->winclass field if > $subdir/.directory.eet doesn't exist, but still tries to add it as an > app. That sounds buggy to me - it should either invent a default winclass value, or not try to add it as an app. But that's just my opinion based on a quarter century of coding - the actual author of the code gets final call here. ;) pgpotz6jHEj09.pgp Description: PGP signature
[E-devel] Re: enlightenment-cvs digest, Vol 1 #3130 - 3 msgs
On Mon, 07 Feb 2005 20:18:35 PST, [EMAIL PROTECTED] said: > +It is now possible to execute scripts on startup and/or restart. Use > +$ eesh set misc.session.cmd_init
Re: [E-devel] Re: enlightenment-cvs digest, Vol 1 #3130 - 3 msgs
On Tue, 08 Feb 2005 20:07:47 +0100, Kim Woelders said: > Yeah, that was somewhat ambiguous. I hope it's more clear now :) Yeah, that looks better.. ;) pgpSVaIMRY9vd.pgp Description: PGP signature
Re: [E-devel] problems detecting X11 headers?
On Mon, 28 Feb 2005 18:29:32 GMT, =?ISO-8859-1?Q?Andr=E9_Medeiros?= said: > export LDFLAGS="-L/usr/X11R6/lib -lXxf86vm" always works for me. > > Careful though, because it might overwrite whatever you have stored on > that env. var. export LDFLAGS="-L/usr/X11R6/lib -lXxf86vm $LDFLAGS" will prepend to the original with the obvious change to append. Note however that what you probably wanted if it's the C headers is: export CFLAGS="-I/usr/X11R6/include $CFLAGS" (Actually, if your build environment is that scrozzled to need that hint, you probably need *both* CFLAGS and LDFLAGS tweaks...) pgpcxUu7EMIPB.pgp Description: PGP signature
Re: [E-devel] problems detecting X11 headers?
On Mon, 28 Feb 2005 16:44:10 EST, Steev Klimaszewski said: > Personally, I just pass --x-includes=/usr/include and > --x-libraries=/usr/lib with autogen.sh - I use Gentoo with Xorg 6.8.2, > and Donnie is moving everything to /usr instead of /usr/X11R6,. You might want to investigate why those particular x-include and x-library values are needed at all - is your system *not* checking /usr/include and /usr/lib by default already? pgp3TQKLty7j9.pgp Description: PGP signature
Re: [E-devel] problems detecting X11 headers?
On Mon, 28 Feb 2005 19:46:13 EST, Mike Frysinger said: > if test "x$x_libraries" != "x"; then > x_libs="-L$x_libraries -lX11 -lXext" > fi Gaak. :) > x_libs is only set to '-lX11 -lXext' if x_library is a valid path ? talk > about bogus ... here's some sane logic: > > x_libs="-lX11 -lXext" > if test "x$x_libraries" != "x"; then > x_libs="-L$x_libraries $x_libs" > fi > Yes, that looks more sane indeed. pgpDHCbX76fta.pgp Description: PGP signature
