java-gnome-developer Mailing List for The java-gnome language bindings project (Page 113)
Brought to you by:
afcowie
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: Andrew <ac...@on...> - 2003-11-12 18:59:50
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Luca De Rugeriis wrote: | Il mer, 2003-11-12 alle 15:07, Jeffrey Morgan ha scritto: | |>Thanks for the effort. I will try to apply the |>patch sometime today. How close is this application |>to being functional? |> |>-Jeff |> | | I think it's near to be functional, however I really miss the point at | which the work was left... have to investigate further. | The gui is complete although it needs to be higyfied specially the new | crontab entry window. | There is some implementation of the logic used to create a new crontab | entry, but it is limited to System.out outputs. | It seems like the methods used to append new entries in the main window | TreeView are broken or left unimplemented... | My main problem to continue developing on this, is to really figure out | what the original authors wanted to do in every methods, and at what | point these were left uncompleted. | I've also seen the original cromagnon and it seems to be limited to | print to stdout the new crontab entry. | | However don't take this statements as if I see a negative future for it, | I'm a newbie, so someone with more skills could bring it to | functionality easily ;) | | Luca. Is this available to be checked out of cvs, or do I need to download a new copy of it? I'd be interested in poking about in it perhaps. - -- [A n d r e w] - ------|------ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQE/soSR2AKtqeH2FeQRAt1kAJ9oXFaBMuQLRsNT8hABOcwfbMuZGQCfXFzl Bn2mg6XY07R2Jgj4owo1Cnw= =sVNr -----END PGP SIGNATURE----- |
From: Luca De R. <pie...@li...> - 2003-11-12 18:42:46
|
Il mer, 2003-11-12 alle 15:07, Jeffrey Morgan ha scritto: > Thanks for the effort. I will try to apply the > patch sometime today. How close is this application > to being functional? > > -Jeff > I think it's near to be functional, however I really miss the point at which the work was left... have to investigate further. The gui is complete although it needs to be higyfied specially the new crontab entry window. There is some implementation of the logic used to create a new crontab entry, but it is limited to System.out outputs. It seems like the methods used to append new entries in the main window TreeView are broken or left unimplemented... My main problem to continue developing on this, is to really figure out what the original authors wanted to do in every methods, and at what point these were left uncompleted. I've also seen the original cromagnon and it seems to be limited to print to stdout the new crontab entry. However don't take this statements as if I see a negative future for it, I'm a newbie, so someone with more skills could bring it to functionality easily ;) Luca. -- Luca De Rugeriis <pie...@li...> |
From: Luca De R. <pie...@li...> - 2003-11-12 18:26:48
|
Il mer, 2003-11-12 alle 15:10, Jeffrey Morgan ha scritto: > > 1. If you use the AppBar, all is fine except that, if you also > > install menu hints, instead of the right hint you will end up > > with a lot of libgnomeui information strings sticked in the > > appbar, but only when the mouse hovers the main menu > > items i.e. > > file, edit,...etc. (If you want I can explain myself > > better with > > a couple of pics). > > This seems to be working now. I believe this was addressed with > the reworking of String arrays. Please try again and let me know > if you are still having problems. > > > 2. It seems there is a couple of visible bug in the TestGtk > > application: when you click on the toolbar button and on the > > handle box button the program crash with signal 11 with a long > > trace. I think these traces are easily reproducible, but again > > if you want I will post them. > > The fix for this problem is in cvs. > > > 3. In the TreeExample app, it cannot render a pixbuf on the tree. > > It complaints with this message: > > This item will requrie further work. I will look into it over > the next few days. > > > And finally, altought maybe it isn't a bug report, can we > > have chance to > > see the MessageDialog class finished? > > This class is already complete. I just added an example of > its' usage to TestGTK. Check out cvs. Wonderful ;) I'll checkout cvs and I'll let you know. Luca. -- Luca De Rugeriis <pie...@li...> |
From: Elias M. <el...@al...> - 2003-11-12 17:34:09
|
Tiago Cogumbreiro wrote: > On Qua, 2003-11-12 at 15:50, Elias Martenson wrote: > >>I find this to be a huge issue. I'd much rather place a dependency on a >>VM and then be able to distribute the same PRM, DEB or whatever archive >>to everybody. The user would still never have to bother about the VM. > > Well, when i think of desktop users i think of lazy ones, and IMO > installing the sun's java vm is more difficult then loading a shared lib > (libgcj.so), yet this is a problem for distribution only. Don't forget > that on linux usually java is not installed by default, it's always > thought of a big extra. Yet applications such as PostgreSQL and 3 billion useless(:-)) KDE libraries are usually installed on a standard Linux distribution and it's not considered being a "big" extra. In my mind, what needs to be done is to lobby for the inclusing of VM's on the linux distributions. apt-based distros such as debian and fedors shouldn't be a problem at all technically, since a Java-GNOME app can include the dependencies on it and it will get downloaded automatically. >>I don't think one problem (allergic users) should be solved by >>introducing another (platform dependency and neutering the Java platform). > > I don't think it's yet another dependency, JVM is switched by a shared > lib, you lose one you gain another. No, the big difference is that every single Java-GNOME app can be distributed in a platform _indepdendent_ form. This is an anormous advantage in my opinion. >>>also there's the Swing >>>factor which gives a bit of a bad impression in all java programs that >>>they are slugish. >> >>Solved with Java-GNOME. That's why we're here, isn't it? :-) > > My point was that swing applications being slugish makes some users > think that _all_ java aplications are slugish. Of course. The funny thing is that the same people happily run various graphical applications written in for example Python and Perl without any complaints about speed. This is an education issue really. Prove a point, preferably using a killer app. That way you can tell them: "did you know that this app actually runs in the Java VM? It's not slow, is it?" >>>This way when they open up our java-gnome program >>>they'll think it's an ordinary C/Gnome program. >> >>My Java-GNOME test applications look exactly like any GNOME application >>written in C, and it runs in the Sun VM. Exactly what is the problem? > > java -Djava.library.path=/usr/lib/jni -classpath > /usr/lib/java/gnome.jar:/usr/lib/java/gtk.jar org.foo.MyCoolClass This would look like so, on a properly installed system: java -jar myapp.jar > I know that shell scripts soulve this very gracefully, but again, when > natively compiled you just don't have to think about yet another > problem, and that's basicly the difference between running a C app and a > Java app. A shellscript wrapping the above command is prefectly adequate for me. The screllscript is platform independent too, in the context of Java-GNOME. >>Sorry for sounding a bit harsh. I usually try to avoid that on public >>forums but this is an issue that I care a lot about. > > I don't think that you were harsh :) And i care alot about this topic so > i find this discussion very productive ;) Thank you. :-) Elias |
From: Elias M. <el...@al...> - 2003-11-12 17:21:49
|
fer...@lo... wrote: > I have used both Apache FOP and JasperReports. The second is much speedier and > easier. iReports is an open source project that provides a visual designer for > JasperReports, but it uses Swing. A GTK+ version of iReports would be a nice start. What is iReports? Also, a report designer for a database tool woul dhave slightly different needs than a standard report designer (such as the one for jasper reports). > Using the same glade XML format has advantages: ability to work on form > rendering/execution before the visual designer is working; use of a > stabilished, well-known format; option to import/export forms to other > development tools; use the libglade infrastructure already provided bu java-gnome. Yes, but the needs of a form designer and a GUI builder like glade are very different, don't you shink? One would have to shoehorn the functionality into the glade format, unless it is extremely flexible. Elias |
From: <fer...@lo...> - 2003-11-12 17:09:49
|
Hi there, > > The biggest problem I could forsee (with that one and this) is the > > generating of reports. I think they could be done (and thought of apache > > fo, but I don't know a whole lot about that technology. > > I've used it in production and I like it a lot. The only complaint I > have is that it doesn't handle characters outside 8859-1 very well. > Cyrillic didn't work at all, the euro symbol got the wrong width etc... > Could be simple font issues on my end though. I have used both Apache FOP and JasperReports. The second is much speedier and easier. iReports is an open source project that provides a visual designer for JasperReports, but it uses Swing. A GTK+ version of iReports would be a nice start. > > The other thing that pokes my mind is form desiging, and I thought some > > use of glade could be done, but not quite sure as I haven't given it a > > whole lot of time. > > I would expect that the end user should be able to design his own forms. > Are you saying that he should design them using glade? Or that the forms > designed should genrade glade XML data? The first case won't work I > think, and the second would effectively mean that we'd be redesigning > the glade tool in Java. Using the same glade XML format has advantages: ability to work on form rendering/execution before the visual designer is working; use of a stabilished, well-known format; option to import/export forms to other development tools; use the libglade infrastructure already provided bu java-gnome. []s, Fernando Lozano |
From: Mark H. <mh...@ca...> - 2003-11-12 17:03:37
|
On Wed, Nov 12, 2003 at 09:10:39AM -0500, Jeffrey Morgan wrote: > > 3. In the TreeExample app, it cannot render a pixbuf on the tree. > > It complaints with this message: > > This item will requrie further work. I will look into it over > the next few days. I've just been looking at it but not found a solution. It is very strange - very similar code is working perfectly in my applications. -- .''`. Mark Howard : :' : `. `' http://www.tildemh.com `- mh...@de... | mh...@ti... | mh...@ca... |
From: Elias M. <el...@al...> - 2003-11-12 16:50:34
|
Tiago Cogumbreiro wrote: > Sun's JVM is superior to GCJ and Kaffe, but nowadays these Open options > are way more mature and present as a viable choice, such fact didn't > happened in a few months ago. GCJ has a great advantage: being able to > compile classes to native code is great to package/destribute and your > program (in a single platform) I find this to be a huge issue. I'd much rather place a dependency on a VM and then be able to distribute the same PRM, DEB or whatever archive to everybody. The user would still never have to bother about the VM. The lack of properly working dynamic class loading and reflection is also a problem for me. > and don't forget that there are alot of > Java alergics out there (especially in the Linux community) that don't > want anything to do with java classes and jars, I don't think one problem (allergic users) should be solved by introducing another (platform dependency and neutering the Java platform). > also there's the Swing > factor which gives a bit of a bad impression in all java programs that > they are slugish. Solved with Java-GNOME. That's why we're here, isn't it? :-) > This way when they open up our java-gnome program > they'll think it's an ordinary C/Gnome program. My Java-GNOME test applications look exactly like any GNOME application written in C, and it runs in the Sun VM. Exactly what is the problem? Sorry for sounding a bit harsh. I usually try to avoid that on public forums but this is an issue that I care a lot about. Elias |
From: Tiago C. <cog...@li...> - 2003-11-12 16:40:22
|
On Qua, 2003-11-12 at 15:18, Elias Martenson wrote: > Tiago Cogumbreiro wrote: > > > Yes, i agree that the 2D functionalities of the upcoming version of > > sun's jvm is not something relevant to java-gnome, more important are > > system responsiveness and a good JNI performance. IMO we need a JVM that > > has a low memory footprint and good execution speed, i didn't see any > > recent becnhmarks but GCJ and Kaffe seem to be a good option in either > > cases. > > I find Sun Java to be superiour in every respect. It's the fastest, it > (obviously) has the lastest stuff first, and it it only _very_ slightly > larger than kaffe (if the numbers posted are accurate). > > GCJ has it's own slew os problems, inability to do reflection being the > largest one. > > Elias > Sun's JVM is superior to GCJ and Kaffe, but nowadays these Open options are way more mature and present as a viable choice, such fact didn't happened in a few months ago. GCJ has a great advantage: being able to compile classes to native code is great to package/destribute and your program (in a single platform) and don't forget that there are alot of Java alergics out there (especially in the Linux community) that don't want anything to do with java classes and jars, also there's the Swing factor which gives a bit of a bad impression in all java programs that they are slugish. This way when they open up our java-gnome program they'll think it's an ordinary C/Gnome program. Btw, i don't know if anyone tried this, but it seems very easy to compile Java-Gnome programs to native code with this tool: http://www.freestyler-toolkit.org/ |
From: Jeffrey M. <Jef...@Br...> - 2003-11-12 16:01:39
|
Can you provide some details and code to reproduce this problem? Thanks -Jeff > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Mark Howard wrote: > | On Tue, Nov 11, 2003 at 08:40:44PM +0000, Andrew wrote: > | > |>What do we do in this situation? (I haven't researched if > it is a known > |>bug)? > |>Do we log it somewhere, or raise it to this list? > | > | > | In general, bugs should be submitted to the sourceforge bug > tracker for > | the java-gnome project. Any changes there get sent here > anyway, so we > | can all help out if somebody knows the answer. Unfortunately I can't > | help with this one - I'm sure you'll get a more helpful > response from > | somebody else soon though. > | > | > |>org.gnu.glade.GladeXMLException: Unknown GtkToolbar child property: > |>border_width > | > | > > > Hey, > This error is coming from a native call > private native void glade_xml_new_from_buffer(byte[] buffer, String > rootname) throws GladeXMLException; > > I did a search on the net, but I don't know much about glade and C > libraries in gnome (only about java type stuff :P) > > I did just remove the gtk toolbar that I was using and used a normal > horizontal button panel and it works just the same. > > > - -- > [A n d r e w] > - ------|------ > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.3 (GNU/Linux) > Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org > > iD8DBQE/sVHH2AKtqeH2FeQRAiXKAKCUZRmg6rOhIvaD8lMFKqZxWh9UQwCcCEJz > uKMXFPEQ2ptSOul9dJo/uhU= > =lz4H > -----END PGP SIGNATURE----- > > > > ------------------------------------------------------- > This SF.Net email sponsored by: ApacheCon 2003, > 16-19 November in Las Vegas. Learn firsthand the latest > developments in Apache, PHP, Perl, XML, Java, MySQL, > WebDAV, and more! http://www.apachecon.com/ > _______________________________________________ > java-gnome-developer mailing list > jav...@li... > https://lists.sourceforge.net/lists/listinfo/java-gnome-developer > |
From: Elias M. <el...@al...> - 2003-11-12 15:58:20
|
Tiago Cogumbreiro wrote: > If that is really going to happen then it is a good sign but you must > agree that until that happens the desktop application that run on a JVM > in a windows machine is faster then on a linux machine. In this context, we don't care much do we? Java-GNOME doesn't use Java2D (for good or worse). Elias |
From: Jeffrey M. <Jef...@Br...> - 2003-11-12 14:11:05
|
> 1. If you use the AppBar, all is fine except that, if you also > install menu hints, instead of the right hint you will end up > with a lot of libgnomeui information strings sticked in the > appbar, but only when the mouse hovers the main menu > items i.e. > file, edit,...etc. (If you want I can explain myself > better with > a couple of pics). This seems to be working now. I believe this was addressed with the reworking of String arrays. Please try again and let me know if you are still having problems. > 2. It seems there is a couple of visible bug in the TestGtk > application: when you click on the toolbar button and on the > handle box button the program crash with signal 11 with a long > trace. I think these traces are easily reproducible, but again > if you want I will post them. The fix for this problem is in cvs. > 3. In the TreeExample app, it cannot render a pixbuf on the tree. > It complaints with this message: This item will requrie further work. I will look into it over the next few days. > And finally, altought maybe it isn't a bug report, can we > have chance to > see the MessageDialog class finished? This class is already complete. I just added an example of its' usage to TestGTK. Check out cvs. |
From: Jeffrey M. <Jef...@Br...> - 2003-11-12 14:07:36
|
Thanks for the effort. I will try to apply the patch sometime today. How close is this application to being functional? -Jeff > Here it is, the app is ported to java-gnome-0.8. > It is at the same level of functionality it was before, only it uses > gtk2. > I would like to continue developing this app but I haven't fully > understood the crontab entry creation logic, cause it seems > it was left > half implemented... > I would very glad if someone has any suggestion on which are the major > missing methods to implement, or wants to help to reach basic > functionality ;) > > Luca. > -- > Luca De Rugeriis <pie...@li...> > |
From: Tiago C. <cog...@li...> - 2003-11-12 13:48:25
|
On Qua, 2003-11-12 at 11:27, Clemens Eisserer wrote: > Hi Tiago! >=20 > Your benchmarks ment nice, but they are not really useful. >=20 > 1.) What does the *9, *3 mean for memory-consumption? > Have you ever heard of threads? Threads are under Linux(o.k., thats not= =20 > really correct) processes that share stuff between. That means Sun=B4s = JVM=20 > does need 7,5% of your memory, not 67,5%. Thanks for clearing this up, i did know that threads under sun JVM appear as processes, i did not know about the shared memory issue, but still Kaffe is the winner in this field, it has 6,5% while sun's JVM has a 7.5% consumption >=20 > 2.) Your startup-results seem that you=B4ve started the JVMs already, n= ot=20 > really fair, because JVMs like Sun=B4s one need to load a lot of shared= =20 > stuff. It would be fair to measure with and without haveing started it=20 > before and listing both values. Don't the other JVM's load shared libs too? How do i control this? =20 >=20 > 3.) You forgot a really important point: Execution speed. > GIJ wasnt designed to be fast, it runs java-code only in interpreted=20 > mode, because it was ment to be a way for GCJ to load dynamically=20 > classes. Kaffe has two fifferent JITs. > IBMs JVM is the fastest one, when only measuring execution speed. -> 10= 0% > After IBM theres Sun. -> 90% > after Sun theres kaffe ->25-45% > after kaffe theres GIJ -> 5-10% >=20 > However these percentages are not 100% correct and depend heavily on=20 > what you do with the VMs. >=20 > However IBMs VM is based on Suns classes, that means Kaffe is the=20 > fastest OPenSource-JVM laying arround. Suns&IBMs JVMs are as fast or=20 > faster than GCJ, which compiles java statically. Even GCJ developers=20 > commented that Sun-JVM is impressive! I did not forgot, i just didn't measured this. Yet from what I've seen it seems that GCJ is faster then Kaffe: http://klomp.org/mark/free-vm-benchmarks/ http://www.shudo.net/jit/perf/ >=20 > 4.) > > better choice then Sun's java. This makes you wonder how good is Sun > > investing in the Linux platform. > Sun is currently rewriting their whole Java2D/x11 rendering pipeline. > They will use OpenGL starting from 1.5 for drawing things. I=B4ve spoke= n=20 > with a Java2D-enginier and he told me that the results are impressive. > If Linux would not matter, they wouldnt spend so much time improving it= =20 > on Linux/Desktop. If that is really going to happen then it is a good sign but you must agree that until that happens the desktop application that run on a JVM in a windows machine is faster then on a linux machine. Finally i just wanted to make clear that these benchmarks are nothing "official" nor it ever tried to be. I was benchmarking load speed and memory consumption of an idle application, ignoring the fact that i've made the processes error, and i think i can believe in those 2 benchmarks. Kaffe is still the most light JVM around Sun's JVM is of course the fastest, the other JVM options are still, IMO, very good. |
From: J. M. B. <jmb...@co...> - 2003-11-12 13:34:13
|
Clemens Eisserer wrote: > 1.) What does the *9, *3 mean for memory-consumption? > Have you ever heard of threads? Threads are under Linux(o.k., thats > not really correct) processes that share stuff between. That means > Sun´s JVM does need 7,5% of your memory, not 67,5%. > Actually, threads aren't a problem under Linux if you are using the NPTL thread library and JDK1.4.2. > 4.) > > better choice then Sun's java. This makes you wonder how good is Sun > > investing in the Linux platform. > Sun is currently rewriting their whole Java2D/x11 rendering pipeline. > They will use OpenGL starting from 1.5 for drawing things. I´ve spoken > with a Java2D-enginier and he told me that the results are impressive. > If Linux would not matter, they wouldnt spend so much time improving > it on Linux/Desktop. I'm glad to see that they are moving to OpenGL for this. I hadn't heard this information before. Does anyone know if GCJ will be updated to include the language changes coming with 1.5 (Java ?3?) |
From: Clemens E. <lin...@we...> - 2003-11-12 12:28:36
|
Hi Tiago! Your benchmarks ment nice, but they are not really useful. 1.) What does the *9, *3 mean for memory-consumption? Have you ever heard of threads? Threads are under Linux(o.k., thats not really correct) processes that share stuff between. That means Sun´s JVM does need 7,5% of your memory, not 67,5%. 2.) Your startup-results seem that you´ve started the JVMs already, not really fair, because JVMs like Sun´s one need to load a lot of shared stuff. It would be fair to measure with and without haveing started it before and listing both values. 3.) You forgot a really important point: Execution speed. GIJ wasnt designed to be fast, it runs java-code only in interpreted mode, because it was ment to be a way for GCJ to load dynamically classes. Kaffe has two fifferent JITs. IBMs JVM is the fastest one, when only measuring execution speed. -> 100% After IBM theres Sun. -> 90% after Sun theres kaffe ->25-45% after kaffe theres GIJ -> 5-10% However these percentages are not 100% correct and depend heavily on what you do with the VMs. However IBMs VM is based on Suns classes, that means Kaffe is the fastest OPenSource-JVM laying arround. Suns&IBMs JVMs are as fast or faster than GCJ, which compiles java statically. Even GCJ developers commented that Sun-JVM is impressive! 4.) > better choice then Sun's java. This makes you wonder how good is Sun > investing in the Linux platform. Sun is currently rewriting their whole Java2D/x11 rendering pipeline. They will use OpenGL starting from 1.5 for drawing things. I´ve spoken with a Java2D-enginier and he told me that the results are impressive. If Linux would not matter, they wouldnt spend so much time improving it on Linux/Desktop. lg Clemens |
From: Tiago C. <cog...@li...> - 2003-11-12 01:11:59
|
Hi all, i've been playing around with open-source JVMs and decided to perform a benchmark on those who could run my CDMirror project, as mentioned before in the mailing list the great loss is SableVM, unfornatly they are with some JNI problems, and until that solves Java-gnome won't be able to run on it. I did 2 simple benchmarks, startup time and memory consumption, Kaffe was the great winner with impressive results, makes me realize how heavy sun's VM really is on memory. Check it out: http://s1x.homelinux.net/content.php?id=9 |
From: Fernando L. <fer...@lo...> - 2003-11-12 00:18:03
|
Hi Andrew, > | I think most "killer apps" for a desktop were already made. I'd love to > | see Java-Gnome popular for developing information systems and > | line-of-businness apps, in place of Swing. This is what Linux and Java > | miss today: something to fill the role of VB, Delphi or PowerBuilder. > | Rich UIs, easy but powerfull language, with performance. Java is the > | language that can do that, and GTK+ is the toolkit. Something > | integrating Eclipse and Glade would fill the gap > Actually, the thing that I think is missing (or I can't find) is a > substitute for MS Access. I think something could be done that uses > HSQLDB (for example) and a front end is tacked on to it to provide a > schema builder, forms, reports. I have seen some around, but they are > really poor quality and usually cost $. OpenOffice has some head start on this. I'm amazed at how few people know OpenOffice can connect to databases, create tables, run SQL statements and visually design queryes just like Access. It doesn't have forms (well, you can draw then on Writer or Calc). But this is a nice example of what could be developed using GCJ to become the "killer" app. Use Rhino or BeanShell for the scripting language, Glade for forms, and HSQLDB to build an Access-Like low profile development tool. :-) []s, Fernando Lozano |
From: Elias M. <el...@al...> - 2003-11-11 21:30:18
|
tis 2003-11-11 klockan 19.57 skrev Andrew: > The biggest problem I could forsee (with that one and this) is the > generating of reports. I think they could be done (and thought of apache > fo, but I don't know a whole lot about that technology. I've used it in production and I like it a lot. The only complaint I have is that it doesn't handle characters outside 8859-1 very well. Cyrillic didn't work at all, the euro symbol got the wrong width etc... Could be simple font issues on my end though. There are some other tools though, that are simpler to use. One of them is being used by a colleague of mine to print reports for a system we're building. I'll check with him tomorrow to see how it works. He seems to like it a lot though. > The other thing that pokes my mind is form desiging, and I thought some > use of glade could be done, but not quite sure as I haven't given it a > whole lot of time. I would expect that the end user should be able to design his own forms. Are you saying that he should design them using glade? Or that the forms designed should genrade glade XML data? The first case won't work I think, and the second would effectively mean that we'd be redesigning the glade tool in Java. Generally, the form designer shouldn't be very hard to write. All you have to do is let the user place the form element manually on a canvas without any layout applied. Right? I never tried such a thing using GTK+ but all in other toolkits I used it was a piece of cake. > But the rest of it would be reasonable ok to do I think, well, I can't > think why it would be too hard. Don't you agree? I mean, surely if > people can write CD Burning software, a nice database tool would be > straight forward!! Definately. And I'd like to reiterate the enormous advangate of being able to simple drop these JDBC driver in. Look at the way DbVisualizer handles JDBC drivers. It's really cool. :-) Elias |
From: Elias M. <el...@al...> - 2003-11-11 21:23:33
|
tis 2003-11-11 klockan 19.40 skrev Luca De Rugeriis: > Il mar, 2003-11-11 alle 13:34, Elias Martenson ha scritto: > > Andrew - Dev wrote: > > > > > Actually, the thing that I think is missing (or I can't find) is a > > > substitute for MS Access. I think something could be done that uses > > > HSQLDB (for example) and a front end is tacked on to it to provide a > > > schema builder, forms, reports. I have seen some around, but they are > > > really poor quality and usually cost $. > > > > That's a brilliant suggestion really. I consider JDBC to be superiour to > > the non-java alternatives. ODBC isn't much of a standard on Unix, and > > Gnome-DB isn't working properly yet and still doesn't support many > > databases. Every single database out there has working JDBC drivers > > available today. > > > > This is the best suggestion so far for a killer app that leverages Java > > technologies. Don't you all think? > I think this suggestion it's even better than the former ones, but I'm > wondering who is going (and got the skills) to implement such a nice > project? I'd be interested in helping out. Definately. Not only forms support but also general database manager, similar to DbVisualizer (http://www.minq.se/products/dbvis/) would be cool. I'm not the person to start such a project by myself though. Elias |
From: Tiago C. <cog...@li...> - 2003-11-11 21:20:56
|
On Ter, 2003-11-11 at 19:05, Luca De Rugeriis wrote: > Il mar, 2003-11-11 alle 20:28, Dan Pilone ha scritto: > > > That was one thing that came into my mind too. It is the main windows > > > app thats missing from linux (although some would say it encourages bad > > > practice, so shouldn't be ported). I think this could be put in the same > > > > My suggestion would be a good CD burning application. I considered > > porting k3b to JavaGnome but haven't gotten very far. I don't think it > > would be too difficult to port an existing C++ app to JavaGnome, Gnome would > > get a really nice CD/DVD authoring tool, and once it's in Java I think > > development could progress much more quickly than the C++ version. I don't > > want to cause trouble or get people all worked up by forking k3b, but I > > think it would be a great "proof-of-concept" without running into problems > > of either writing an application no one uses or an application getting > > started but having people lose interest. Just my $.02. -- Dan > Tiago Cogumbreiro is writing something like that he said: > I am going to create a sourceforge project and put the files there, i > have a somewhat outdated version of cdmirror in: > http://s1x.homelinux.net/downloads/cdmirror.tar.bz2 > > Finally i want to underline that the initial concept of the project is > not to create a masterizing program, the first concept is to write/read > images from CDs (it already includes tools to convert images like .bin > to .isos, and raw audio to .wav). To create a fullfeatured remastering > program natively will be hard work, you have to create a library for ISO > managing, thus should be developed in a future stage. > > I hope we can help on developing this. > > Luca. I've just submitted CDMirror project to Sourceforge.net, so i'm just waiting for theyre review, as soon as i get a reply i'll anounce it so we can start some work. One of my main interests/objectives in this project is to provide an easy library for readung and recording data to CDs yet this process is still in a very early stage so one thing we could do is to create the ability to chose from the native drivers or to use cdrdao/cdrecord to perform the reading and writing operations. I've been playing around with gij and gcj to try to run cdmirror. All the classes in the main lib were compiled with gcj correctly but gij didn't succeded running the entire app. I need to check how kaffe handles with it too. Tiago Cogumbreiro |
From: Andrew <ac...@on...> - 2003-11-11 21:10:52
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mark Howard wrote: | On Tue, Nov 11, 2003 at 08:40:44PM +0000, Andrew wrote: | |>What do we do in this situation? (I haven't researched if it is a known |>bug)? |>Do we log it somewhere, or raise it to this list? | | | In general, bugs should be submitted to the sourceforge bug tracker for | the java-gnome project. Any changes there get sent here anyway, so we | can all help out if somebody knows the answer. Unfortunately I can't | help with this one - I'm sure you'll get a more helpful response from | somebody else soon though. | | |>org.gnu.glade.GladeXMLException: Unknown GtkToolbar child property: |>border_width | | Hey, This error is coming from a native call private native void glade_xml_new_from_buffer(byte[] buffer, String rootname) throws GladeXMLException; I did a search on the net, but I don't know much about glade and C libraries in gnome (only about java type stuff :P) I did just remove the gtk toolbar that I was using and used a normal horizontal button panel and it works just the same. - -- [A n d r e w] - ------|------ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQE/sVHH2AKtqeH2FeQRAiXKAKCUZRmg6rOhIvaD8lMFKqZxWh9UQwCcCEJz uKMXFPEQ2ptSOul9dJo/uhU= =lz4H -----END PGP SIGNATURE----- |
From: Mark H. <mh...@ca...> - 2003-11-11 20:52:58
|
On Tue, Nov 11, 2003 at 08:40:44PM +0000, Andrew wrote: > What do we do in this situation? (I haven't researched if it is a known > bug)? > Do we log it somewhere, or raise it to this list? In general, bugs should be submitted to the sourceforge bug tracker for the java-gnome project. Any changes there get sent here anyway, so we can all help out if somebody knows the answer. Unfortunately I can't help with this one - I'm sure you'll get a more helpful response from somebody else soon though. > org.gnu.glade.GladeXMLException: Unknown GtkToolbar child property: > border_width -- .''`. Mark Howard : :' : `. `' http://www.tildemh.com `- mh...@de... | mh...@ti... | mh...@ca... |
From: Andrew <ac...@on...> - 2003-11-11 20:34:39
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 What do we do in this situation? (I haven't researched if it is a known bug)? Do we log it somewhere, or raise it to this list? org.gnu.glade.GladeXMLException: Unknown GtkToolbar child property: border_width - -- [A n d r e w] - ------|------ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQE/sUlM2AKtqeH2FeQRAsZQAJ9W0Ok6zIqcN8cRvHif5zIrU7pI5ACdG8R0 hmxIXb/I2sPrc/CAM5McJ34= =wl6Y -----END PGP SIGNATURE----- |
From: Luca De R. <pie...@li...> - 2003-11-11 20:07:39
|
Il mar, 2003-11-11 alle 20:28, Dan Pilone ha scritto: > > That was one thing that came into my mind too. It is the main windows > > app thats missing from linux (although some would say it encourages bad > > practice, so shouldn't be ported). I think this could be put in the same > > My suggestion would be a good CD burning application. I considered > porting k3b to JavaGnome but haven't gotten very far. I don't think it > would be too difficult to port an existing C++ app to JavaGnome, Gnome would > get a really nice CD/DVD authoring tool, and once it's in Java I think > development could progress much more quickly than the C++ version. I don't > want to cause trouble or get people all worked up by forking k3b, but I > think it would be a great "proof-of-concept" without running into problems > of either writing an application no one uses or an application getting > started but having people lose interest. Just my $.02. -- Dan Tiago Cogumbreiro is writing something like that he said: I am going to create a sourceforge project and put the files there, i have a somewhat outdated version of cdmirror in: http://s1x.homelinux.net/downloads/cdmirror.tar.bz2 Finally i want to underline that the initial concept of the project is not to create a masterizing program, the first concept is to write/read images from CDs (it already includes tools to convert images like .bin to .isos, and raw audio to .wav). To create a fullfeatured remastering program natively will be hard work, you have to create a library for ISO managing, thus should be developed in a future stage. I hope we can help on developing this. Luca. -- Luca De Rugeriis <pie...@li...> |