From pierre.marechal at scilab.org Sun Nov 2 23:04:45 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Sun, 02 Nov 2008 23:04:45 +0100 Subject: [Scilab-Dev] Gitview not updated? In-Reply-To: <490A257F.70205@free.fr> References: <4908B901.6000002@free.fr> <4909BAA5.5080704@scilab.org> <490A257F.70205@free.fr> Message-ID: <490E23FD.1070005@scilab.org> Fran?ois Vogel a ?crit : > Pierre MARECHAL said on 30/10/2008 14:46: >>> Also, since I'm here, another question: is there any way in gitview >>> to show only the commits pertaining to a given branch? Currently >>> everything is mixed up and it's not easy to find what applies to >>> what branch. >> Yes, links are in the "heads" section : >> http://gitweb.scilab.org/?p=scilab;a=shortlog;h=refs/heads/5.0 > > OK, found, thanks. > > Btw, did you perhaps miss the other question I had yesterday: > > How can I see the commits in a remote branch, say origin/master > without fetching or pulling them first? > > Francois Hi Fran?ois, I don't think it's possible because it's not the philosophy of a distributed revision control : Every Git working directory is a full-fledged repository with complete history and full revision tracking capabilities. So why ask the remote repository ? "git fecth" can't break your local build : This command downloads objects and refs from the origin repository but not impact on your working directory. Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 1B - Bureau 008 Email : pierre.marechal at inria.fr ============================================== From fvogelnew1 at free.fr Sun Nov 2 23:07:40 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Sun, 02 Nov 2008 23:07:40 +0100 Subject: [Scilab-Dev] Gitview not updated? In-Reply-To: <490E23FD.1070005@scilab.org> References: <4908B901.6000002@free.fr> <4909BAA5.5080704@scilab.org> <490A257F.70205@free.fr> <490E23FD.1070005@scilab.org> Message-ID: <490E24AC.7060704@free.fr> Pierre MARECHAL said on 02/11/2008 23:04: > Fran?ois Vogel a ?crit : >> Pierre MARECHAL said on 30/10/2008 14:46: >>>> Also, since I'm here, another question: is there any way in gitview >>>> to show only the commits pertaining to a given branch? Currently >>>> everything is mixed up and it's not easy to find what applies to >>>> what branch. >>> Yes, links are in the "heads" section : >>> http://gitweb.scilab.org/?p=scilab;a=shortlog;h=refs/heads/5.0 >> >> OK, found, thanks. >> >> Btw, did you perhaps miss the other question I had yesterday: >> >> How can I see the commits in a remote branch, say origin/master >> without fetching or pulling them first? >> >> Francois > Hi Fran?ois, > > I don't think it's possible because it's not the philosophy of a > distributed revision control : Every Git working directory is a > full-fledged repository with complete history and full revision tracking > capabilities. So why ask the remote repository ? > "git fecth" can't break your local build : This command downloads > objects and refs from the origin repository but not impact on your > working directory. Understood. Then, one I have fetched, how can I see the diff between the newly fetched commits and my local sandbox? I could not find this in the git help, I must have looked at the wrong place. Francois From pierre.marechal at scilab.org Mon Nov 3 08:19:37 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Mon, 03 Nov 2008 08:19:37 +0100 Subject: [Scilab-Dev] Gitview not updated? In-Reply-To: <490E24AC.7060704@free.fr> References: <4908B901.6000002@free.fr> <4909BAA5.5080704@scilab.org> <490A257F.70205@free.fr> <490E23FD.1070005@scilab.org> <490E24AC.7060704@free.fr> Message-ID: <490EA609.20100@scilab.org> Fran?ois Vogel a ?crit : > Pierre MARECHAL said on 02/11/2008 23:04: >> Fran?ois Vogel a ?crit : >>> Pierre MARECHAL said on 30/10/2008 14:46: >>>>> Also, since I'm here, another question: is there any way in >>>>> gitview to show only the commits pertaining to a given branch? >>>>> Currently everything is mixed up and it's not easy to find what >>>>> applies to what branch. >>>> Yes, links are in the "heads" section : >>>> http://gitweb.scilab.org/?p=scilab;a=shortlog;h=refs/heads/5.0 >>> >>> OK, found, thanks. >>> >>> Btw, did you perhaps miss the other question I had yesterday: >>> >>> How can I see the commits in a remote branch, say origin/master >>> without fetching or pulling them first? >>> >>> Francois >> Hi Fran?ois, >> >> I don't think it's possible because it's not the philosophy of a >> distributed revision control : Every Git working directory is a >> full-fledged repository with complete history and full revision >> tracking capabilities. So why ask the remote repository ? >> "git fecth" can't break your local build : This command downloads >> objects and refs from the origin repository but not impact on your >> working directory. > > Understood. Then, one I have fetched, how can I see the diff between > the newly fetched commits and my local sandbox? I could not find this > in the git help, I must have looked at the wrong place. > > Francois "git log HEAD" shows commit logs in your working tree and "git log refs/remotes/origin/HEAD" shows commit logs in the origin repository. git log HEAD > /tmp/local ; git log refs/remotes/origin/HEAD > /tmp/origin ; diff -w /tmp/origin /tmp/local Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 1B - Bureau 008 Email : pierre.marechal at scilab.org ============================================== From sylvestre.ledru at scilab.org Mon Nov 3 09:55:47 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Mon, 03 Nov 2008 09:55:47 +0100 Subject: [Scilab-Dev] Gitview not updated? In-Reply-To: <490EA609.20100@scilab.org> References: <4908B901.6000002@free.fr> <4909BAA5.5080704@scilab.org> <490A257F.70205@free.fr> <490E23FD.1070005@scilab.org> <490E24AC.7060704@free.fr> <490EA609.20100@scilab.org> Message-ID: <1225702547.29623.156.camel@korcula.inria.fr> > > Understood. Then, one I have fetched, how can I see the diff between > > the newly fetched commits and my local sandbox? I could not find this > > in the git help, I must have looked at the wrong place. > > > > Francois > "git log HEAD" shows commit logs in your working tree and "git log > refs/remotes/origin/HEAD" shows commit logs in the origin repository. > > git log HEAD > /tmp/local ; git log refs/remotes/origin/HEAD > > /tmp/origin ; diff -w /tmp/origin /tmp/local I guess this: git log HEAD..origin is doing the same thing. Sylvestre From mwe012008 at gmx.net Tue Nov 4 11:05:40 2008 From: mwe012008 at gmx.net (Michael Welle) Date: Tue, 04 Nov 2008 11:05:40 +0100 Subject: Matrix -> String representation-> Matrix Message-ID: <878wrz956z.fsf@luisa.c0t0d0s0.de> Hi, I need to convert matrices containing various types to a string representation. Later I need to convert the string representation back to the matrices with their original types. I can do this in a C module (that I need anyway) or with SciLab built ins. Maybe string( m ) and type( m ) is the way to go for the first conversation step? Any wise words on this? I need this for an ODBC interface that supports among other operations sqlLoad(), sqlSave() and sqlClear(), that should behave like their SciLab counterparts. Regards hmw -- biff4emacsen - A biff-like tool for (X)Emacs http://www.c0t0d0s0.de/biff4emacsen/biff4emacsen.html Flood - Your friendly network packet generator http://www.c0t0d0s0.de/flood/flood.html From enrico.segre at weizmann.ac.il Tue Nov 4 11:15:03 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Tue, 04 Nov 2008 12:15:03 +0200 Subject: [Scilab-Dev] Matrix -> String representation-> Matrix In-Reply-To: <878wrz956z.fsf@luisa.c0t0d0s0.de> References: <878wrz956z.fsf@luisa.c0t0d0s0.de> Message-ID: <491020A7.8050009@weizmann.ac.il> Michael Welle wrote: > Hi, > > I need to convert matrices containing various types to a string > representation. Later I need to convert the string representation back > to the matrices with their original types. I can do this in a C module > (that I need anyway) or with SciLab built ins. Maybe string( m ) and > type( m ) is the way to go for the first conversation step? Any wise > words on this? > > pedestrian way is sci2exp(), but it suffered of limitations (at the moment I don't recall which). Have a look at FormatStringsForWatch(), there we thought at several ad hoc solutions for odd types. Enrico From shiqi.yu at gmail.com Thu Nov 6 04:45:37 2008 From: shiqi.yu at gmail.com (Shiqi Yu) Date: Thu, 6 Nov 2008 11:45:37 +0800 Subject: Image showing in Scilab 5 Message-ID: <296f12ef0811051945v225447f1vcda14547c0389c2a@mail.gmail.com> Dear All, I had a nice met with Dr. Serge Steer and Dr. Claude Gomez on the Scilab Workshop in Liuzhou, China. Serge Steer said Scilab will have a imshow function to show an image. I remember Sylvestre Ledru also said that. Now my toolbox SIVP uses Tcl/Tk for showing image, and it's very inefficient. I'm not familar with Java. It will take me a long time to add showing image function into the Scilab Figure. If Scilab Team can do that, it will be more efficient. Several years ago, I was told that Scilab 5 will solve the problem. I have waited it for many years, ;-) and many SIVP/SIP users also expect it. Will Scilab Team put a high priority on it? Regards, Shiqi Yu -------------- next part -------------- An HTML attachment was scrubbed... URL: From enrico.segre at weizmann.ac.il Thu Nov 6 17:02:47 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Thu, 06 Nov 2008 18:02:47 +0200 Subject: [Scilab-Dev] Git / Scilab In-Reply-To: <1224250230.18336.330.camel@korcula.inria.fr> References: <1224250230.18336.330.camel@korcula.inria.fr> Message-ID: <49131527.3050207@weizmann.ac.il> I sit behind a firewall and git clone git://git.scilab.org/scilab just times out. Is there an option to open the access on a less efficient http: protocol, or something the like, on your side? Otherwise, proposed workaraounds? Enrico From pierre.marechal at scilab.org Fri Nov 7 10:28:33 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Fri, 07 Nov 2008 10:28:33 +0100 Subject: Scilab 5.0.3 is released ! Message-ID: <49140A41.90303@scilab.org> Hello, The Scilab Team is pleased to announce the release of Scilab 5.0.3. Misc information about this version: http://www.scilab.org/download/index_download.php?page=5.0.3 Main changes between Scilab 4 & Scilab 5 : http://www.scilab.org/changes_5/ Changes between Scilab 5.0.1 & Scilab 5.0.2 : http://www.scilab.org/download/index_download.php?page=CHANGES_5.0.2 Changes between Scilab 5.0.2 & Scilab 5.0.3 : http://www.scilab.org/download/index_download.php?page=CHANGES_5.0.3 Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 23 - Bureau 24 Email : pierre.marechal at scilab.org ============================================== From sebastienbarre at free.fr Fri Nov 7 13:33:36 2008 From: sebastienbarre at free.fr (Sebastien Barre) Date: Fri, 07 Nov 2008 13:33:36 +0100 Subject: call C function in Scilab 5.0.2 Message-ID: <1226061216.491435a0303fb@imp.free.fr> Dear Scilab users / Developpers, I am currently working on the development of an interface between measurement devices and the great Scilab machine. The idea is to be able to command the devices connected to the computer directly for the scilab command Line. In this aim, I use the Scilab c_link function to link c function of a DLL, then the Scilab call function. It looks like this: if (~c_link('QueryDevices')) funcload = ['QueryDevices']; link('SCILabIF.dll', funcload, 'c'); end [NbrOfDevices] = call('QueryDevices', 'out', [1,1], 1, 'i'); if (NbrOfDevices <= 0) //error handling no devices found rc = -1; end My problem is: The command: [Name, Channels] = call('GetChannelName', 0 , 1, 'i', 'out', [32, 1], 2, 'c', [1,1], 3, 'i'); works properly with version 4.1.2 of Scilab, but with Version 5.0.2 I get the message Warning !!! Scilab has found a critical error (EXCEPTION_ACCESS_VIOLATION) with "call" function. Save your data and restart Scilab. Warning: stack problem..., cleared. Does anybody know why? and how could I solve it. Thanks in advance, S?bastien -- S?bastien Barr? Sinus Messtechnik GmbH Foepplstrasse 13 D-04347 Leipzig Germany Tel: +49 341 244 29 28 From allan.cornet at scilab.org Fri Nov 7 13:41:56 2008 From: allan.cornet at scilab.org (Allan CORNET) Date: Fri, 7 Nov 2008 13:41:56 +0100 Subject: [Scilab-Dev] call C function in Scilab 5.0.2 In-Reply-To: <1226061216.491435a0303fb@imp.free.fr> References: <1226061216.491435a0303fb@imp.free.fr> Message-ID: <001f01c940d6$385e1a40$a91a4ec0$@cornet@scilab.org> Hi, Can you attach your script and your dll ? Thanks you Allan CORNET -----Message d'origine----- De?: Sebastien Barre [mailto:sebastienbarre at free.fr] Envoy??: vendredi 7 novembre 2008 13:34 ??: dev at lists.scilab.org Objet?: [Scilab-Dev] call C function in Scilab 5.0.2 Dear Scilab users / Developpers, I am currently working on the development of an interface between measurement devices and the great Scilab machine. The idea is to be able to command the devices connected to the computer directly for the scilab command Line. In this aim, I use the Scilab c_link function to link c function of a DLL, then the Scilab call function. It looks like this: if (~c_link('QueryDevices')) funcload = ['QueryDevices']; link('SCILabIF.dll', funcload, 'c'); end [NbrOfDevices] = call('QueryDevices', 'out', [1,1], 1, 'i'); if (NbrOfDevices <= 0) //error handling no devices found rc = -1; end My problem is: The command: [Name, Channels] = call('GetChannelName', 0 , 1, 'i', 'out', [32, 1], 2, 'c', [1,1], 3, 'i'); works properly with version 4.1.2 of Scilab, but with Version 5.0.2 I get the message Warning !!! Scilab has found a critical error (EXCEPTION_ACCESS_VIOLATION) with "call" function. Save your data and restart Scilab. Warning: stack problem..., cleared. Does anybody know why? and how could I solve it. Thanks in advance, S?bastien -- S?bastien Barr? Sinus Messtechnik GmbH Foepplstrasse 13 D-04347 Leipzig Germany Tel: +49 341 244 29 28 From allan.cornet at scilab.org Wed Nov 12 17:13:09 2008 From: allan.cornet at scilab.org (Allan CORNET) Date: Wed, 12 Nov 2008 17:13:09 +0100 Subject: Windows build prerequisites for scilab (GIT master) Message-ID: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> Hello, We have updated PVM for Windows on GIT master. You need to update your Windows build prerequisites. http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequiremen ts/Windows/ or http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequiremen ts/Windows_x64/ Best Regards Allan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fvogelnew1 at free.fr Wed Nov 12 19:59:33 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Wed, 12 Nov 2008 19:59:33 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> Message-ID: <491B2795.1070400@free.fr> Hi, Is it normal that the web query from for recent SVN changes returns nothing for, say, the last week or day? http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform This worked before git migration. Since I was told prerequirements will continue to be managed by SVN, so I was expecting this form to continue to work. Francois Allan CORNET said on 12/11/2008 17:13: > Hello, > > > > We have updated PVM for Windows on GIT master. > > You need to update your Windows build prerequisites. > > > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ > > or > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ > > > > Best Regards > > > > Allan > > > > > From ycollet at freesurf.fr Wed Nov 12 21:09:53 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Wed, 12 Nov 2008 21:09:53 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B2795.1070400@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> Message-ID: <491B3811.5040200@freesurf.fr> Hello, I was about to get the linux 64 bits prerequisite and get just an empty directory. So, it looks like a global problem ... YC Fran?ois Vogel a ?crit : > Hi, > > Is it normal that the web query from for recent SVN changes returns > nothing for, say, the last week or day? > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform > > This worked before git migration. Since I was told prerequirements > will continue to be managed by SVN, so I was expecting this form to > continue to work. > > Francois > > > Allan CORNET said on 12/11/2008 17:13: >> Hello, >> >> >> >> We have updated PVM for Windows on GIT master. >> >> You need to update your Windows build prerequisites. >> >> >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ >> >> >> or >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ >> >> >> >> >> Best Regards >> >> >> >> Allan >> >> >> >> >> > From pierre.marechal at scilab.org Wed Nov 12 21:41:22 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Wed, 12 Nov 2008 21:41:22 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B2795.1070400@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> Message-ID: <491B3F72.3050003@scilab.org> Hi Fran?ois, Automatic update of Viewvc database is now fixed. Sorry for the trouble. Pierre [1] http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform Fran?ois Vogel a ?crit : > Hi, > > Is it normal that the web query from for recent SVN changes returns > nothing for, say, the last week or day? > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform > > This worked before git migration. Since I was told prerequirements > will continue to be managed by SVN, so I was expecting this form to > continue to work. > > Francois > > > Allan CORNET said on 12/11/2008 17:13: >> Hello, >> >> >> >> We have updated PVM for Windows on GIT master. >> >> You need to update your Windows build prerequisites. >> >> >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ >> >> >> or >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ >> >> >> >> >> Best Regards >> >> >> >> Allan >> >> >> >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From fvogelnew1 at free.fr Thu Nov 13 01:21:02 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 13 Nov 2008 01:21:02 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B3F72.3050003@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> Message-ID: <491B72EE.3090505@free.fr> Thank you for your efforts. However I'm sorry, but it still does not work for me. I can get no answer from the server, neither with the link provided by Allan below nor with the link you marked [1] in your message. Francois Pierre MARECHAL said on 12/11/2008 21:41: > Hi Fran?ois, > > Automatic update of Viewvc database is now fixed. > Sorry for the trouble. > > Pierre > > [1] http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform > > > Fran?ois Vogel a ?crit : >> Hi, >> >> Is it normal that the web query from for recent SVN changes returns >> nothing for, say, the last week or day? >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform >> >> This worked before git migration. Since I was told prerequirements >> will continue to be managed by SVN, so I was expecting this form to >> continue to work. >> >> Francois >> >> >> Allan CORNET said on 12/11/2008 17:13: >>> Hello, >>> >>> >>> >>> We have updated PVM for Windows on GIT master. >>> >>> You need to update your Windows build prerequisites. >>> >>> >>> >>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ >>> >>> >>> or >>> >>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ >>> >>> >>> >>> >>> Best Regards >>> >>> >>> >>> Allan From fvogelnew1 at free.fr Thu Nov 13 01:35:15 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 13 Nov 2008 01:35:15 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B72EE.3090505@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> Message-ID: <491B7643.6050102@free.fr> Also, I can't connect anymore to the compilation chain results web site. What should I do? Thanks, Francois Fran?ois Vogel said on 13/11/2008 01:21: > Thank you for your efforts. However I'm sorry, but it still does not > work for me. > > I can get no answer from the server, neither with the link provided by > Allan below nor with the link you marked [1] in your message. > > Francois > > > Pierre MARECHAL said on 12/11/2008 21:41: >> Hi Fran?ois, >> >> Automatic update of Viewvc database is now >> fixed. >> Sorry for the trouble. >> >> Pierre >> >> [1] http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform >> >> >> Fran?ois Vogel a ?crit : >>> Hi, >>> >>> Is it normal that the web query from for recent SVN changes returns >>> nothing for, say, the last week or day? >>> >>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform >>> >>> This worked before git migration. Since I was told prerequirements >>> will continue to be managed by SVN, so I was expecting this form to >>> continue to work. >>> >>> Francois >>> >>> >>> Allan CORNET said on 12/11/2008 17:13: >>>> Hello, >>>> >>>> >>>> >>>> We have updated PVM for Windows on GIT master. >>>> >>>> You need to update your Windows build prerequisites. >>>> >>>> >>>> >>>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ >>>> >>>> >>>> or >>>> >>>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ >>>> >>>> >>>> >>>> >>>> Best Regards >>>> >>>> >>>> >>>> Allan > From fvogelnew1 at free.fr Thu Nov 13 08:22:45 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 13 Nov 2008 08:22:45 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> Message-ID: <491BD5C5.4010501@free.fr> Hi Allan, Is it supposed to be OK for the f2c solution as well? I see some problems. Thanks, Francois Allan CORNET said on 12/11/2008 17:13: > Hello, > > > > We have updated PVM for Windows on GIT master. > > You need to update your Windows build prerequisites. > > > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ > > or > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ > > > > Best Regards > > > > Allan > > > > > From allan.cornet at scilab.org Thu Nov 13 08:50:14 2008 From: allan.cornet at scilab.org (Allan CORNET) Date: Thu, 13 Nov 2008 08:50:14 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491BD5C5.4010501@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491BD5C5.4010501@free.fr> Message-ID: <000e01c94564$767f4c30$637de490$@cornet@scilab.org> Hi, Fixed for f2c solution (dependency on localization) 1>sci_pvm_config.obj : error LNK2019: unresolved external symbol _localeToUTF referenced in function _intspvm_config Thanks Allan -----Message d'origine----- De?: Fran?ois Vogel [mailto:fvogelnew1 at free.fr] Envoy??: jeudi 13 novembre 2008 08:23 ??: dev at lists.scilab.org Objet?: Re: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) Hi Allan, Is it supposed to be OK for the f2c solution as well? I see some problems. Thanks, Francois From ycollet at freesurf.fr Thu Nov 13 09:25:44 2008 From: ycollet at freesurf.fr (ycollet at freesurf.fr) Date: Thu, 13 Nov 2008 09:25:44 +0100 Subject: Problem with windows nightly build Message-ID: <20081113092544.99114cot0fcuzog0@webmail-html.freesurf.fr> Hello, The last windows nightly build (1226506170) has a problem. When I start Scilex or WScilex, the following error message is printed: C:\Program Files\scilab-master-1226506170\bin>Scilex scilex can't launch scilab. Error code : 5 Acc?s refus?. YC From pierre.marechal at scilab.org Thu Nov 13 11:04:46 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Thu, 13 Nov 2008 11:04:46 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B3811.5040200@freesurf.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3811.5040200@freesurf.fr> Message-ID: <491BFBBE.5050005@scilab.org> Hi Yann, The linux 64 bits prerequisites are now available. [1] http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 Pierre Collette Yann a ?crit : > Hello, > > I was about to get the linux 64 bits prerequisite and get just an > empty directory. > So, it looks like a global problem ... > > YC > > Fran?ois Vogel a ?crit : >> Hi, >> >> Is it normal that the web query from for recent SVN changes returns >> nothing for, say, the last week or day? >> >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/?view=queryform >> >> This worked before git migration. Since I was told prerequirements >> will continue to be managed by SVN, so I was expecting this form to >> continue to work. >> >> Francois >> >> >> Allan CORNET said on 12/11/2008 17:13: >>> Hello, >>> >>> >>> >>> We have updated PVM for Windows on GIT master. >>> >>> You need to update your Windows build prerequisites. >>> >>> >>> >>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows/ >>> >>> >>> or >>> >>> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/Windows_x64/ >>> >>> >>> >>> >>> Best Regards >>> >>> >>> >>> Allan >>> >>> >>> >>> >>> >> > -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 23 - Bureau 24 Email : pierre.marechal at scilab.org ============================================== From enrico.segre at weizmann.ac.il Thu Nov 13 11:26:16 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Thu, 13 Nov 2008 12:26:16 +0200 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491BFBBE.5050005@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3811.5040200@freesurf.fr> <491BFBBE.5050005@scilab.org> Message-ID: <491C00C8.2020305@weizmann.ac.il> Pierre MARECHAL wrote: > > The linux 64 bits prerequisites are now available. > > [1] > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 > > any plan of putting there a libmatio.so as well? I hadn't been able to build it from sources on my system. And btw, what about git access behind firewall? Enrico From pierre.marechal at scilab.org Thu Nov 13 11:40:52 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Thu, 13 Nov 2008 11:40:52 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C00C8.2020305@weizmann.ac.il> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3811.5040200@freesurf.fr> <491BFBBE.5050005@scilab.org> <491C00C8.2020305@weizmann.ac.il> Message-ID: <491C0434.7070800@scilab.org> Hi Enrico : Enrico Segre a ?crit : > Pierre MARECHAL wrote: >> >> The linux 64 bits prerequisites are now available. >> >> [1] >> http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 >> >> > > any plan of putting there a libmatio.so as well? I hadn't been able to > build it from sources on my system. Done : http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 > And btw, what about git access behind firewall? Do you have your ssh port opened ? > Enrico Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 23 - Bureau 24 Email : pierre.marechal at scilab.org ============================================== From enrico.segre at weizmann.ac.il Thu Nov 13 12:17:14 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Thu, 13 Nov 2008 13:17:14 +0200 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C0434.7070800@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3811.5040200@freesurf.fr> <491BFBBE.5050005@scilab.org> <491C00C8.2020305@weizmann.ac.il> <491C0434.7070800@scilab.org> Message-ID: <491C0CBA.7020500@weizmann.ac.il> Pierre MARECHAL wrote: >> any plan of putting there a libmatio.so as well? I hadn't been able >> to build it from sources on my system. > Done : > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 > Thanks, I see. Copied the relevant files, but configure still gives me checking for Tk_BindEvent in -ltk8.5... yes checking for MATIO... configure: error: Package requirements (matio >= 1.3.3) were not met: No package 'matio' found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables MATIO_CFLAGS and MATIO_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. will sort out the problem later. No time now. >> And btw, what about git access behind firewall? > Do you have your ssh port opened ? Possibly not. I can ssh to internal hosts, so the block is not on my workstation, but it doesn't occur me of an external host to test the firewall with. Enrico From fvogelnew1 at free.fr Thu Nov 13 20:22:44 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 13 Nov 2008 20:22:44 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491B7643.6050102@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> Message-ID: <491C7E84.6060800@free.fr> > Fran?ois Vogel said on 13/11/2008 01:21: >> Thank you for your efforts. However I'm sorry, but it still does not >> work for me. >> >> I can get no answer from the server, neither with the link provided by >> Allan below nor with the link you marked [1] in your message. Works for me now, thanks. Not sure why it didn't yesterday. >> Also, I can't connect anymore to the compilation chain results web site. http://compilation_chain.scilab.org/index.php Can't connect. Did I forget my password maybe? Thanks for a hint. Francois From ycollette at free.fr Thu Nov 13 22:12:03 2008 From: ycollette at free.fr (Collette Yann) Date: Thu, 13 Nov 2008 22:12:03 +0100 Subject: linux_x64 problem Message-ID: <491C9823.3010304@free.fr> Hello, I think there is a problem with the jdk shipped in the linux_x64 requirements. It is not possible to use this jdk to build a java application. if I do a ls -l in the java/jdk1.... directory, there is a circular symlink: drwxr-xr-x 3 collette collette 4096 2008-11-13 18:39 bin/ -rw-r--r-- 1 collette collette 3767 2008-11-13 18:39 COPYRIGHT drwxr-xr-x 8 collette collette 4096 2008-11-13 18:36 db/ drwxr-xr-x 10 collette collette 4096 2008-11-13 18:36 demo/ drwxr-xr-x 4 collette collette 4096 2008-11-13 18:35 include/ *lrwxrwxrwx 1 collette collette 11 2008-11-13 18:39 jdk1.6.0_10 -> jdk1.6.0_10* drwxr-xr-x 6 collette collette 4096 2008-11-13 18:35 jre/ drwxr-xr-x 4 collette collette 4096 2008-11-13 18:39 lib/ -rw-r--r-- 1 collette collette 17058 2008-11-13 18:39 LICENSE drwxr-xr-x 5 collette collette 4096 2008-11-13 18:39 man/ -rw-r--r-- 1 collette collette 28599 2008-11-13 18:39 README.html -rw-r--r-- 1 collette collette 25390 2008-11-13 18:39 README_ja.html -rw-r--r-- 1 collette collette 20768 2008-11-13 18:39 README_zh_CN.html -rw-r--r-- 1 collette collette 5334 2008-11-13 18:39 register.html -rw-r--r-- 1 collette collette 5766 2008-11-13 18:39 register_ja.html -rw-r--r-- 1 collette collette 4944 2008-11-13 18:39 register_zh_CN.html drwxr-xr-x 9 collette collette 4096 2008-11-13 18:35 sample/ -rw-r--r-- 1 collette collette 19009557 2008-11-13 18:39 src.zip -rw-r--r-- 1 collette collette 228347 2008-11-13 18:39 THIRDPARTYLICENSEREADME.txt YC From ycollet at freesurf.fr Thu Nov 13 22:41:53 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Thu, 13 Nov 2008 22:41:53 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C7E84.6060800@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> Message-ID: <491C9F21.4020608@freesurf.fr> Hello, By the way, for your matio problem: have you added a --enable-shared option to the configure script of matio ? By default, matio builds a static library which is not detected by scilab (it detects only shared library). >>> Also, I can't connect anymore to the compilation chain results web >>> site. > > http://compilation_chain.scilab.org/index.php > > Can't connect. Did I forget my password maybe? With a good passwaord, it works. > Thanks for a hint. > > Francois > YC From sylvestre.ledru at inria.fr Thu Nov 13 22:46:53 2008 From: sylvestre.ledru at inria.fr (Sylvestre Ledru) Date: Thu, 13 Nov 2008 22:46:53 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C9F21.4020608@freesurf.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> Message-ID: <1226612813.3544.27.camel@zlarin> Le jeudi 13 novembre 2008 ? 22:41 +0100, Collette Yann a ?crit : > Hello, > > By the way, for your matio problem: have you added a --enable-shared > option to the configure script of matio ? > By default, matio builds a static library which is not detected by > scilab (it detects only shared library). Btw, that is a problem that we should report to the dev of matio. Sylvestre From fvogelnew1 at free.fr Thu Nov 13 22:59:34 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 13 Nov 2008 22:59:34 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C9F21.4020608@freesurf.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> Message-ID: <491CA346.9010001@free.fr> Collette Yann said on 13/11/2008 22:41: >> http://compilation_chain.scilab.org/index.php >> >> Can't connect. Did I forget my password maybe? > > With a good passwaord, it works. The page says password is the subversion password. I can't make it work with my subversion password. I could login some time ago, can't say exactly when. I guess I will need help from Pierre. Francois From ycollet at freesurf.fr Fri Nov 14 14:45:01 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Fri, 14 Nov 2008 14:45:01 +0100 Subject: Git questions Message-ID: <491D80DD.7010001@freesurf.fr> Hello, I have 2 questions related to the use of git: I have cloned the git repository. I have made some changes, for example in the file modules/optim/help/en_US/optim.xml. I want to put the changes into the scilab master repository. Do I must enter the following commands ? git branch master refs/remotes/origin/master git push To get the YaSp branch, what kind of git commands must I enter ? git branch YaSp refs/remotes/origin/YaSp git pull YC From sylvestre.ledru at scilab.org Fri Nov 14 17:09:15 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Fri, 14 Nov 2008 17:09:15 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C0CBA.7020500@weizmann.ac.il> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3811.5040200@freesurf.fr> <491BFBBE.5050005@scilab.org> <491C00C8.2020305@weizmann.ac.il> <491C0434.7070800@scilab.org> <491C0CBA.7020500@weizmann.ac.il> Message-ID: <1226678955.29623.643.camel@korcula.inria.fr> Le jeudi 13 novembre 2008 ? 13:17 +0200, Enrico Segre a ?crit : > Pierre MARECHAL wrote: > >> any plan of putting there a libmatio.so as well? I hadn't been able > >> to build it from sources on my system. > > Done : > > http://viewvc.scilab.org/bin/cgi/viewvc.cgi/trunk/Dev-Tools/SE/Prerequirements/linux_x64 > > > > Thanks, I see. Copied the relevant files, but configure still gives me > > checking for Tk_BindEvent in -ltk8.5... yes > checking for MATIO... configure: error: Package requirements (matio >= > 1.3.3) were not met: > > No package 'matio' found > > Consider adjusting the PKG_CONFIG_PATH environment variable if you > installed software in a non-standard prefix. > > Alternatively, you may set the environment variables MATIO_CFLAGS > and MATIO_LIBS to avoid the need to call pkg-config. > See the pkg-config man page for more details. If you try: export PKG_CONFIG_PATH=path/to/matio/with/the/.pc/file before the configure, it should work. If it doesn't, please let me know. Sylvestre From pierre.marechal at scilab.org Fri Nov 14 17:34:46 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Fri, 14 Nov 2008 17:34:46 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491CA346.9010001@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> <491CA346.9010001@free.fr> Message-ID: <491DA8A6.4040505@scilab.org> Fran?ois Vogel a ?crit : > Collette Yann said on 13/11/2008 22:41: >>> http://compilation_chain.scilab.org/index.php >>> >>> Can't connect. Did I forget my password maybe? >> >> With a good passwaord, it works. > > The page says password is the subversion password. I can't make it > work with my subversion password. I could login some time ago, can't > say exactly when. > > I guess I will need help from Pierre. > > Francois > Hi Fran?ois, What is your web browser ? It seems that Microsoft Internet Explorer 7 (IE7) loses the PHP Session ID cookie. Do you have the possibility to try with another browser (mozilla, opera, chrome, ...) ? Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 23 - Bureau 24 Email : pierre.marechal at scilab.org ============================================== From pierre.marechal at scilab.org Fri Nov 14 18:04:35 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Fri, 14 Nov 2008 18:04:35 +0100 Subject: [Scilab-Dev] linux_x64 problem In-Reply-To: <491C9823.3010304@free.fr> References: <491C9823.3010304@free.fr> Message-ID: <491DAFA3.7090908@scilab.org> Collette Yann a ?crit : > Hello, > > I think there is a problem with the jdk shipped in the linux_x64 > requirements. > It is not possible to use this jdk to build a java application. > if I do a ls -l in the java/jdk1.... directory, there is a circular > symlink: > > drwxr-xr-x 3 collette collette 4096 2008-11-13 18:39 bin/ > -rw-r--r-- 1 collette collette 3767 2008-11-13 18:39 COPYRIGHT > drwxr-xr-x 8 collette collette 4096 2008-11-13 18:36 db/ > drwxr-xr-x 10 collette collette 4096 2008-11-13 18:36 demo/ > drwxr-xr-x 4 collette collette 4096 2008-11-13 18:35 include/ > *lrwxrwxrwx 1 collette collette 11 2008-11-13 18:39 jdk1.6.0_10 > -> jdk1.6.0_10* > drwxr-xr-x 6 collette collette 4096 2008-11-13 18:35 jre/ > drwxr-xr-x 4 collette collette 4096 2008-11-13 18:39 lib/ > -rw-r--r-- 1 collette collette 17058 2008-11-13 18:39 LICENSE > drwxr-xr-x 5 collette collette 4096 2008-11-13 18:39 man/ > -rw-r--r-- 1 collette collette 28599 2008-11-13 18:39 README.html > -rw-r--r-- 1 collette collette 25390 2008-11-13 18:39 README_ja.html > -rw-r--r-- 1 collette collette 20768 2008-11-13 18:39 > README_zh_CN.html > -rw-r--r-- 1 collette collette 5334 2008-11-13 18:39 register.html > -rw-r--r-- 1 collette collette 5766 2008-11-13 18:39 > register_ja.html > -rw-r--r-- 1 collette collette 4944 2008-11-13 18:39 > register_zh_CN.html > drwxr-xr-x 9 collette collette 4096 2008-11-13 18:35 sample/ > -rw-r--r-- 1 collette collette 19009557 2008-11-13 18:39 src.zip > -rw-r--r-- 1 collette collette 228347 2008-11-13 18:39 > THIRDPARTYLICENSEREADME.txt > > YC > Hello Yann, I've removed this symbolic link. Pierre -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 23 - Bureau 24 Email : pierre.marechal at scilab.org ============================================== From ycollet at freesurf.fr Fri Nov 14 18:46:44 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Fri, 14 Nov 2008 18:46:44 +0100 Subject: [Scilab-Dev] linux_x64 problem In-Reply-To: <491DAFA3.7090908@scilab.org> References: <491C9823.3010304@free.fr> <491DAFA3.7090908@scilab.org> Message-ID: <491DB984.4080603@freesurf.fr> Thanks, now there is still a problem. When I launch javac from the jdk directory, I get: ./javac: error while loading shared libraries: libjli.so: cannot open shared object file: No such file or directory I really need to use this jdk because the one shipped with mandriva has a font problem. YC >> > Hello Yann, > > I've removed this symbolic link. > > Pierre From fvogelnew1 at free.fr Fri Nov 14 20:37:49 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Fri, 14 Nov 2008 20:37:49 +0100 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491DA8A6.4040505@scilab.org> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> <491CA346.9010001@free.fr> <491DA8A6.4040505@scilab.org> Message-ID: <491DD38D.6080308@free.fr> Pierre MARECHAL said on 14/11/2008 17:34: > Fran?ois Vogel a ?crit : >> Collette Yann said on 13/11/2008 22:41: >>>> http://compilation_chain.scilab.org/index.php >>>> >>>> Can't connect. Did I forget my password maybe? >>> >>> With a good passwaord, it works. >> >> The page says password is the subversion password. I can't make it >> work with my subversion password. I could login some time ago, can't >> say exactly when. >> >> I guess I will need help from Pierre. >> > > What is your web browser ? It seems that Microsoft Internet Explorer 7 > (IE7) loses the PHP Session ID cookie. > Do you have the possibility to try with another browser (mozilla, opera, > chrome, ...) ? I'm usually using IE6 (was the case yesterday when it was NOK). Now it works with both IE6 and IE7. How fishy... Did you change anything on your side since yesterday night? Tonight it's now working and I'm doing just the same as yesterday. Well, when I say "it works" I really mean that I can log in. Clicking of anything then re-asks me to log in (and login works), then click, then login... and so on. A pain you know where. Francois From ycollet at freesurf.fr Fri Nov 14 21:42:07 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Fri, 14 Nov 2008 21:42:07 +0100 Subject: [Scilab-Dev] linux_x64 problem In-Reply-To: <491DB984.4080603@freesurf.fr> References: <491C9823.3010304@free.fr> <491DAFA3.7090908@scilab.org> <491DB984.4080603@freesurf.fr> Message-ID: <491DE29F.5080303@freesurf.fr> I found an interesting thread related to this problem: https://bugs.launchpad.net/ubuntu/+source/sun-java6/+bug/103933 YC Collette Yann a ?crit : > Thanks, now there is still a problem. When I launch javac from the jdk > directory, I get: > > ./javac: error while loading shared libraries: libjli.so: cannot open > shared object file: No such file or directory > > I really need to use this jdk because the one shipped with mandriva > has a font problem. > > YC >>> >> Hello Yann, >> >> I've removed this symbolic link. >> >> Pierre > From sylvestre.ledru at inria.fr Sat Nov 15 04:49:51 2008 From: sylvestre.ledru at inria.fr (Sylvestre Ledru) Date: Sat, 15 Nov 2008 04:49:51 +0100 Subject: [Scilab-Dev] linux_x64 problem In-Reply-To: <491DE29F.5080303@freesurf.fr> References: <491C9823.3010304@free.fr> <491DAFA3.7090908@scilab.org> <491DB984.4080603@freesurf.fr> <491DE29F.5080303@freesurf.fr> Message-ID: <1226720991.3967.0.camel@zlarin> Le vendredi 14 novembre 2008 ? 21:42 +0100, Collette Yann a ?crit : > I found an interesting thread related to this problem: > https://bugs.launchpad.net/ubuntu/+source/sun-java6/+bug/103933 It is the first time that I hear about libjli in the JVM (btw, any idea about the meaning?) Please let me know if you find a fix. Sylvestre From ycollet at freesurf.fr Sat Nov 15 18:17:40 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Sat, 15 Nov 2008 18:17:40 +0100 Subject: [Scilab-Dev] linux_x64 problem In-Reply-To: <1226720991.3967.0.camel@zlarin> References: <491C9823.3010304@free.fr> <491DAFA3.7090908@scilab.org> <491DB984.4080603@freesurf.fr> <491DE29F.5080303@freesurf.fr> <1226720991.3967.0.camel@zlarin> Message-ID: <491F0434.1020202@freesurf.fr> JLI = /Java/ Legacy Interface YC Sylvestre Ledru a ?crit : > Le vendredi 14 novembre 2008 ? 21:42 +0100, Collette Yann a ?crit : > >> I found an interesting thread related to this problem: >> https://bugs.launchpad.net/ubuntu/+source/sun-java6/+bug/103933 >> > It is the first time that I hear about libjli in the JVM (btw, any idea > about the meaning?) > Please let me know if you find a fix. > > Sylvestre > > From enrico.segre at weizmann.ac.il Sun Nov 16 10:03:15 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 11:03:15 +0200 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491C9F21.4020608@freesurf.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> Message-ID: <491FE1D3.3070208@weizmann.ac.il> > By the way, for your matio problem: have you added a --enable-shared > option to the configure script of matio ? > By default, matio builds a static library which is not detected by > scilab (it detects only shared library). no, in fact I didn't think of, and was surprised that at the end of the build I had only .a. Now make install finally puts libmatio.so in /usr/local/lib/, thanks. > If you try: > export PKG_CONFIG_PATH=path/to/matio/with/the/.pc/file > before the configure, it should work. > Indeed it works with /usr/local/lib/pkgconfig/ where the matio.pc created today is. Thanks for that too, Enrico From enrico.segre at weizmann.ac.il Sun Nov 16 10:42:55 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 11:42:55 +0200 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491FE1D3.3070208@weizmann.ac.il> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> <491FE1D3.3070208@weizmann.ac.il> Message-ID: <491FEB1F.3000808@weizmann.ac.il> Enrico Segre wrote: > >> By the way, for your matio problem: have you added a --enable-shared >> option to the configure script of matio ? >> By default, matio builds a static library which is not detected by >> scilab (it detects only shared library). > > no, in fact I didn't think of, and was surprised that at the end of > the build I had only .a. Now make install finally puts libmatio.so in > /usr/local/lib/, thanks. > For the record, it turns out that for matio I need also export CFLAGS=-fPIC; ./configure --enable-shared and for scilab ln -s /usr/local/lib/libmatio.so.0 $SCI/bin/ to make scilab build happily. Enrico From enrico.segre at weizmann.ac.il Sun Nov 16 10:46:28 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 11:46:28 +0200 Subject: [Scilab-Dev] Windows build prerequisites for scilab (GIT master) In-Reply-To: <491FE1D3.3070208@weizmann.ac.il> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> <491FE1D3.3070208@weizmann.ac.il> Message-ID: <491FEBF4.20308@weizmann.ac.il> just to get a # An unexpected error has been detected by Java Runtime Environment: # # SIGSEGV (0xb) at pc=0x00002aaab559c80e, pid=20983, tid=46912684240576 # # Java VM: Java HotSpot(TM) 64-Bit Server VM (10.0-b19 mixed mode linux-amd64) # Problematic frame: # C [libmatio.so.0+0x880e] InflateData+0x395 # # An error report file with more information is saved as: # /usr/share/scilab-master-1225963445/hs_err_pid20983.log # the first time I try to loadmatfile something. ha ha. From roberto.bucher at supsi.ch Sun Nov 16 11:12:37 2008 From: roberto.bucher at supsi.ch (Roberto Bucher) Date: Sun, 16 Nov 2008 11:12:37 +0100 Subject: Modifying permanent variables for Scicos Message-ID: <200811161112.38034.roberto.bucher@supsi.ch> In scilab-4.1.2 I added a new popup menu to the scicos menu and 3 new palettes to the palette list. I'm now trying to generate a new toolbox for scilab-5.0.3 in order to reproduce the same behaviour I had before, but every temptative to modify the "%scicos_menu" and the "scicos_pal" variable gives the same error Redefining permanent variable. How can I change the two variables? I don't want to use the palette editor or similar tools. I'd like to change these variables in the SCI/ contrib/XXXXX/etc/XXXXX.start script. Roberto -- University of Applied Sciences of Southern Switzerland Dept. Innovative Technologies CH-6928 Lugano-Manno http://www.dti.supsi.ch/~bucher From enrico.segre at weizmann.ac.il Sun Nov 16 11:41:52 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 12:41:52 +0200 Subject: [Scilab-Dev] Modifying permanent variables for Scicos In-Reply-To: <200811161112.38034.roberto.bucher@supsi.ch> References: <200811161112.38034.roberto.bucher@supsi.ch> Message-ID: <491FF8F0.4090205@weizmann.ac.il> > but every temptative to > modify the "%scicos_menu" and the "scicos_pal" variable gives the same > error > > Redefining permanent variable. > > How can I change the two variables? > lower predef() before modifying and restore it afterwards. From roberto.bucher at supsi.ch Sun Nov 16 12:16:47 2008 From: roberto.bucher at supsi.ch (Roberto Bucher) Date: Sun, 16 Nov 2008 12:16:47 +0100 Subject: [Scilab-Dev] Modifying permanent variables for Scicos In-Reply-To: <491FF8F0.4090205@weizmann.ac.il> References: <200811161112.38034.roberto.bucher@supsi.ch> <491FF8F0.4090205@weizmann.ac.il> Message-ID: <200811161216.48119.roberto.bucher@supsi.ch> Thanks for the prompt answer. I did the following: predef('c') scicos_pal($+1,:)=['RTAI-Lib','SCI/share/scilab/contrib/RTAI/macros/RTAI- Lib.cosf']; but when I try to protect the variables again I have the error -->predef('a') !--error 13 Redefining permanent variable. Roberto On Sunday 16 November 2008 11:41:52 Enrico Segre wrote: > > but every temptative to > > modify the "%scicos_menu" and the "scicos_pal" variable gives the same > > error > > > > Redefining permanent variable. > > > > How can I change the two variables? > > lower predef() before modifying and restore it afterwards. -- University of Applied Sciences of Southern Switzerland Dept. Innovative Technologies CH-6928 Lugano-Manno http://www.dti.supsi.ch/~bucher From enrico.segre at weizmann.ac.il Sun Nov 16 12:27:18 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 13:27:18 +0200 Subject: [Scilab-Dev] Modifying permanent variables for Scicos In-Reply-To: <200811161216.48119.roberto.bucher@supsi.ch> References: <200811161112.38034.roberto.bucher@supsi.ch> <491FF8F0.4090205@weizmann.ac.il> <200811161216.48119.roberto.bucher@supsi.ch> Message-ID: <49200396.6070208@weizmann.ac.il> > but when I try to protect the variables again I have the error > > -->predef('a') > !--error 13 > Redefining permanent variable. > > bug of predef in scilab5 then? In fact, predef('a') gives me error even in a fresh session. I have an idea about where this could come from: -->predef() ans = 84. //be that, so there are 84 protected variables -->predef(84) ans = 84. 84. //so far so good, 84 were protected before, 84 protected now -->predef('a') !--error 13 Redefining permanent variable. -->predef() !--error 13 Redefining permanent variable. -->predef(84) ans = 85. 84. //wait a moment -- 85 variables before? How come? Perhaps ans is taken into account? Enrico From enrico.segre at weizmann.ac.il Sun Nov 16 12:47:35 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Sun, 16 Nov 2008 13:47:35 +0200 Subject: [Scilab-Dev] Modifying permanent variables for Scicos In-Reply-To: <49200396.6070208@weizmann.ac.il> References: <200811161112.38034.roberto.bucher@supsi.ch> <491FF8F0.4090205@weizmann.ac.il> <200811161216.48119.roberto.bucher@supsi.ch> <49200396.6070208@weizmann.ac.il> Message-ID: <49200857.40502@weizmann.ac.il> Enrico Segre wrote: > > bug of predef in scilab5 then? no, present in 4.1.2 as well, with side implications. See for example http://bugzilla.scilab.org/show_bug.cgi?id=686 http://bugzilla.scilab.org/show_bug.cgi?id=667 I think anyway predef() is crap. Other pearls: -->predef([1 2]) !--error 204 predef: Wrong type for argument 1: Scalar, 'min' or 'max' expected. -->predef('max') !--error 901 Unknown option and: -->predef('a');predef('a') !--error 13 Redefining permanent variable. -->quit Warning: recursion problem..., cleared. From pierre.marechal at scilab.org Sun Nov 16 23:25:22 2008 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Sun, 16 Nov 2008 23:25:22 +0100 Subject: [Scilab-Dev] Git / Scilab In-Reply-To: <49131527.3050207@weizmann.ac.il> References: <1224250230.18336.330.camel@korcula.inria.fr> <49131527.3050207@weizmann.ac.il> Message-ID: <49209DD2.5060205@scilab.org> Enrico Segre a ?crit : > I sit behind a firewall and git clone git://git.scilab.org/scilab > just times out. Is there an option to open the access on a less > efficient http: protocol, or something the like, on your side? > Otherwise, proposed workaraounds? > > Enrico Hi Enrico, I've set up an http access (read-only) : git clone http://git.scilab.org/scilab.git This access is updated every ten minutes : /"git update-server-info : A dumb server that does not do on-the-fly pack generations must have some auxiliary information files in $GIT_DIR/info and $GIT_OBJECT_DIRECTORY/info directories to help clients discover what references and packs the server has. This command generates such auxiliary files."/ Pierre [1] http://www.kernel.org/pub/software/scm/git/docs/git-update-server-info.html -- ============================================== Pierre MARECHAL INRIA - Unit? de Recherche de Rocquencourt Domaine de Voluceau - B.P. 105 78153 Le Chesnay Cedex ============================================== Projet Scilab B?timent 1B - Bureau 008 Email : pierre.marechal at inria.fr ============================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: From enrico.segre at weizmann.ac.il Mon Nov 17 09:36:15 2008 From: enrico.segre at weizmann.ac.il (Enrico Segre) Date: Mon, 17 Nov 2008 10:36:15 +0200 Subject: [Scilab-Dev] Git / Scilab In-Reply-To: <49209DD2.5060205@scilab.org> References: <1224250230.18336.330.camel@korcula.inria.fr> <49131527.3050207@weizmann.ac.il> <49209DD2.5060205@scilab.org> Message-ID: <49212CFF.3050404@weizmann.ac.il> Pierre MARECHAL wrote: > I've set up an http access (read-only) : > git clone http://git.scilab.org/scilab.git git clone http://git.scilab.org/scilab.git .... walk 925017849e14166f3cbb280179a6725318b29250 got ed6c588c4bf575397ed18ee58cf0cb79c9fbf563 got 1f70c4246d05d7d64eeba65212eb4c5818e40ac4 error: The requested URL returned error: 500 (curl_result = 22, http_code = 500, sha1 = eea9d542c56b6a43f47ec27346fcc1d44394a273) error: Unable to find eea9d542c56b6a43f47ec27346fcc1d44394a273 under http://git.scilab.org/scilab.git Cannot obtain needed tree eea9d542c56b6a43f47ec27346fcc1d44394a273 while processing commit 925017849e14166f3cbb280179a6725318b29250. and apparently no subdir scilab is created. Never mind, I don't have time for that now. Enrico From fvogelnew1 at free.fr Mon Nov 17 22:04:25 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Mon, 17 Nov 2008 22:04:25 +0100 Subject: What dev environment for java? Message-ID: <4921DC59.2030101@free.fr> Hi all, Because bug 3733 is a problem for me, I'm currently considering how to debug java source code on Windows. I'm currently using VC Express + f2c to edit and compile C and Fortran files, but I have no development environment capable of editing and debugging java nicely. I suspect there is a solution with Eclipse, and I have to try this (what eclipse package should I use exactly, btw?), but I wanted first to get feedback from you about what is your dev environment, especially on windows? Any recommendation perhaps? Thanks, Francois From fvogelnew1 at free.fr Wed Nov 19 05:15:51 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Wed, 19 Nov 2008 05:15:51 +0100 Subject: What dev environment for java? Message-ID: <492392F7.205@free.fr> Hi all, Because bug 3733 is a problem for me, I'm currently considering how to debug java source code on Windows. I'm currently using VC Express + f2c to edit and compile C and Fortran files, but I have no development environment capable of editing and debugging java nicely. I suspect there is a solution with Eclipse, and I have to try this (what eclipse package should I use exactly, btw?), but I wanted first to get feedback from you about what is your dev environment, especially on windows? Any recommendation perhaps? Thanks, Francois From sylvestre.ledru at scilab.org Wed Nov 19 09:39:39 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Wed, 19 Nov 2008 09:39:39 +0100 Subject: [Scilab-Dev] What dev environment for java? In-Reply-To: <492392F7.205@free.fr> References: <492392F7.205@free.fr> Message-ID: <1227083979.5910.7.camel@korcula.inria.fr> Le mercredi 19 novembre 2008 ? 05:15 +0100, Fran?ois Vogel a ?crit : > Hi all, > > Because bug 3733 is a problem for me, I'm currently considering how to > debug java source code on Windows. > > I'm currently using VC Express + f2c to edit and compile C and Fortran > files, but I have no development environment capable of editing and > debugging java nicely. Debugging Java within Scilab is quite tricky. I don't even know if it is possible because of JNI. (Jb?) However, what I do to debug it is adding a .*main.* method into the Java class just to get interesting the piece of code. You will find an example of this here: modules/helptools/src/java/org/scilab/modules/helptools/BuildDocObject.java About eclipse & co, I cannot answer you. I am using emacs & ant. Sylvestre From ycollet at freesurf.fr Wed Nov 19 09:48:13 2008 From: ycollet at freesurf.fr (ycollet at freesurf.fr) Date: Wed, 19 Nov 2008 09:48:13 +0100 Subject: Catch a CTRL+C Message-ID: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> Hello, Is it possible to catch a CTRL+C (entered in the scilab shell) in a C interface ? I have a C interface which takes a lot of time to execute and I want to interrupt this C interface while I am fine tuning some algorithms. YC From sylvestre.ledru at scilab.org Wed Nov 19 17:43:55 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Wed, 19 Nov 2008 17:43:55 +0100 Subject: [Scilab-Dev] Catch a CTRL+C In-Reply-To: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> References: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> Message-ID: <1227113035.5910.69.camel@korcula.inria.fr> Hello, man signal Calling this function (under Linux) from your interface could do the trick. Sylvestre Le mercredi 19 novembre 2008 ? 09:48 +0100, ycollet at freesurf.fr a ?crit : > Hello, > > Is it possible to catch a CTRL+C (entered in the scilab shell) in a C > interface ? > I have a C interface which takes a lot of time to execute and I want > to interrupt this C interface while I am fine tuning some algorithms. > > YC > > > From ycollet at freesurf.fr Wed Nov 19 20:19:24 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Wed, 19 Nov 2008 20:19:24 +0100 Subject: [Scilab-Dev] Catch a CTRL+C In-Reply-To: <1227113035.5910.69.camel@korcula.inria.fr> References: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> <1227113035.5910.69.camel@korcula.inria.fr> Message-ID: <492466BC.1080501@freesurf.fr> I have already tried under windows and it doesn't work. I am not able to interrupt the interface via a ctrl+c. I will try under linux to see if it works. YC Sylvestre Ledru a ?crit : > Hello, > > man signal > Calling this function (under Linux) from your interface could do the > trick. > > Sylvestre > > > Le mercredi 19 novembre 2008 ? 09:48 +0100, ycollet at freesurf.fr a > ?crit : > >> Hello, >> >> Is it possible to catch a CTRL+C (entered in the scilab shell) in a C >> interface ? >> I have a C interface which takes a lot of time to execute and I want >> to interrupt this C interface while I am fine tuning some algorithms. >> >> YC >> >> >> >> > > From Serge.Steer at scilab.org Wed Nov 19 21:19:35 2008 From: Serge.Steer at scilab.org (steer) Date: Wed, 19 Nov 2008 21:19:35 +0100 Subject: [Scilab-Dev] Catch a CTRL+C In-Reply-To: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> References: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> Message-ID: <492474D7.7000207@scilab.org> ycollet at freesurf.fr a ?crit : > Hello, > > Is it possible to catch a CTRL+C (entered in the scilab shell) in a C > interface ? > I have a C interface which takes a lot of time to execute and I want > to interrupt this C interface while I am fine tuning some algorithms. > > YC > > > > Le Ctrl-C est trapp? par Scilab qui assigne true a la variable C2F(basbrk).iflag Il suffit donc de tester cette valeur et ne pas oublier de la remettre a false une fois prise en compte Serge From ycollet at freesurf.fr Thu Nov 20 18:24:41 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Thu, 20 Nov 2008 18:24:41 +0100 Subject: [Scilab-Dev] Catch a CTRL+C In-Reply-To: <492474D7.7000207@scilab.org> References: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> <492474D7.7000207@scilab.org> Message-ID: <49259D59.7080302@freesurf.fr> Hello, Thanks to Serge answer. The variable C2F(basbrk).iflag is not visible in scilab. Impossible to link a program with such a symbol. But I managed to find a function in the core module: csignal.[ch]. This function defines a handler to deal with SIGINT. I was wondering if it was possible to extend this function so as to a user defined handler before the scilab one. Something like this: csignal.c: #include "banier.h" #include "csignal.h" #include "sigbas.h" extern int C2F(sigbas)(); typedef void (*sig_func_t)(int); void controlC_handler (int); static user_controlC_handler = controlC_handler; void set_controlC_handler (sig_func_t myHandler) { user_controlC_handler = myHandler; } sig_func_t get_controlC_handler () { return user_controlC_handler; } void controlC_handler (int sig) { int j = SIGINT; C2F(sigbas)(&j); } int C2F(csignal)(void) { signal (SIGINT, user_controlC_handler); return(0); } csignal.h: #ifndef __CSIGNAL_H__ #define __CSIGNAL_H__ #include #include "machine.h" void controlC_handler (int sig); typedef void (*sig_func_t)(int); void set_usercontrolC_handler(sig_func_t handler); sig_func_t get_controlC_handler (); int C2F(csignal)(void); #endif /* __CSIGNAL_H__ */ With such functions, to add a new SIGINT handler is easier: sig_func_t old_handler = get_controlC_handler(); void my_controlC_handler(int sig) { // Do something to close cleanly the C interface // Now call the scilab handler (*old_handler)(sig); } set_controlC_handler(my_controlC_handler); I am open to suggestions ... Can I commit these changes ? YC steer a ?crit : > ycollet at freesurf.fr a ?crit : > >> Hello, >> >> Is it possible to catch a CTRL+C (entered in the scilab shell) in a C >> interface ? >> I have a C interface which takes a lot of time to execute and I want >> to interrupt this C interface while I am fine tuning some algorithms. >> >> YC >> >> >> >> >> > Le Ctrl-C est trapp? par Scilab qui assigne true a la variable > C2F(basbrk).iflag > Il suffit donc de tester cette valeur et ne pas oublier de la remettre a > false une fois prise en compte > > Serge > > From sylvestre.ledru at scilab.org Thu Nov 20 19:03:43 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Thu, 20 Nov 2008 19:03:43 +0100 Subject: For fun: Scilab under Mac OS X Message-ID: <1227204223.5910.120.camel@korcula.inria.fr> Hello, It is not finished but most of the work is done. Most of the modification are available in the branch scilab-macosx in the git repository. There is still some crappy bugs and stuff that I must fix but it is coming. I also have some interesting results with Solaris/x86 (Btw, thx to Gerard Henry) but since it is less sexy, I won't send a screenshot here :) Sylvestre -------------- next part -------------- A non-text attachment was scrubbed... Name: scilab_macosx.jpg Type: image/jpeg Size: 291155 bytes Desc: not available URL: From Serge.Steer at scilab.org Thu Nov 20 19:30:21 2008 From: Serge.Steer at scilab.org (steer) Date: Thu, 20 Nov 2008 19:30:21 +0100 Subject: [Scilab-Dev] Catch a CTRL+C In-Reply-To: <49259D59.7080302@freesurf.fr> References: <20081119094813.15953ir4xm9yx2w4@webmail-html.freesurf.fr> <492474D7.7000207@scilab.org> <49259D59.7080302@freesurf.fr> Message-ID: <4925ACBD.7020508@scilab.org> Your proposal seems to me a good idea. Note however that if you want to cancel the execution of a C code you will have to generate a Scilab error to clean the calling environnement, so it is not ok to call the (*old_handler)(sig); but you need to reset the default controlC_handler... Serge Steer Collette Yann a ?crit : > Hello, > > Thanks to Serge answer. > The variable C2F(basbrk).iflag is not visible in scilab. Impossible to > link a program with such a symbol. > But I managed to find a function in the core module: > csignal.[ch]. This function defines a handler to deal with SIGINT. > I was wondering if it was possible to extend this function so as to a > user defined handler before the scilab one. > > Something like this: > > csignal.c: > #include "banier.h" > #include "csignal.h" > #include "sigbas.h" > > extern int C2F(sigbas)(); > > typedef void (*sig_func_t)(int); > > void controlC_handler (int); > > static user_controlC_handler = controlC_handler; > > void set_controlC_handler (sig_func_t myHandler) > { > user_controlC_handler = myHandler; > } > > sig_func_t get_controlC_handler () > { > return user_controlC_handler; > } > > void controlC_handler (int sig) > { > int j = SIGINT; > C2F(sigbas)(&j); > } > > int C2F(csignal)(void) > { > signal (SIGINT, user_controlC_handler); > return(0); > } > > > csignal.h: > #ifndef __CSIGNAL_H__ > #define __CSIGNAL_H__ > #include > #include "machine.h" > > void controlC_handler (int sig); > > typedef void (*sig_func_t)(int); > > void set_usercontrolC_handler(sig_func_t handler); > sig_func_t get_controlC_handler (); > > int C2F(csignal)(void); > > #endif /* __CSIGNAL_H__ */ > > > With such functions, to add a new SIGINT handler is easier: > > sig_func_t old_handler = get_controlC_handler(); > > void my_controlC_handler(int sig) > { > // Do something to close cleanly the C interface > // Now call the scilab handler > (*old_handler)(sig); > } > > set_controlC_handler(my_controlC_handler); > > I am open to suggestions ... > Can I commit these changes ? > > YC > steer a ?crit : >> ycollet at freesurf.fr a ?crit : >> >>> Hello, >>> >>> Is it possible to catch a CTRL+C (entered in the scilab shell) in a C >>> interface ? >>> I have a C interface which takes a lot of time to execute and I want >>> to interrupt this C interface while I am fine tuning some algorithms. >>> >>> YC >>> >>> >>> >>> >>> >> Le Ctrl-C est trapp? par Scilab qui assigne true a la variable >> C2F(basbrk).iflag >> Il suffit donc de tester cette valeur et ne pas oublier de la remettre a >> false une fois prise en compte >> >> Serge >> >> > > From BlanchardJ at ieee.org Thu Nov 20 21:27:16 2008 From: BlanchardJ at ieee.org (Jonathan Blanchard) Date: Thu, 20 Nov 2008 16:27:16 -0400 Subject: [Scilab-Dev] For fun: Scilab under Mac OS X In-Reply-To: <1227204223.5910.120.camel@korcula.inria.fr> References: <1227204223.5910.120.camel@korcula.inria.fr> Message-ID: Since I've been using Scilab under Solaris for quite a while now I am quite interested by this, does it mean you might be considering an official build under Solaris? Jonathan Blanchard On Thu, Nov 20, 2008 at 2:03 PM, Sylvestre Ledru wrote: > Hello, > > It is not finished but most of the work is done. > Most of the modification are available in the branch scilab-macosx in > the git repository. There is still some crappy bugs and stuff that I > must fix but it is coming. > I also have some interesting results with Solaris/x86 (Btw, thx to > Gerard Henry) but since it is less sexy, I won't send a screenshot > here :) > > Sylvestre > > > From sylvestre.ledru at inria.fr Fri Nov 21 01:01:28 2008 From: sylvestre.ledru at inria.fr (Sylvestre Ledru) Date: Fri, 21 Nov 2008 01:01:28 +0100 Subject: [Scilab-Dev] For fun: Scilab under Mac OS X In-Reply-To: References: <1227204223.5910.120.camel@korcula.inria.fr> Message-ID: <1227225688.5691.129.camel@zlarin> It is not (for now at least) in our todo list but I think that building regulary a software on misc archs, os and compilers makes a software better (and it is making life of contributors and users easier). Sylvestre Le jeudi 20 novembre 2008 ? 16:27 -0400, Jonathan Blanchard a ?crit : > Since I've been using Scilab under Solaris for quite a while now I am > quite interested by this, does it mean you might be considering an > official build under Solaris? > > Jonathan Blanchard > > > > On Thu, Nov 20, 2008 at 2:03 PM, Sylvestre Ledru > wrote: > > Hello, > > > > It is not finished but most of the work is done. > > Most of the modification are available in the branch scilab-macosx in > > the git repository. There is still some crappy bugs and stuff that I > > must fix but it is coming. > > I also have some interesting results with Solaris/x86 (Btw, thx to > > Gerard Henry) but since it is less sexy, I won't send a screenshot > > here :) > > > > Sylvestre > > > > > > From fvogelnew1 at free.fr Fri Nov 21 01:21:37 2008 From: fvogelnew1 at free.fr (=?UTF-8?B?RnJhbsOnb2lzIFZvZ2Vs?=) Date: Fri, 21 Nov 2008 01:21:37 +0100 Subject: [Scilab-Dev] What dev environment for java? In-Reply-To: <1227083979.5910.7.camel@korcula.inria.fr> References: <492392F7.205@free.fr> <1227083979.5910.7.camel@korcula.inria.fr> Message-ID: <4925FF11.6020000@free.fr> Sylvestre, Thanks for your answer. Btw I sent my question twice because the first time I didn't receive it from the mailing list (and the second time neither). I can now see both messages by browsing the list on the Internet. Did you perhaps change something recently in the mailing list settings? Some option preventing the poster to receive his own messages back? Sylvestre Ledru said on 19/11/2008 09:39: > Debugging Java within Scilab is quite tricky. I don't even know if it is > possible because of JNI. (Jb?) > However, what I do to debug it is adding a .*main.* method into the Java > class just to get interesting the piece of code. > > You will find an example of this here: > modules/helptools/src/java/org/scilab/modules/helptools/BuildDocObject.java Sorry I do not see what you mean here. Where in the above file did you add the main method? Could you please be more specific? > About eclipse & co, I cannot answer you. I am using emacs & ant. Could perhaps other members of the opteam answer the question too, especially on Windows? Thanks in advance. Francois From fvogelnew1 at free.fr Fri Nov 21 01:38:08 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Fri, 21 Nov 2008 01:38:08 +0100 Subject: Disable Tk when making doc Message-ID: <492602F0.8070007@free.fr> Hi, I've see the work today around disabling Tk when making doc. I'm wondering what are the "issues when using Scilab through ssh". Could you please explain? I'm asking because I think it's strange you apparently only need to disable Tk and not Tcl (that's what you did in the commits today). I suspect the problem could perhaps be better fixed in Tk, but I can't be sure since I don't know what's the problem exactly. Also, I don't see a reason for keeping Tcl while disabling Tk. This is certainly unexpected. Using Tcl without Tk is not the usual assumption in Scilab. Even if this is only during make doc, overall there might be a better fix that would not need yet another environment variable to tune Tcl/Tk init. Thanks for a few words of clarification. Francois From sylvestre.ledru at inria.fr Fri Nov 21 08:34:22 2008 From: sylvestre.ledru at inria.fr (Sylvestre Ledru) Date: Fri, 21 Nov 2008 08:34:22 +0100 Subject: [Scilab-Dev] Disable Tk when making doc In-Reply-To: <492602F0.8070007@free.fr> References: <492602F0.8070007@free.fr> Message-ID: <1227252862.5691.141.camel@zlarin> Hello, Le vendredi 21 novembre 2008 ? 01:38 +0100, Fran?ois Vogel a ?crit : > Hi, > > I've see the work today around disabling Tk when making doc. > I'm wondering what are the "issues when using Scilab through ssh". > Could you please explain? I don't know if you had a look to the code but it is not a feature available to the lambda user. The issue that I fixed here was that when you start Scilab through a ssh connexion, Tk tries to connect itself to the X server which obviously doesn't exist. It sets an error which is breaking Scilab. Sylvestre From sylvestre.ledru at scilab.org Fri Nov 21 14:18:10 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Fri, 21 Nov 2008 14:18:10 +0100 Subject: [Scilab-Dev] What dev environment for java? In-Reply-To: <4925FF11.6020000@free.fr> References: <492392F7.205@free.fr> <1227083979.5910.7.camel@korcula.inria.fr> <4925FF11.6020000@free.fr> Message-ID: <1227273490.5910.128.camel@korcula.inria.fr> Le vendredi 21 novembre 2008 ? 01:21 +0100, Fran?ois Vogel a ?crit : > Sylvestre, > > Thanks for your answer. Btw I sent my question twice because the first > time I didn't receive it from the mailing list (and the second time > neither). I can now see both messages by browsing the list on the > Internet. Did you perhaps change something recently in the mailing > list settings? Some option preventing the poster to receive his own > messages back? AFAIK, no. I received both emails from you. > Sylvestre Ledru said on 19/11/2008 09:39: > > Debugging Java within Scilab is quite tricky. I don't even know if it is > > possible because of JNI. (Jb?) > > However, what I do to debug it is adding a .*main.* method into the Java > > class just to get interesting the piece of code. > > > > You will find an example of this here: > > modules/helptools/src/java/org/scilab/modules/helptools/BuildDocObject.java > > Sorry I do not see what you mean here. Where in the above file did you > add the main method? Could you please be more specific? [14:16:33][sylvestre at korcula] ~/dev/git/scilab.master/scilab$ cat modules/helptools/src/java/org/scilab/modules/helptools/BuildDocObject.java |grep -n main 295: public static void main(String[] arg) { The method here is just a stand alone example when I want to test directly the java feature without going into Scilab. public static void main(String[] arg) { try { BuildDocObject d = new BuildDocObject(); d.setOutputDirectory("/tmp/"); d.setExportFormat(JH_FORMAT); d.setDocbookPath("/usr/share/xml/docbook/stylesheet/nwalsh/"); d.process(SCI+"/modules/helptools/master_en_US_help.xml", SCI+"/modules/helptools/css/javahelp.css"); } catch (FileNotFoundException e) { System.err.println("Exception catched: " + e.getMessage()); } } Sylvestre From sylvestre.ledru at scilab.org Fri Nov 21 18:55:14 2008 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Fri, 21 Nov 2008 18:55:14 +0100 Subject: Scilab packages for Debian & Ubuntu Message-ID: <1227290114.5910.219.camel@korcula.inria.fr> Hello, To simplify the diffusion of Scilab, I have been working to produce Debian (sid/experimental) & Ubuntu (intrepid) packages. They are available here: http://www.scilab.org/team/sylvestre.ledru/ Note that doesn't mean that we are going to release Scilab all the time like that (even if I probably will). It is a trial balloon. Please read the FAQ (especially the last point)! I also have report of Mandriva, Gentoo & Fedora packages coming soon. Btw, I am going to be off for the next two weeks doing bubbles diving. Sylvestre From fvogelnew1 at free.fr Sun Nov 23 14:55:02 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Sun, 23 Nov 2008 14:55:02 +0100 Subject: [Scilab-Dev] compilation chain website continuously asks for re-login In-Reply-To: <491DD38D.6080308@free.fr> References: <002d01c944e1$8e23efe0$aa6bcfa0$@cornet@scilab.org> <491B2795.1070400@free.fr> <491B3F72.3050003@scilab.org> <491B72EE.3090505@free.fr> <491B7643.6050102@free.fr> <491C7E84.6060800@free.fr> <491C9F21.4020608@freesurf.fr> <491CA346.9010001@free.fr> <491DA8A6.4040505@scilab.org> <491DD38D.6080308@free.fr> Message-ID: <492960B6.7050106@free.fr> Hi Pierre, >> http://compilation_chain.scilab.org/index.php > > Clicking on anything re-asks me to log in (and login works), then > click, then login... and so on. Would it be possible to fix this? Thanks, Francois From ycollette at free.fr Wed Nov 26 20:25:44 2008 From: ycollette at free.fr (Collette Yann) Date: Wed, 26 Nov 2008 20:25:44 +0100 Subject: Catch a ctrl+C Message-ID: <492DA2B8.1020904@free.fr> Hello, I send to the list the example of C interface which uses the C2F(basbrk).iflag to catch the ctrl+c to stop a C interface. I have tested the script under linux up to now. I need to test it under windows. Just one question: what is the function of C2F(basbrk).interruptible ? Have a good day, YC -------------- next part -------------- A non-text attachment was scrubbed... Name: ctrlc.zip Type: application/zip Size: 2188 bytes Desc: not available URL: From fvogelnew1 at free.fr Thu Nov 27 21:17:30 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 27 Nov 2008 21:17:30 +0100 Subject: [Scilab-Dev] Catch a ctrl+C In-Reply-To: <492DA2B8.1020904@free.fr> References: <492DA2B8.1020904@free.fr> Message-ID: <492F005A.1020505@free.fr> Hi, Collette Yann said on 26/11/2008 20:25: > Just one question: what is the function of C2F(basbrk).interruptible ? Did you have an answer to this one? I see there has been activity in the master branch around it today. Francois From ycollet at freesurf.fr Thu Nov 27 22:56:51 2008 From: ycollet at freesurf.fr (Collette Yann) Date: Thu, 27 Nov 2008 22:56:51 +0100 Subject: [Scilab-Dev] Catch a ctrl+C In-Reply-To: <492F005A.1020505@free.fr> References: <492DA2B8.1020904@free.fr> <492F005A.1020505@free.fr> Message-ID: <492F17A3.3010406@freesurf.fr> Fran?ois Vogel a ?crit : > Hi, > > Collette Yann said on 26/11/2008 20:25: >> Just one question: what is the function of C2F(basbrk).interruptible ? > > Did you have an answer to this one? > > I see there has been activity in the master branch around it today. > > Francois > Not yet, but Allan exported the basbrk variable in a dll. This variable was accessible under linux, but not under windows. YC From fvogelnew1 at free.fr Thu Nov 27 23:08:39 2008 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Thu, 27 Nov 2008 23:08:39 +0100 Subject: [Scilab-Dev] Catch a ctrl+C In-Reply-To: <492F17A3.3010406@freesurf.fr> References: <492DA2B8.1020904@free.fr> <492F005A.1020505@free.fr> <492F17A3.3010406@freesurf.fr> Message-ID: <492F1A67.70001@free.fr> Collette Yann said on 27/11/2008 22:56: >>> Just one question: what is the function of C2F(basbrk).interruptible ? OK, here is what (I think) I know about this: This variable has been introduced when the sync option of ScilabEval has been added approximately 3-4 years ago. Without this variable Scilab execution flow could be interrupted at (almost) any time. Now, it is possible to instruct Scilab to execute something without being possibly interrupted. This is the case with ScilabEval "..." "sync". Try the examples in help ScilabEval as a starting point to understand this. You can also have a further look at the internals in the source code. Start in ScilabEval.c, dynamic_menus.c (especially the flag attribute in the commandRec struct), and parse.c. When an uninterruptible command is launched it gets executed in syncexec. Francois