java-gnome-developer Mailing List for The java-gnome language bindings project (Page 59)
Brought to you by:
afcowie
Menu
▾
▴
java-gnome-developer — For people using the bindings to write their own applications
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(37) |
Dec
(14) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(2) |
Feb
(20) |
Mar
(20) |
Apr
(8) |
May
|
Jun
(1) |
Jul
(6) |
Aug
(39) |
Sep
(37) |
Oct
(34) |
Nov
(50) |
Dec
(22) |
| 2002 |
Jan
(7) |
Feb
(13) |
Mar
(32) |
Apr
(16) |
May
(26) |
Jun
(20) |
Jul
(32) |
Aug
(7) |
Sep
(2) |
Oct
(11) |
Nov
(3) |
Dec
(35) |
| 2003 |
Jan
(11) |
Feb
(3) |
Mar
(8) |
Apr
(3) |
May
(11) |
Jun
(20) |
Jul
(11) |
Aug
(29) |
Sep
(13) |
Oct
(91) |
Nov
(185) |
Dec
(207) |
| 2004 |
Jan
(108) |
Feb
(171) |
Mar
(207) |
Apr
(113) |
May
(22) |
Jun
(53) |
Jul
(69) |
Aug
(43) |
Sep
(34) |
Oct
(182) |
Nov
(101) |
Dec
(61) |
| 2005 |
Jan
(86) |
Feb
(45) |
Mar
(106) |
Apr
(67) |
May
(70) |
Jun
(47) |
Jul
(19) |
Aug
(34) |
Sep
(24) |
Oct
(45) |
Nov
(20) |
Dec
(58) |
| 2006 |
Jan
(21) |
Feb
(21) |
Mar
(16) |
Apr
(24) |
May
(24) |
Jun
(47) |
Jul
(20) |
Aug
(8) |
Sep
(13) |
Oct
(7) |
Nov
(23) |
Dec
(2) |
| 2007 |
Jan
|
Feb
(14) |
Mar
(3) |
Apr
(11) |
May
(1) |
Jun
(15) |
Jul
(2) |
Aug
(5) |
Sep
(10) |
Oct
(5) |
Nov
(1) |
Dec
|
| 2008 |
Jan
|
Feb
(13) |
Mar
(13) |
Apr
(4) |
May
(2) |
Jun
(1) |
Jul
(5) |
Aug
(7) |
Sep
(2) |
Oct
(14) |
Nov
(11) |
Dec
(12) |
| 2009 |
Jan
(30) |
Feb
(4) |
Mar
(16) |
Apr
(9) |
May
(9) |
Jun
(7) |
Jul
(6) |
Aug
(3) |
Sep
(14) |
Oct
(8) |
Nov
(12) |
Dec
(9) |
| 2010 |
Jan
(4) |
Feb
(27) |
Mar
(6) |
Apr
(4) |
May
(3) |
Jun
(13) |
Jul
(6) |
Aug
(15) |
Sep
(15) |
Oct
(12) |
Nov
(11) |
Dec
(9) |
| 2011 |
Jan
(12) |
Feb
(11) |
Mar
|
Apr
(3) |
May
|
Jun
(3) |
Jul
(1) |
Aug
|
Sep
(1) |
Oct
(8) |
Nov
(1) |
Dec
|
| 2012 |
Jan
|
Feb
(10) |
Mar
|
Apr
|
May
|
Jun
|
Jul
(6) |
Aug
(2) |
Sep
(7) |
Oct
(7) |
Nov
|
Dec
(4) |
| 2013 |
Jan
(8) |
Feb
(1) |
Mar
(1) |
Apr
(2) |
May
(3) |
Jun
(3) |
Jul
(16) |
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(1) |
| 2014 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2015 |
Jan
|
Feb
|
Mar
|
Apr
(2) |
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
| 2016 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2017 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
| 2018 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2020 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Skip C. <sco...@gm...> - 2004-11-17 20:48:16
|
On Tue, 16 Nov 2004 09:13:03 +0000, Mark Howard <mh...@ti...> wrote: > Quoting Skip Coon <sco...@gm...>: > > > Hey there, > > > > Well now we are on to something. > > > > aclocal (GNU automake) 1.4-p6 > > > > I did an apt search for aclocal and nothing comes up. Where can I find this. > > automake perhaps? > (I'm not at my ubuntu system at the moment so can't check) > dpkg -S /usr/bin/aclocal should tell you the actual current package. It's > probably possible to have multiple versions installed though, with different > binary names, e.g. aclocal-1.6. > > -- > .""`. Mark Howard > > > : :" : > `. `" http://www.tildemh.com > `- mh...@de... | mh...@ti... > Hey there, I still have not gotten this compiled yet. Could you attache your output of dpkg -l so that I have something to compare with ? scoon |
|
From: Mark H. <mh...@ti...> - 2004-11-17 16:44:55
|
When are you calling this?
From an event handler?
> CustomEvents.addEvent(new Runnable()
> {
> public void run()
> {
> System.out.println("Hello');
> }
> });
>
> The program "Abort" automatically.
This is strange. Please submit a bug to bugzilla, giving the full error message.
--
.""`. Mark Howard
: :" :
`. `" http://www.tildemh.com
`- mh...@de... | mh...@ti...
|
|
From: Nick G. <ar...@ya...> - 2004-11-17 13:04:18
|
I'd like to edit and add some wiki pages on your site. One of them is useful information on how to compile to native binaries using GCJ. However, the wiki registration page gives me an error. The URL http://java-gnome.sourceforge.net/cgi-bin/bin/view/TWiki/TWikiRegistration gives the following error: TWiki Installation Error Template file view.tmpl not found or template directory /home/groups/j/ja/java-gnome/twiki/templates not found. Check the $templateDir variable in TWiki.cfg. Btw, it also cost me a lot of time to find the wiki registration page... maybe it's just me. :) Regards, Nick __________________________________ Do you Yahoo!? The all-new My Yahoo! - Get yours free! http://my.yahoo.com |
|
From: Kevin B. <kb...@ya...> - 2004-11-17 08:14:27
|
Hi there,
I've tried to run the following code in a thread.
CustomEvents.addEvent(new Runnable()
{
public void run()
{
System.out.println("Hello');
}
});
The program "Abort" automatically.
Anybody experienced the same before? What is the reason for that?
Thanks
Kev
---------------------------------
Do you Yahoo!?
Meet the all-new My Yahoo! Try it today! |
|
From: Mark H. <mh...@ti...> - 2004-11-16 09:13:31
|
Quoting Skip Coon <sco...@gm...>: > Hey there, > > Well now we are on to something. > > aclocal (GNU automake) 1.4-p6 > > I did an apt search for aclocal and nothing comes up. Where can I find this. automake perhaps? (I'm not at my ubuntu system at the moment so can't check) dpkg -S /usr/bin/aclocal should tell you the actual current package. It's probably possible to have multiple versions installed though, with different binary names, e.g. aclocal-1.6. -- .""`. Mark Howard : :" : `. `" http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Nicholas R. <ni...@mn...> - 2004-11-16 07:51:51
|
On Mon, 2004-11-15 at 12:49 -0800, Andrew Cowie wrote: > > also, what would be nice would be to > > have the 2.8 ebuilds in stable > > Had to wait until Gnome 2.8 itself was marked stable; further, they'll > probably want to wait a brief while but the positive test reports that > I've had from several people here mean we can get on with it pretty > fast. > gnome 2.8 is now stable (since this weekend, i think) in portage. > [Problem with arch masked in a case like this is that there are too many > idiots out there who run entirely ~arch systems and so the presence of > any 2.9 ebuilds in their portage at this point would cause it to get > brought in - making a mess of the otherwise overall stable gnome 2.8 > present. > > Which is why the next set of Gnome builds tend to be hard masked while > they are being prepared, and stay that way until the whole stack is > ready] > yeah, didn't think about that part. since 2.9 will/does depend on gnome-2.9/gtk-2.5/etc those would have to be brought in somehow. not so nice for idiots like me who run fully ~arched systems. :-) nick |
|
From: Skip C. <sco...@gm...> - 2004-11-15 22:31:08
|
Hey there, Well now we are on to something. aclocal (GNU automake) 1.4-p6 I did an apt search for aclocal and nothing comes up. Where can I find this. scoon On Mon, 15 Nov 2004 19:27:44 +0000, Mark Howard <mh...@ti...> wrote: > Ok, trying another approach, could you possibly check the version of aclocal > you're running (aclocal --version). I have 1.8.5 (also ubuntu) and it seems to > work fine. > > -- > .''`. Mark Howard > > > : :' : > `. `' http://www.tildemh.com > `- mh...@de... | mh...@ti... > |
|
From: Andrew C. <an...@op...> - 2004-11-15 20:49:42
|
On Mon, 2004-15-11 at 12:08 +0100, Nicholas Rahn wrote: > I just emerged java-gnome with your ebuilds. everything seemed to go > fine and the emerge completed without errors. Appreciate the report, thanks. > Just to note that i had to uninstall the previous 2.6 version before i > could install the 2.8. the emerge was blocked by 2.6, but said it was > blocked by 2.8. The way that emerge reports that is confusing, to say the least. The presence of 2.6 was reported as blocking the possibility of 2.8 is how that should read, which was my intention > i had to look in your java-gnome ebuild to see that i (for the reasons that spit out when you merge java-gnome). > As an avid gentoo user, i'd just like to say thanks. when do you think > these will make it into portage? Soon? :) They're making me a dev, so it could be they want me to do the import, but I'll bug karltk and see if he'll import it for me sooner. > also, what would be nice would be to > have the 2.8 ebuilds in stable Had to wait until Gnome 2.8 itself was marked stable; further, they'll probably want to wait a brief while but the positive test reports that I've had from several people here mean we can get on with it pretty fast. > and some 2.9 ebuilds as arch masked. :-) Would depend on 1) there being tarballs of java-gnome 2.9 somewhere and 2) there being a set of ebuilds for the gnome 2.9 libraries to subsequently depend on.=20 [Problem with arch masked in a case like this is that there are too many idiots out there who run entirely ~arch systems and so the presence of any 2.9 ebuilds in their portage at this point would cause it to get brought in - making a mess of the otherwise overall stable gnome 2.8 present. Which is why the next set of Gnome builds tend to be hard masked while they are being prepared, and stay that way until the whole stack is ready] Personally, I'm going to focus on ensuring a 2.10 ebuild is ready for when the time comes. That won't be a problem now, as I've made the jump over the curve to account for the multiplexed libraries. AfC Atlanta =20 --=20 Andrew Frederick Cowie OPERATIONAL DYNAMICS Operations Consultants and Infrastructure Engineers http://www.operationaldynamics.com/ Sydney: +61 2 9977 6866 New York: +1 646 472 5054 Toronto: +1 416 848 6072 London: +44 207 1019201 |
|
From: Mark H. <mh...@ti...> - 2004-11-15 19:27:53
|
Ok, trying another approach, could you possibly check the version of aclocal you're running (aclocal --version). I have 1.8.5 (also ubuntu) and it seems to work fine. -- .''`. Mark Howard : :' : `. `' http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Skip C. <sco...@gm...> - 2004-11-15 16:13:41
|
Hey there, Thanks for the time, but I do not have any older versions of java-gnome installed. I am no slouch with java, so if I could get SCJP then I bet I will be able to resolve this as well. scoon On Mon, 15 Nov 2004 15:47:50 +0000, Mark Howard <mh...@ti...> wrote: > Do you have an older version of java-gnome installed, with files like this: > /usr/.../share/.../libgtk???-java(or java-gnome??)/macros/*.m4 > > If so, do they contain the term AM_PATH_GTK? > > > > -- > .""`. Mark Howard > : :" : > `. `" http://www.tildemh.com > `- mh...@de... | mh...@ti... > |
|
From: Mark H. <mh...@ti...> - 2004-11-15 15:48:15
|
Do you have an older version of java-gnome installed, with files like this: /usr/.../share/.../libgtk???-java(or java-gnome??)/macros/*.m4 If so, do they contain the term AM_PATH_GTK? -- .""`. Mark Howard : :" : `. `" http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Mark H. <mh...@ti...> - 2004-11-15 14:32:14
|
Quoting Skip Coon <sco...@gm...>: > Hey there, > > I do have gnome-devel installed. When I run autogen.sh, I get this: > > -------------------------------------------------------------------------------- > -- Running ./autogen.sh in libgtk-java > > aclocal: configure.in: 42: macro `AM_PATH_GTK' not found in library > ./configure: line 1240: JG_COMMON: command not found Try editing libgtk-java/autogen, adding an exta -I /usr/share/aclocal/ (or wherever libgtk...-dev installes the M4 macro which defines AM_PATH_GTK) -- .""`. Mark Howard : :" : `. `" http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Skip C. <sco...@gm...> - 2004-11-15 14:09:51
|
Hey there, I do have gnome-devel installed. When I run autogen.sh, I get this: -------------------------------------------------------------------------------- -- Running ./autogen.sh in libgtk-java aclocal: configure.in: 42: macro `AM_PATH_GTK' not found in library ./configure: line 1240: JG_COMMON: command not found configure: creating ./config.status config.status: creating Makefile config.status: creating gtk2-java.pc config.status: creating gtk2-java.pc.buildplace config.status: creating doc/Makefile config.status: creating doc/examples/runExample.sh config.status: creating src/Makefile config.status: creating src/java//org/gnu/gtk/Gtk.java config.status: creating src/java/org/gnu/gdk/Gdk.java config.status: creating src/java/org/gnu/glib/GObject.java -------------------------------------------------------------------------------- -- Running ./autogen.sh in libgnome-java using macro dir /home/scoon/java-gnome/src/java-gnome/libgtk-java/macros/ aclocal: configure.in: 40: macro `AM_PATH_GTK' not found in library ./configure: line 1236: syntax error near unexpected token `2.4' ./configure: line 1236: `JG_GTK_JAVA(2.4)' I am not really certain on what to do next. I looked at the configure file and at least around the above complaint seems ok. I am getting java-gnome like this: CVSROOT=":pserver:ano...@an...:/cvs/gnome" cvs -z3 co -r MAINT_2_8 java-gnome Maybe that is my problem, but everything checks out all right. Thanks for any input, scoon On Mon, 15 Nov 2004 13:08:51 +0000, Mark Howard <mh...@ti...> wrote: > Do you have libgtk...-dev packages installed? In fact, it's probably best to > install gnome-devel. > > -- > .""`. Mark Howard > > > : :" : > `. `" http://www.tildemh.com > `- mh...@de... | mh...@ti... > |
|
From: Mark H. <mh...@ti...> - 2004-11-15 13:09:16
|
Do you have libgtk...-dev packages installed? In fact, it's probably best to install gnome-devel. -- .""`. Mark Howard : :" : `. `" http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Nicholas R. <ni...@mn...> - 2004-11-15 11:09:13
|
I just emerged java-gnome with your ebuilds. everything seemed to go fine and the emerge completed without errors. i haven't had a chance to test what was actually installed, but i would assume that it works. Just to note that i had to uninstall the previous 2.6 version before i could install the 2.8. the emerge was blocked by 2.6, but said it was blocked by 2.8. i had to look in you gnome-java ebuild to see that i needed to unmerge the 2.6 first. As an avid gentoo user, i'd just like to say thanks. when do you think these will make it into portage? also, what would be nice would be to have the 2.8 ebuilds in stable and some 2.9 ebuilds as arch masked. :-) nick On Tue, 2004-11-09 at 01:36 +1100, Andrew Cowie wrote: > For those running Gentoo: > > I have created new ebuilds for libgtk-java, libgnome-java, etc and > replaced java-gnome with a meta package which depends on the various > pieces. > > I've tested it on two systems and I think I have the glitches knocked > out of the ebuild. As those on #gnome-java and #java-gnome who have been > listening to me bash away at this thing for the last couple days will > attest, things were massively complicated by the fact that java-gnome is > a fully autoconf'd gnome project. I've got a number of issues that I'll > fix in the various .in files during 2.9 > > If you're interested, I've put the ebuilds I wrote up at: > > http://www.operationaldynamics.com/reference/software/gentoo/ > > We'll get them into Portage shortly. > > If you want to try them, there's a tarball there if you want to put it > into your local overlay. > > ACCEPT_KEYWORDS=~x86 emerge =dev-java/java-gnome-2.8.2 > > [You'll need to have your Gentoo system on GNOME 2.8 (which last I > checked is still masked, though I've been running it for a while now no > problems)] > > I have various programs which I'm working on which quite handily > exercise various bits of these packages, so I'm content that everything > builds, ends up in sensible places, and works. > > AfC > Sydney > -- |
|
From: Skip C. <sco...@gm...> - 2004-11-14 18:19:48
|
Hey all, I am also running ubuntu and am trying to compile the 2.8 from source. I get this err: aclocal: configure.in: 42: macro `AM_PATH_GTK' not found in library ./configure: line 1240: JG_COMMON: command not found aclocal: configure.in: 40: macro `AM_PATH_GTK' not found in library ./configure: line 1236: syntax error near unexpected token `2.4' ./configure: line 1236: `JG_GTK_JAVA(2.4)' Any ideas ? scoon On Mon, 1 Nov 2004 19:52:27 +0000, Mark Howard <mh...@ti...> wrote: > On Sat, Oct 30, 2004 at 07:03:21PM +0100, Sean Coughlan wrote: > > On a clean ubuntu install with all the appropiate packages i had to > > configure like this > > ./configure CFLAGS=-I/usr/lib/gcc/i486-linux/3.4.2/include/ > This should not be needed. We've been talking to the gcj developers about this > -- if the gcj includes cannot be found then your distribution has broken gcj. > > > and do this before a make > > export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig/:/usr/lib/pkgconfig/ > > to get java-gnome up and running. > This is probably needed for installing anything if you have a lot of things > installed in /usr/local/ > > -- > .''`. Mark Howard > : :' : > `. `' http://www.tildemh.com > `- mh...@de... | mh...@ti... > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > java-gnome-developer mailing list > jav...@li... > https://lists.sourceforge.net/lists/listinfo/java-gnome-developer > |
|
From: Robert M. <ro...@ma...> - 2004-11-10 14:24:18
|
On Tue, 2004-11-09 at 22:03, Laurent Martelli wrote: > >>>>> "Andrew" == Andrew Cowie <an...@op...> writes: > > [...] > > Andrew> Now of course, we've pointed out... > > Andrew> (I thought `ant` had something to do with it, but of course, > Andrew> that's just ant doing a single javac invocation. I stand > Andrew> corrected) > > Ant also has tha ability to run java "in process" when you use Sun's > JDK. So the overhead of an invocation is only paid once. > I have been following this discussion, and I think that can be interesting if ant could be used with the Eclipse compiler adapter, as I understand eclipse JDT compiler is smarter than javac detecting which files changed. The people from Tomcat are integrating this compiler with Tomcat 5.5 in order to not depend on javac and to increase JSP compilation speed > Andrew> ...that javac actually has a lot of smarts in it - in fact, > Andrew> you could hand javac nothing but your top level Main.java > Andrew> and it would probably reach out and compile your entire > Andrew> application. > > Andrew> [for a long time I didn't trust that, but then I noticed it > Andrew> doing it for me and covering my ass when I'd in fact made a > Andrew> Makefile mistake, so now I feel better about it!] > > [...] > > Andrew> I've attached the simple diff (and again, in patch form) > Andrew> that shows the makefile differences. As I said, all it does > Andrew> is loose the individual compile steps. > > Andrew> I'm not sure that this is a good idea, overall... > > Andrew> obviously, if you're hacking on the bindings then you don't > Andrew> really want to wait even 4 minutes for a recompile. But If > Andrew> you've already built it, and you're only working on the Java > Andrew> side of things, then 1 minute isn't so bad for a full > Andrew> rebuild [shit, you probably have to make install before you > Andrew> test it anyway] > > Given my optimized includes, "make nativelib" does not win anything > from your patch. So I would personnaly reject that part. Also you can > easily win some time in the "rebuild one java class" case by using $? > instead of $^ in the javac invocation. > > I've ended up making symlinks to my build dir so that I don't have to > "sudo make install" every time I rebuild. > > Andrew> ...but 4 minutes beats the hell out of 18 minutes if you're > Andrew> just an end user building the package for installation on > Andrew> your system. > > Andrew> Food for thought. ________________________________________ Robert Marcano |
|
From: Mark H. <mh...@ti...> - 2004-11-10 09:20:25
|
Did you miss the message I sent last night about this? Perhaps I only sent it to java-gnome-hackers. I have tested a similar solution, except rather than passing in everything, I only pass in modified java source files. This is done by listing all java source files explicitly as prerequisites and then using $? for the ones to compile. I only had time to test this with java compilation though. -- .""`. Mark Howard : :" : `. `" http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Ray A. <ra...@do...> - 2004-11-10 03:14:23
|
Haven't checked it out yet myself (as both my home workstation and server decided that they didn't like their harddisks anymore on the same day... ;( , luckily I use NFS for my home shares on a second server disk...). -----Forwarded Message----- > From: Radu Racaru <rad...@gm...> > To: ja...@gc... > Subject: Eclipse GCJ Builder > Date: Mon, 08 Nov 2004 19:36:52 +0200 > > Hi list, > Just what to announce that I've developed an Eclipse plugin to > facilitate Java development with Eclipse and GCJ. > To quote from plugin's description: > > "GCJ Builder is an Eclipse plugin that integrates GCJ compiler within > Eclipse build infrastructure. Make is used to build GCJ enabled > projects with the plugin's generated makefile." > > The plugin is hosted here (source code included and licensed as EPL 1.0): > http://gcjbuilder.sourceforge.net/ > > Regards, > Radu |
|
From: Laurent M. <la...@ao...> - 2004-11-10 02:03:27
|
>>>>> "Andrew" == Andrew Cowie <an...@op...> writes: [...] Andrew> Now of course, we've pointed out... Andrew> (I thought `ant` had something to do with it, but of course, Andrew> that's just ant doing a single javac invocation. I stand Andrew> corrected) Ant also has tha ability to run java "in process" when you use Sun's JDK. So the overhead of an invocation is only paid once. Andrew> ...that javac actually has a lot of smarts in it - in fact, Andrew> you could hand javac nothing but your top level Main.java Andrew> and it would probably reach out and compile your entire Andrew> application. Andrew> [for a long time I didn't trust that, but then I noticed it Andrew> doing it for me and covering my ass when I'd in fact made a Andrew> Makefile mistake, so now I feel better about it!] [...] Andrew> I've attached the simple diff (and again, in patch form) Andrew> that shows the makefile differences. As I said, all it does Andrew> is loose the individual compile steps. Andrew> I'm not sure that this is a good idea, overall... Andrew> obviously, if you're hacking on the bindings then you don't Andrew> really want to wait even 4 minutes for a recompile. But If Andrew> you've already built it, and you're only working on the Java Andrew> side of things, then 1 minute isn't so bad for a full Andrew> rebuild [shit, you probably have to make install before you Andrew> test it anyway] Given my optimized includes, "make nativelib" does not win anything from your patch. So I would personnaly reject that part. Also you can easily win some time in the "rebuild one java class" case by using $? instead of $^ in the javac invocation. I've ended up making symlinks to my build dir so that I don't have to "sudo make install" every time I rebuild. Andrew> ...but 4 minutes beats the hell out of 18 minutes if you're Andrew> just an end user building the package for installation on Andrew> your system. Andrew> Food for thought. -- Laurent Martelli la...@ao... Java Aspect Components http://www.aopsys.com/ http://jac.objectweb.org |
|
From: Andrew C. <an...@op...> - 2004-11-09 23:11:02
|
[long email, but worth it - please read]
++
Mark rightly points out that there are two modes:
1) build the whole damn thing, from scratch
2) compile just what I have updated, because I'm hacking on it.
I have some thoughts on the subject:
++
Having grown up in the grand old Unix tradition, I have always
understood Makefiles, and the rational for such a program: don't want to
recompile everything just because of one little change. So, you'd have
blah: blah.o foo.o bar.o bling.o
ld -o blah blah.o foo.o bar.o bling.o
blah.o: blah.c
cc -o blah.o -c blah.c
foo.o: foo.c
cc -o foo.o -c foo.c
...
and so on. Of course, if you were compiling from scratch, it'd do
everything which was good, and if you made a single source change, you'd
only get a single compile and a single link.
For the longest time, I always thought that this was good.
++
In recent years, I've started to wonder about this. For example, 99% of
the time, when I compile something, it's not software I'm developing but
some package I'm building to install on my system. And so I sit there
and watch endless cc commands (each with two pages of -I flags) scroll
by.
The economics of CPU, memory, and the nature of the language have
changed. This is why languages like python (and Java, for that matter)
are feasible - the overhead of an interpreter (+ VM, etc) is not
terribly significant given today's available processing power.
It's not so obvious with gcc, but with javac, oh man.
Because it runs in a Java VM, javac takes a fair while to fire up. Java
source files themselves tend to be very small. There is a lot of
repetitive lookups to check references in all the imports. So you do
this long process for each of thousands of .java files, and you're there
a while.
++
Now of course, we've pointed out...
(I thought `ant` had something to do with it, but of course, that's just
ant doing a single javac invocation. I stand corrected)
...that javac actually has a lot of smarts in it - in fact, you could
hand javac nothing but your top level Main.java and it would probably
reach out and compile your entire application.
[for a long time I didn't trust that, but then I noticed it doing it for
me and covering my ass when I'd in fact made a Makefile mistake, so now
I feel better about it!]
So I started thinking about what would happen if reworked my own
projects' make files to just do one single javac invocation (to build
the .class files) and one single gcj invocation (straight to target .so
or executable) with all my .java files as input - regardless of whether
or not they are new.
Of course, I was horrified at the thought of getting it to compile
things it didn't have to, but
<speculation>
the effort of compiling, and the effort of looking up the
classes for import is about the same regardless, and the gain
would be not doing n invocations for n files
</speculation>
So I tried it.
Wow. Fast fast fast! Much faster than waiting for the individual gcj
and/or javac calls to scroll by. So fast, in fact, that a rebuild after
a trivial change to one file was still faster on the monolithic model
than on the one-by-one model.
++
So in view of this week's conversation, I decided to try it with
libgtk-java.
With my system otherwise idle, libgtk-java takes this long to build
(this is `make` in a clean (but configured) top level directory):
real 17m40.494s
user 14m49.584s
sys 1m8.221s
[This box is a dual P3; my laptop takes over 40 minutes :(]
I then did some quick hacks to libgtk-java/src/Makefile to replace the
individual invocations of javac (.class files) and gcj (.o files) and
gcc (JNI .o files) with single steps
[ie a single javac with all the source files listed, a single gcj
command going straight to the .so with all the .java source files
listed, and a single gcc command going straight to the jni .so file with
all the .c files listed]
real 4m37.743s
user 4m11.783s
sys 0m22.127s
Wow!
The amazing thing is that the slowest part of all is the jni C
compiling! Take that out, and the time to compile the java source (both
javac and gcj) was only
real 1m5.500s
user 1m1.663s
sys 0m1.988s
that's 65 seconds!
Wow!
++
I've attached the simple diff (and again, in patch form) that shows the
makefile differences. As I said, all it does is loose the individual
compile steps.
I'm not sure that this is a good idea, overall...
obviously, if you're hacking on the bindings then you don't really want
to wait even 4 minutes for a recompile. But If you've already built it,
and you're only working on the Java side of things, then 1 minute isn't
so bad for a full rebuild [shit, you probably have to make install
before you test it anyway]
...but 4 minutes beats the hell out of 18 minutes if you're just an end
user building the package for installation on your system.
Food for thought.
AfC
--
Andrew Frederick Cowie
OPERATIONAL DYNAMICS
Operations Consultants and Infrastructure Engineers
Australia: +61 2 9977 6866 North America: +1 646 472 5054
http://www.operationaldynamics.com/
|
|
From: Laurent M. <la...@ao...> - 2004-11-09 22:03:08
|
>>>>> "Mark" == Mark Howard <mh...@ti...> writes: Mark> Looks good. Does it make any difference to the stripped size Mark> of the so files? I'm guessing not. But the speed improvements Mark> are certainly worth it. Mark> On my system, strip *.so.* changes the main gtk so file from Mark> 27MB to 626K The stripped size remains the same, as expected. -- Laurent Martelli la...@ao... Java Aspect Components http://www.aopsys.com/ http://jac.objectweb.org |
|
From: Mark H. <mh...@ti...> - 2004-11-09 20:11:45
|
Looks good. Does it make any difference to the stripped size of the so files? I'm guessing not. But the speed improvements are certainly worth it. On my system, strip *.so.* changes the main gtk so file from 27MB to 626K -- .''`. Mark Howard : :' : `. `' http://www.tildemh.com `- mh...@de... | mh...@ti... |
|
From: Laurent M. <la...@ao...> - 2004-11-09 13:53:44
|
Dear java-gnome-developer,
Would you like to reduce the size of your generated shared object file
by as mush as 90% ? If you do then read on, because I want to share a
technique I've just discovered. And it's *free* of all charge!!!!
First, you should know that including <gtk/gtk/h> when you are only
using a GtkButton is waste of resource. If you actually only use
GtkButton, just include <gtk/gtkbutton.h>, the size of the object file
will be drastically smaller. And compilation will be faster. I will
post a patch to clean includes in the BTS.
New versions of gcc use the DWARF2 debugging format by default instead
of stabs. DWARF2 is theoretically more compact than stabs, but by
default debugging info for all the symbols encountered during
compilation is produced, even for the symbols which are not
used. Also, the gcc implementation of DWARF2 does not by default
eliminate duplicate debugging info. So if you include <gtk/gtk.h> in
each every c file, you end up with debugging info for all the symbols
in <gtk/gtk.h> multiplied by the number of c files. The use of the
-feliminate-dwarf2-dups flag can fix that.
I also discovered that gcc version 3.4 is about 30% faster than
version 3.3.
Here are few numbers comparing before and after (use f gcc 3.4 with
-feliminate-dwarf2-dups and only include the necessary header files):
before after
compile time 2m20sec 1m5sec
.so size 27M 2.4M
du -hs libgtk-java/src/jni/ 60M 17M
Dear sir, I hope you are now convinced and will chose to follow my proven
method.
Sincerely yours,
Laurent Martelli
--
Laurent Martelli
la...@ao... Java Aspect Components
http://www.aopsys.com/ http://jac.objectweb.org
|
|
From: Nicholas R. <ni...@mn...> - 2004-11-09 10:16:03
|
Yes, i've had this same problem on my stable gentoo box. My arch-masked
("~x86") gentoo machine did not have this problem. And like some of the
other problems i've had, i couldn't figure out a "proper" solution.
If i remember correctly, this is what was happening: the "api" directory
was getting deleted and then rebuilt. however, the "resources"
directory was not getting rebuilt and probably shouldn't have been
deleted in the first place because there was some image file in there.
i tried serveral things, but what i think worked was that i modified the
Makefiles, commenting the removal of the api directory in the install
section. i did this during the emerge, while the compilation was
taking place and after the configuration step.
hope this helps.
nick
On Tue, 2004-11-09 at 09:50 +0000, co...@ye... wrote:
> Hi
>
> Ive tryed this morning to emerge java-gnome on Gentoo and aftter about 1 hour it
> falls over with this error:-
>
> cannot stat the 'api/resources/*'
>
> Anyone had the same problems on Gentoo or even has anyone got it installed okay
>
>
> Many Thanks
>
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> java-gnome-developer mailing list
> jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-gnome-developer
--
|
59 messages has been excluded from this view by a project administrator.
