From laurent.berger at univ-lemans.fr Wed Nov 4 13:02:08 2009 From: laurent.berger at univ-lemans.fr (Laurent Berger) Date: Wed, 04 Nov 2009 13:02:08 +0100 Subject: DLL thread wxWidgets Message-ID: <4AF16D40.80606@univ-lemans.fr> Hi, I have written a dll which launch wxMinimal a wxWidgets sample. In my project executable file load dll and can call a proc which run a wxMinimal. This proc can be a thread or not. With my own application everything is ok using a thread or not. Now if I use scilab with commands link and call without thread everything is fined and with a thread scilab crashed with Acces Violation at line 476 of toplevel.cpp (wxWidgets source file). I have check that if thread is empty scibab doesn't crached. I don't understand what's wrong. Any idea? Laurent From allan.cornet at scilab.org Wed Nov 4 13:21:33 2009 From: allan.cornet at scilab.org (Allan CORNET) Date: Wed, 4 Nov 2009 13:21:33 +0100 Subject: [Scilab-Dev] DLL thread wxWidgets In-Reply-To: <4AF16D40.80606@univ-lemans.fr> References: <4AF16D40.80606@univ-lemans.fr> Message-ID: <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> Hi, Please attach a mininal example. Thanks you Allan CORNET -----Message d'origine----- De?: Laurent Berger [mailto:laurent.berger at univ-lemans.fr] Envoy??: mercredi 4 novembre 2009 13:02 ??: dev at lists.scilab.org Objet?: [Scilab-Dev] DLL thread wxWidgets Hi, I have written a dll which launch wxMinimal a wxWidgets sample. In my project executable file load dll and can call a proc which run a wxMinimal. This proc can be a thread or not. With my own application everything is ok using a thread or not. Now if I use scilab with commands link and call without thread everything is fined and with a thread scilab crashed with Acces Violation at line 476 of toplevel.cpp (wxWidgets source file). I have check that if thread is empty scibab doesn't crached. I don't understand what's wrong. Any idea? Laurent From laurent.berger at univ-lemans.fr Wed Nov 4 13:40:46 2009 From: laurent.berger at univ-lemans.fr (Laurent Berger) Date: Wed, 04 Nov 2009 13:40:46 +0100 Subject: [Scilab-Dev] DLL thread wxWidgets In-Reply-To: <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> References: <4AF16D40.80606@univ-lemans.fr> <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> Message-ID: <4AF1764E.8080304@univ-lemans.fr> In attached zip file you will find *source file of dll -> MiniMalWX/ dllmain.cpp MinimalWx.cpp Minimalwx.h *source file of exe -> AppliDllWx/ main.cpp with exe file in debugwx29 *the dll ->debugwx29/minimalwx.dll *scilab toolbox ->debugwx29/wxMinimal with two scilab functions WxRun() and WxThread() WxRun() is Ok and WxThread() crashed exe file AppliDllWx load dll and call first TestScilabRun and TestScilabThread. This exe file needs wxwidgets dll. Thanks for your interest Allan CORNET a ?crit : > Hi, > > Please attach a mininal example. > > Thanks you > > Allan CORNET > > -----Message d'origine----- > De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] > Envoy? : mercredi 4 novembre 2009 13:02 > ? : dev at lists.scilab.org > Objet : [Scilab-Dev] DLL thread wxWidgets > > Hi, > > I have written a dll which launch wxMinimal a wxWidgets sample. > In my project executable file load dll and can call a proc which run a > wxMinimal. This proc can be a thread or not. > With my own application everything is ok using a thread or not. > > Now if I use scilab with commands link and call without thread > everything is fined and with a thread scilab crashed with Acces > Violation at line 476 of toplevel.cpp (wxWidgets source file). > I have check that if thread is empty scibab doesn't crached. > I don't understand what's wrong. > > Any idea? > > Laurent > > > > > -------------- next part -------------- A non-text attachment was scrubbed... Name: zipscilabdll.zip Type: application/x-zip-compressed Size: 45960 bytes Desc: not available URL: From vincent.lejeune at u-psud.fr Thu Nov 5 08:47:23 2009 From: vincent.lejeune at u-psud.fr (Vincent Lejeune) Date: Thu, 5 Nov 2009 08:47:23 +0100 Subject: Problem using double API Message-ID: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> Hi, I don't know if it's a bug or if something is missing in my code... The issue is : using the following code, I cannot get scilab to display the mB matrix : #include "stack-c.h" #include "api_scilab.h" #include "Scierror.h" int oclexec(char* fname) { SciErr sciErr; ? int *pB; ? int rowsB,colsB; ? double* mB; ? int k; ? mB=(double*)malloc(rowsB*colsB*sizeof(double)); ? for(k=0;k From yann.collette at scilab.org Thu Nov 5 09:44:56 2009 From: yann.collette at scilab.org (Yann Collette) Date: Thu, 05 Nov 2009 09:44:56 +0100 Subject: [Scilab-Dev] Problem using double API In-Reply-To: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> References: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> Message-ID: <4AF29088.8090908@scilab.org> And what about this: the value of rowsB and colsB is not defined ? YC Vincent Lejeune a ?crit : > > Hi, > > I don't know if it's a bug or if something is missing in my code... > The issue is : using the following code, I cannot get scilab to > display the > mB matrix : > > #include "stack-c.h" > #include "api_scilab.h" > #include "Scierror.h" > > int oclexec(char* fname) > { > SciErr sciErr; > int *pB; > int rowsB,colsB; > double* mB; > int k; > > mB=(double*)malloc(rowsB*colsB*sizeof(double)); > > > for(k=0;k { > mB[k]=1.0; > } > > sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); > if(sciErr.iErr) > { > printError(&sciErr, 0); > return 0; > } > > LhsVar(1)=Rhs+1; > > return 0; > } > > The gateway is compiling fine, but when I type B=ocltest(A); , the B > matrix > is equal to A, whereas it is expected to be a ones matrix. > > On the other hand, the unary test double_writing_api.tst works. > I'am currently using the windows 1257195239 build (taken from the nightly > build of scilab 5.2) > > Regards, > Vincent > -- ----------------------------- Yann COLLETTE Software Development Engineer ----------------------------- The Scilab Consortium Digiteo Foundation Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay France Phone: +33.1.39.63.57.82 From allan.cornet at scilab.org Thu Nov 5 11:19:27 2009 From: allan.cornet at scilab.org (Allan CORNET) Date: Thu, 5 Nov 2009 11:19:27 +0100 Subject: [Scilab-Dev] DLL thread wxWidgets In-Reply-To: <4AF1764E.8080304@univ-lemans.fr> References: <4AF16D40.80606@univ-lemans.fr> <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> <4AF1764E.8080304@univ-lemans.fr> Message-ID: <004d01ca5e01$7494d690$5dbe83b0$@cornet@scilab.org> Hi, This example seems to work (see minimal.cpp) with a dedicated thread for wxwidgets. link("Minimal.dll","TestScilabThread","c"); call("TestScilabThread") Best Regards Allan -- ------------------------- Allan CORNET ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France -----Message d'origine----- De?: Laurent Berger [mailto:laurent.berger at univ-lemans.fr] Envoy??: mercredi 4 novembre 2009 13:41 ??: dev at lists.scilab.org Objet?: Re: [Scilab-Dev] DLL thread wxWidgets In attached zip file you will find *source file of dll -> MiniMalWX/ dllmain.cpp MinimalWx.cpp Minimalwx.h *source file of exe -> AppliDllWx/ main.cpp with exe file in debugwx29 *the dll ->debugwx29/minimalwx.dll *scilab toolbox ->debugwx29/wxMinimal with two scilab functions WxRun() and WxThread() WxRun() is Ok and WxThread() crashed exe file AppliDllWx load dll and call first TestScilabRun and TestScilabThread. This exe file needs wxwidgets dll. Thanks for your interest Allan CORNET a ?crit : > Hi, > > Please attach a mininal example. > > Thanks you > > Allan CORNET > > -----Message d'origine----- > De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] > Envoy? : mercredi 4 novembre 2009 13:02 ? : dev at lists.scilab.org Objet > : [Scilab-Dev] DLL thread wxWidgets > > Hi, > > I have written a dll which launch wxMinimal a wxWidgets sample. > In my project executable file load dll and can call a proc which run a > wxMinimal. This proc can be a thread or not. > With my own application everything is ok using a thread or not. > > Now if I use scilab with commands link and call without thread > everything is fined and with a thread scilab crashed with Acces > Violation at line 476 of toplevel.cpp (wxWidgets source file). > I have check that if thread is empty scibab doesn't crached. > I don't understand what's wrong. > > Any idea? > > Laurent > > > > > -------------- next part -------------- A non-text attachment was scrubbed... Name: minimalwx.zip Type: application/octet-stream Size: 2853417 bytes Desc: not available URL: From vincent.lejeune at u-psud.fr Thu Nov 5 12:43:07 2009 From: vincent.lejeune at u-psud.fr (Vincent Lejeune) Date: Thu, 5 Nov 2009 12:43:07 +0100 Subject: [Scilab-Dev] Problem using double API In-Reply-To: <4AF29088.8090908@scilab.org> References: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> <4AF29088.8090908@scilab.org> Message-ID: Sorry, this was a copy/paste failure : the code actually?contains the following lines : rowsB=rowsA; colsB=colsA; (in fact thereal code is here : http://gitweb.scilab.org/?p=cuda;a=blob;f=simd/sci_gateway/c/sci_simd.c;h=d480c8071070ae5596e1bde42babfc862a2f1620;hb=HEAD but it did not work either, but using sciprint I can see that the value returned by gpuCompute is the right one) Le Jeu 5 novembre 2009 09:44, Yann Collette a ?crit : > And what about this: the value of rowsB and colsB is not defined ? > > > YC > > > Vincent Lejeune a ?crit : > >> >> Hi, >> >> >> I don't know if it's a bug or if something is missing in my code... >> The issue is : using the following code, I cannot get scilab to >> display the mB matrix : >> >> #include "stack-c.h" >> #include "api_scilab.h" >> #include "Scierror.h" >> >> >> int oclexec(char* fname) { >> SciErr sciErr; >> int *pB; int rowsB,colsB; double* mB; int k; >> >> mB=(double*)malloc(rowsB*colsB*sizeof(double)); >> >> >> for(k=0;k> mB[k]=1.0; } >> >> >> sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); >> if(sciErr.iErr) { >> printError(&sciErr, 0); return 0; } >> >> >> LhsVar(1)=Rhs+1; >> >> >> return 0; } >> >> >> The gateway is compiling fine, but when I type B=ocltest(A); , the B >> matrix is equal to A, whereas it is expected to be a ones matrix. >> >> On the other hand, the unary test double_writing_api.tst works. >> I'am currently using the windows 1257195239 build (taken from the >> nightly build of scilab 5.2) >> >> Regards, >> Vincent >> >> > > > -- > ----------------------------- > Yann COLLETTE > Software Development Engineer > ----------------------------- > The Scilab Consortium > Digiteo Foundation > Domaine de Voluceau > Rocquencourt - B.P. 105 > 78153 Le Chesnay France > Phone: +33.1.39.63.57.82 > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From laurent.berger at univ-lemans.fr Thu Nov 5 15:39:43 2009 From: laurent.berger at univ-lemans.fr (Laurent Berger) Date: Thu, 05 Nov 2009 15:39:43 +0100 Subject: [Scilab-Dev] DLL thread wxWidgets In-Reply-To: <004d01ca5e01$7494d690$5dbe83b0$@cornet@scilab.org> References: <4AF16D40.80606@univ-lemans.fr> <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> <4AF1764E.8080304@univ-lemans.fr> <004d01ca5e01$7494d690$5dbe83b0$@cornet@scilab.org> Message-ID: <4AF2E3AF.7060007@univ-lemans.fr> Hi, Program is still crashed using your sourcefile in scilab with a call("TestScilabThread"). With a windows program it's OK with minimalwx.dll and minimal.dll. (With your dll some dll cannot be loaded like. With dependacy.exe wer.dll and ieshims.dll are missing. I don't understand why) Size of your dll is 119296 (minimal.dll) and my dll (MinimalWX.dll) 94208 bytes. What are your compile options? my scilab version is scilab-5.1.1 (Apr 14 2009) Best regards Laurent Berger Allan CORNET a ?crit : > Hi, > > This example seems to work (see minimal.cpp) with a dedicated thread for > wxwidgets. > > link("Minimal.dll","TestScilabThread","c"); > call("TestScilabThread") > > Best Regards > > Allan > > -- > ------------------------- > Allan CORNET > ------------------------- > The Scilab Consortium > Digiteo > Domaine de Voluceau > Rocquencourt - B.P. 105 > 78153 Le Chesnay Cedex > France > > -----Message d'origine----- > De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] > Envoy? : mercredi 4 novembre 2009 13:41 > ? : dev at lists.scilab.org > Objet : Re: [Scilab-Dev] DLL thread wxWidgets > > In attached zip file you will find > *source file of dll -> MiniMalWX/ dllmain.cpp MinimalWx.cpp > Minimalwx.h > > *source file of exe -> AppliDllWx/ main.cpp with exe file in > debugwx29 > *the dll ->debugwx29/minimalwx.dll > *scilab toolbox ->debugwx29/wxMinimal with two scilab functions > WxRun() and WxThread() > > WxRun() is Ok and WxThread() crashed > > exe file AppliDllWx load dll and call first TestScilabRun and > TestScilabThread. This exe file needs wxwidgets dll. > > Thanks for your interest > > Allan CORNET a ?crit : >> Hi, >> >> Please attach a mininal example. >> >> Thanks you >> >> Allan CORNET >> >> -----Message d'origine----- >> De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] >> Envoy? : mercredi 4 novembre 2009 13:02 ? : dev at lists.scilab.org Objet >> : [Scilab-Dev] DLL thread wxWidgets >> >> Hi, >> >> I have written a dll which launch wxMinimal a wxWidgets sample. >> In my project executable file load dll and can call a proc which run a >> wxMinimal. This proc can be a thread or not. >> With my own application everything is ok using a thread or not. >> >> Now if I use scilab with commands link and call without thread >> everything is fined and with a thread scilab crashed with Acces >> Violation at line 476 of toplevel.cpp (wxWidgets source file). >> I have check that if thread is empty scibab doesn't crached. >> I don't understand what's wrong. >> >> Any idea? >> >> Laurent >> >> >> >> >> > From allan.cornet at scilab.org Thu Nov 5 15:57:14 2009 From: allan.cornet at scilab.org (Allan CORNET) Date: Thu, 5 Nov 2009 15:57:14 +0100 Subject: [Scilab-Dev] DLL thread wxWidgets In-Reply-To: <4AF2E3AF.7060007@univ-lemans.fr> References: <4AF16D40.80606@univ-lemans.fr> <001501ca5d49$588a12c0$099e3840$@cornet@scilab.org> <4AF1764E.8080304@univ-lemans.fr> <004d01ca5e01$7494d690$5dbe83b0$@cornet@scilab.org> <4AF2E3AF.7060007@univ-lemans.fr> Message-ID: <006201ca5e28$42a060b0$c7e12210$@cornet@scilab.org> Hi, I used visual studio project of : wxWidgets-2.9.0\samples\minimal (modified to do a dll) See attachment I rebuilt wxWidgets-2.9.0 on Windows 7 (maybe the reason of these dependencies) Checked with scilab current master (32 bits) and 5.1.1 On XP and 7even Allan -----Message d'origine----- De?: Laurent Berger [mailto:laurent.berger at univ-lemans.fr] Envoy??: jeudi 5 novembre 2009 15:40 ??: dev at lists.scilab.org Objet?: Re: [Scilab-Dev] DLL thread wxWidgets Hi, Program is still crashed using your sourcefile in scilab with a call("TestScilabThread"). With a windows program it's OK with minimalwx.dll and minimal.dll. (With your dll some dll cannot be loaded like. With dependacy.exe wer.dll and ieshims.dll are missing. I don't understand why) Size of your dll is 119296 (minimal.dll) and my dll (MinimalWX.dll) 94208 bytes. What are your compile options? my scilab version is scilab-5.1.1 (Apr 14 2009) Best regards Laurent Berger Allan CORNET a ?crit : > Hi, > > This example seems to work (see minimal.cpp) with a dedicated thread for > wxwidgets. > > link("Minimal.dll","TestScilabThread","c"); > call("TestScilabThread") > > Best Regards > > Allan > > -- > ------------------------- > Allan CORNET > ------------------------- > The Scilab Consortium > Digiteo > Domaine de Voluceau > Rocquencourt - B.P. 105 > 78153 Le Chesnay Cedex > France > > -----Message d'origine----- > De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] > Envoy? : mercredi 4 novembre 2009 13:41 > ? : dev at lists.scilab.org > Objet : Re: [Scilab-Dev] DLL thread wxWidgets > > In attached zip file you will find > *source file of dll -> MiniMalWX/ dllmain.cpp MinimalWx.cpp > Minimalwx.h > > *source file of exe -> AppliDllWx/ main.cpp with exe file in > debugwx29 > *the dll ->debugwx29/minimalwx.dll > *scilab toolbox ->debugwx29/wxMinimal with two scilab functions > WxRun() and WxThread() > > WxRun() is Ok and WxThread() crashed > > exe file AppliDllWx load dll and call first TestScilabRun and > TestScilabThread. This exe file needs wxwidgets dll. > > Thanks for your interest > > Allan CORNET a ?crit : >> Hi, >> >> Please attach a mininal example. >> >> Thanks you >> >> Allan CORNET >> >> -----Message d'origine----- >> De : Laurent Berger [mailto:laurent.berger at univ-lemans.fr] >> Envoy? : mercredi 4 novembre 2009 13:02 ? : dev at lists.scilab.org Objet >> : [Scilab-Dev] DLL thread wxWidgets >> >> Hi, >> >> I have written a dll which launch wxMinimal a wxWidgets sample. >> In my project executable file load dll and can call a proc which run a >> wxMinimal. This proc can be a thread or not. >> With my own application everything is ok using a thread or not. >> >> Now if I use scilab with commands link and call without thread >> everything is fined and with a thread scilab crashed with Acces >> Violation at line 476 of toplevel.cpp (wxWidgets source file). >> I have check that if thread is empty scibab doesn't crached. >> I don't understand what's wrong. >> >> Any idea? >> >> Laurent >> >> >> >> >> > -------------- next part -------------- A non-text attachment was scrubbed... Name: minimal_vc9.vcproj Type: application/xml Size: 20655 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Capture.JPG Type: image/jpeg Size: 108034 bytes Desc: not available URL: From Serge.Steer at scilab.org Thu Nov 5 16:38:00 2009 From: Serge.Steer at scilab.org (Serge Steer) Date: Thu, 05 Nov 2009 16:38:00 +0100 Subject: [Scilab-Dev] Problem using double API In-Reply-To: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> References: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> Message-ID: <4AF2F158.4090909@scilab.org> In my opinion the code should be (createMatrixOfDouble allocates the array and returns a pointer) #include "stack-c.h" #include "api_scilab.h" #include "Scierror.h" int oclexec(char* fname) { SciErr sciErr; int *pB; int rowsB,colsB; double* mB; int k; sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); if(sciErr.iErr) { printError(&sciErr, 0); return 0; } for(k=0;k > Hi, > > I don't know if it's a bug or if something is missing in my code... > The issue is : using the following code, I cannot get scilab to > display the > mB matrix : > > #include "stack-c.h" > #include "api_scilab.h" > #include "Scierror.h" > > int oclexec(char* fname) > { > SciErr sciErr; > int *pB; > int rowsB,colsB; > double* mB; > int k; > > mB=(double*)malloc(rowsB*colsB*sizeof(double)); > > > for(k=0;k { > mB[k]=1.0; > } > > sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); > if(sciErr.iErr) > { > printError(&sciErr, 0); > return 0; > } > > LhsVar(1)=Rhs+1; > > return 0; > } > > The gateway is compiling fine, but when I type B=ocltest(A); , the B > matrix > is equal to A, whereas it is expected to be a ones matrix. > > On the other hand, the unary test double_writing_api.tst works. > I'am currently using the windows 1257195239 build (taken from the nightly > build of scilab 5.2) > > Regards, > Vincent > From pierre.marechal at scilab.org Thu Nov 5 16:55:18 2009 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Thu, 05 Nov 2009 16:55:18 +0100 Subject: Scilab 5.2 Beta available Message-ID: <4AF2F566.6010704@scilab.org> Hello, The Scilab Consortium announces the release of the first Scilab 5.2.0 Beta version. >> http://www.scilab.org/download/index_download.php?page=5.2.0-beta-1 Changes between Scilab 5.1.1 & Scilab 5.2.0 beta 1 : >> http://www.scilab.org/download/5.2.0-beta-1/CHANGES_5.2.0-beta-1.pdf The release notes : >> http://www.scilab.org/download/index_download.php?page=RELEASE_NOTES_5.2.0-beta-1 Do not hesitate to send us your comments and feedbacks on this version through the Scilab bug tracker . Pierre -- ============================================== Pierre MARECHAL Ing?nieur Support et D?veloppements ------ Consortium Scilab Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex ------ T?l : +33.1.39.63.56.81 ============================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent.lejeune at u-psud.fr Thu Nov 5 22:59:36 2009 From: vincent.lejeune at u-psud.fr (Vincent Lejeune) Date: Thu, 5 Nov 2009 22:59:36 +0100 Subject: [Scilab-Dev] Problem using double API In-Reply-To: <4AF2F158.4090909@scilab.org> References: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> <4AF2F158.4090909@scilab.org> Message-ID: It didn't work either, but I have find out what was breaking the code : I was using the same name for scilab function and c-function, that is to say, ocltest in Scilab called the ocltest C-function. When I changed the name in my builder_gateway_c.sce file, everything worked as expected. However I don't know if this can be considered as a "bug" , as this kind of naming convention is somehow not recommanded... Le Jeu 5 novembre 2009 16:38, Serge Steer a ?crit : > In my opinion the code should be (createMatrixOfDouble allocates the > array and returns a pointer) > > #include "stack-c.h" > #include "api_scilab.h" > #include "Scierror.h" > > > int oclexec(char* fname) { > SciErr sciErr; > int *pB; int rowsB,colsB; double* mB; int k; > > sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); > if(sciErr.iErr) { > printError(&sciErr, 0); return 0; } > > > for(k=0;k mB[k]=1.0; } > LhsVar(1)=Rhs+1; > > > return 0; } > > > > > Vincent Lejeune a ?crit : > >> >> Hi, >> >> >> I don't know if it's a bug or if something is missing in my code... >> The issue is : using the following code, I cannot get scilab to >> display the mB matrix : >> >> #include "stack-c.h" >> #include "api_scilab.h" >> #include "Scierror.h" >> >> >> int oclexec(char* fname) { >> SciErr sciErr; >> int *pB; int rowsB,colsB; double* mB; int k; >> >> mB=(double*)malloc(rowsB*colsB*sizeof(double)); >> >> >> for(k=0;k> mB[k]=1.0; } >> >> >> sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); >> if(sciErr.iErr) { >> printError(&sciErr, 0); return 0; } >> >> >> LhsVar(1)=Rhs+1; >> >> >> return 0; } >> >> >> The gateway is compiling fine, but when I type B=ocltest(A); , the B >> matrix is equal to A, whereas it is expected to be a ones matrix. >> >> On the other hand, the unary test double_writing_api.tst works. >> I'am currently using the windows 1257195239 build (taken from the >> nightly build of scilab 5.2) >> >> Regards, >> Vincent >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent.lejeune at institutoptique.fr Thu Nov 5 08:43:40 2009 From: vincent.lejeune at institutoptique.fr (Vincent Lejeune) Date: Thu, 5 Nov 2009 08:43:40 +0100 Subject: Problem using double API Message-ID: Hi, I don't know if it's a bug or if something is missing in my code... The issue is : using the following code, I cannot get scilab to display the mB matrix : #include "stack-c.h" #include "api_scilab.h" #include "Scierror.h" int oclexec(char* fname) { SciErr sciErr; int *pB; int rowsB,colsB; double* mB; int k; mB=(double*)malloc(rowsB*colsB*sizeof(double)); for(k=0;k References: <43a91f5ba05d475a1269d633f271fc29.squirrel@webmail.u-psud.fr> <4AF2F158.4090909@scilab.org> Message-ID: <56EA91AC2BE648318496C887BDB574C4@PCdevlj> It doesn't work either... Just in case, the builder_gateway_c.sce script is the following (taken from the skeleton toolbox) : if MSDOS then // to manage long pathname includes_src_c = '-I""' + get_absolute_file_path('builder_gateway_c.sce') + '../../include/""'; else includes_src_c = '-I' + get_absolute_file_path('builder_gateway_c.sce') + '../../include/'; end // PutLhsVar managed by user in sci_sum and in sci_sub // if you do not this variable, PutLhsVar is added // in gateway generated (default mode in scilab 4.x and 5.x) WITHOUT_AUTO_PUTLHSVAR = %t; tbx_build_gateway('ocl',.. ['ocltest','oclexec';.. 'gInit','gInit'],.. ['sci_simd.c'], .. get_absolute_file_path('builder_gateway_c.sce'), .. ['../../simd'],'',includes_src_c); clear WITHOUT_AUTO_PUTLHSVAR; clear tbx_build_gateway; -------------------------------------------------- From: "Serge Steer" Sent: Thursday, November 05, 2009 4:38 PM To: Subject: Re: [Scilab-Dev] Problem using double API > In my opinion the code should be (createMatrixOfDouble allocates the > array and returns a pointer) > > #include "stack-c.h" > #include "api_scilab.h" > #include "Scierror.h" > > int oclexec(char* fname) > { > SciErr sciErr; > int *pB; > int rowsB,colsB; > double* mB; > int k; > > sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); > if(sciErr.iErr) > { > printError(&sciErr, 0); > return 0; > } > > for(k=0;k { > mB[k]=1.0; > } > LhsVar(1)=Rhs+1; > > return 0; > } > > > > Vincent Lejeune a ?crit : >> >> Hi, >> >> I don't know if it's a bug or if something is missing in my code... >> The issue is : using the following code, I cannot get scilab to >> display the >> mB matrix : >> >> #include "stack-c.h" >> #include "api_scilab.h" >> #include "Scierror.h" >> >> int oclexec(char* fname) >> { >> SciErr sciErr; >> int *pB; >> int rowsB,colsB; >> double* mB; >> int k; >> >> mB=(double*)malloc(rowsB*colsB*sizeof(double)); >> >> >> for(k=0;k> { >> mB[k]=1.0; >> } >> >> sciErr = createMatrixOfDouble(pvApiCtx, Rhs + 1, rowsB, colsB,mB ); >> if(sciErr.iErr) >> { >> printError(&sciErr, 0); >> return 0; >> } >> >> LhsVar(1)=Rhs+1; >> >> return 0; >> } >> >> The gateway is compiling fine, but when I type B=ocltest(A); , the B >> matrix >> is equal to A, whereas it is expected to be a ones matrix. >> >> On the other hand, the unary test double_writing_api.tst works. >> I'am currently using the windows 1257195239 build (taken from the nightly >> build of scilab 5.2) >> >> Regards, >> Vincent >> > > From kirill.kirillov at gmail.com Thu Nov 5 23:12:07 2009 From: kirill.kirillov at gmail.com (Kirill Kirillov) Date: Fri, 06 Nov 2009 01:12:07 +0300 Subject: [Scilab-Dev] Questions about Xpad menu actions Message-ID: <1257459127.4155.4.camel@faith.universe> Hi! I'm sorry to trouble you, but I hope this one is a right mailing list to ask some questions about Xpad menu actions: 1. What does "Edit -> Indent" action do? 2. What does "Document -> Colorize" action do? May be it should be implemented as check box and then it should enable/disable coloring? 3. I think that it would be more usable if "Execute -> Execute File into Scilab" action saved file automatically instead of opening two dialogs. -- Kirill Kirillov <>< From sylvestre.ledru at scilab.org Thu Nov 5 23:16:11 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Thu, 05 Nov 2009 23:16:11 +0100 Subject: [Scilab-Dev] Questions about Xpad menu actions In-Reply-To: <1257459127.4155.4.camel@faith.universe> References: <1257459127.4155.4.camel@faith.universe> Message-ID: <1257459371.7028.1603.camel@zlarin> Hello, Le vendredi 06 novembre 2009 ? 01:12 +0300, Kirill Kirillov a ?crit : > Hi! > > I'm sorry to trouble you, but I hope this one is a right mailing list > to ask some questions about Xpad menu actions: > 1. What does "Edit -> Indent" action do? Type some text, select all, Edit -> Indent shoud reorganize your code for example if 1==1 then a=2 end will change your text for if 1==1 then a=2 end > 2. What does "Document -> Colorize" action do? Should apply some colors on the variable, function, keywords, etc. > May be it should be > implemented as check box and then it should enable/disable coloring? You are right. > 3. I think that it would be more usable if "Execute -> Execute File into > Scilab" action saved file automatically instead of opening two dialogs. Why not but save automatically to where ? By the way, Xpad was just an internal call name. Now, it is just "Editor" (note the originality ;) and there are still some rather important bugs in this editor. Sylvestre -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From kirill.kirillov at gmail.com Fri Nov 6 00:40:12 2009 From: kirill.kirillov at gmail.com (Kirill Kirillov) Date: Fri, 06 Nov 2009 02:40:12 +0300 Subject: [Scilab-Dev] Questions about Xpad menu actions In-Reply-To: <1257459371.7028.1603.camel@zlarin> References: <1257459127.4155.4.camel@faith.universe> <1257459371.7028.1603.camel@zlarin> Message-ID: <1257464412.4155.75.camel@faith.universe> Hi! Thanks for your answers. ? ???, 05/11/2009 ? 23:16 +0100, Sylvestre Ledru ?????: > Le vendredi 06 novembre 2009 ? 01:12 +0300, Kirill Kirillov a ?crit : > > 1. What does "Edit -> Indent" action do? > Type some text, select all, Edit -> Indent shoud reorganize your code > for example > if 1==1 then > a=2 > end > will change your text for > if 1==1 then > a=2 > end Then it should be named more obviously like "Reformat Selection" http://bugzilla.scilab.org/show_bug.cgi?id=5219 > > 2. What does "Document -> Colorize" action do? > Should apply some colors on the variable, function, keywords, etc. > > May be it should be > > implemented as check box and then it should enable/disable coloring? > You are right. http://bugzilla.scilab.org/show_bug.cgi?id=5220 > > 3. I think that it would be more usable if "Execute -> Execute File into > > Scilab" action saved file automatically instead of opening two dialogs. > Why not but save automatically to where ? Assume I've load rather large file into editor and introduced some changes. Now I want to test them before saving. Scipad can quickly load current unsaved version, but Scilab Editor can't: "Execute -> Load Into Scilab" takes ages to finish loading on my PC, because it prints every line in the file; (bug?) "Execute -> Execute File Into Scilab" needs saving file. Moreover "Save and Execute" button opens second dialog for saved file before. (bug?) > By the way, Xpad was just an internal call name. Now, it is just > "Editor" (note the originality ;) and there are still some rather > important bugs in this editor. IMHO, "Text Editor" would be better, there are many different editors even inside Scilab, e.g., Graphic (Figure, Axes) Editor, etc. -- Kirill Kirillov <>< From sumit.adhikari at gmail.com Fri Nov 6 07:26:42 2009 From: sumit.adhikari at gmail.com (Sumit Adhikari) Date: Fri, 6 Nov 2009 11:56:42 +0530 Subject: [Scilab-Dev] Scilab 5.2 Beta available In-Reply-To: <4AF2F566.6010704@scilab.org> References: <4AF2F566.6010704@scilab.org> Message-ID: Hello, Do this new version scilab requires GLIBC_2.7 ? It means we need to upgrade our OS to use it! /user/sa1/tools/scilab-5.2.0-beta-1/bin/scilab-bin: /lib64/tls/libc.so.6: version `GLIBC_2.7' not found (required by /user/sa1/tools/scilab-5.2.0-beta-1/lib/thirdparty/libhdf5.so.6) Regards, Sumit On Thu, Nov 5, 2009 at 9:25 PM, Pierre MARECHAL wrote: > Hello, > > The Scilab Consortium announces the release of the first Scilab 5.2.0 Beta > version. > >> http://www.scilab.org/download/index_download.php?page=5.2.0-beta-1 > > Changes between Scilab 5.1.1 & Scilab 5.2.0 beta 1 : > >> http://www.scilab.org/download/5.2.0-beta-1/CHANGES_5.2.0-beta-1.pdf > > The release notes : > >> > > http://www.scilab.org/download/index_download.php?page=RELEASE_NOTES_5.2.0-beta-1 > > Do not hesitate to send us your comments and feedbacks on this version > through the Scilab bug tracker . > > Pierre > > -- > ============================================== > Pierre MARECHAL > Ing?nieur Support et D?veloppements > ------ > Consortium Scilab > Digiteo > Domaine de Voluceau > Rocquencourt - B.P. 105 > 78153 Le Chesnay Cedex > ------ > T?l : +33.1.39.63.56.81 > ============================================== > > -- -------------------------------------------- Sumit Adhikari System Design Engineer austriamicrosystems AG Business Unit : Automotive Mob : 00-91-9885271710/00-91-9000161710 -------------- next part -------------- An HTML attachment was scrubbed... URL: From sebastien at celles.net Sat Nov 7 15:07:15 2009 From: sebastien at celles.net (sebastien at celles.net) Date: Sat, 7 Nov 2009 15:07:15 +0100 Subject: Scilab with Mac OS X 10.6.1 Message-ID: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> Hello, I'm trying to install Scilab with Mac OS X 10.6.1. It doesn't work at all when I'm clicking on Applications/scilab-5.1.1-alpha-1 (icon appear and disappear). Using the console I stroke the following command : $ /Applications/scilab-5.1.1-alpha-1.app/Contents/MacOS/bin/scilab-bin I get this error message : dyld: Library not loaded: /Users/pmarecha/compilation_chain_macosx/sci_binary/scilab-5.1.1-alpha-1.app/Contents/MacOS/lib/scilab/libscilab.0.dylib Referenced from: /Applications/scilab-5.1.1-alpha-1.app/Contents/ MacOS/bin/scilab-bin Reason: image not found Trace/BPT trap If someone can compile it again I could try it. Kind regards Sebastien From sylvestre.ledru at scilab.org Sat Nov 7 15:12:06 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Sat, 07 Nov 2009 15:12:06 +0100 Subject: [Scilab-Dev] Scilab with Mac OS X 10.6.1 In-Reply-To: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> References: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> Message-ID: <1257603126.7028.5132.camel@zlarin> Hello Sebastien, You have a few solutions: * Try: rm scilab-5.1-alpha-1.app/Contents/MacOS/lib/scilab/libstdc++.* or scilab-5.1-alpha-1.app/Contents/MacOS/lib/thirdparty/*libstdc++* * download the beta1 of Scilab 5.2.0 which is pretty stable and fixes some other issues under Mac OS X (like keyword shortcuts, docking system, etc). Sylvestre Le samedi 07 novembre 2009 ? 15:07 +0100, sebastien at celles.net a ?crit : > Hello, > > I'm trying to install Scilab with Mac OS X 10.6.1. > It doesn't work at all when I'm clicking on Applications/scilab-5.1.1-alpha-1 > (icon appear and disappear). > Using the console I stroke the following command : > $ /Applications/scilab-5.1.1-alpha-1.app/Contents/MacOS/bin/scilab-bin > > I get this error message : > > dyld: Library not loaded: > /Users/pmarecha/compilation_chain_macosx/sci_binary/scilab-5.1.1-alpha-1.app/Contents/MacOS/lib/scilab/libscilab.0.dylib > Referenced from: /Applications/scilab-5.1.1-alpha-1.app/Contents/ > MacOS/bin/scilab-bin > Reason: image not found > Trace/BPT trap > > If someone can compile it again I could try it. > > Kind regards > > Sebastien -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From s.celles at gmail.com Sat Nov 7 15:22:13 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sat, 7 Nov 2009 15:22:13 +0100 Subject: [Scilab-Dev] Scilab with Mac OS X 10.6.1 In-Reply-To: <1257603126.7028.5132.camel@zlarin> References: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> <1257603126.7028.5132.camel@zlarin> Message-ID: <8e9379c0911070622x4988a50dg1cff5d86dacce4cf@mail.gmail.com> On Sat, Nov 7, 2009 at 3:12 PM, Sylvestre Ledru wrote: > Hello Sebastien, > > You have a few solutions: > * Try: > rm scilab-5.1-alpha-1.app/Contents/MacOS/lib/scilab/libstdc++.* > or > scilab-5.1-alpha-1.app/Contents/MacOS/lib/thirdparty/*libstdc++* > * download the beta1 of Scilab 5.2.0 which is pretty stable and fixes some other issues under Mac OS X (like > keyword shortcuts, docking system, etc). > > Sylvestre Thanks But you shouldn't make the 5.1-alpha-1 for Mac OS X still available on the website because I don't think having an absolute path to Mr "pmarecha" /Users directory is a good idea. because the server is very slow (about 40 minutes to get scilab-5.1-alpha-1... and 40 minutes again to get scilab-5.2-beta-2 !!! Kind regards Sebastien From sylvestre.ledru at scilab.org Sat Nov 7 16:14:01 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Sat, 07 Nov 2009 16:14:01 +0100 Subject: [Scilab-Dev] Scilab with Mac OS X 10.6.1 In-Reply-To: <8e9379c0911070622x4988a50dg1cff5d86dacce4cf@mail.gmail.com> References: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> <1257603126.7028.5132.camel@zlarin> <8e9379c0911070622x4988a50dg1cff5d86dacce4cf@mail.gmail.com> Message-ID: <1257606841.7028.5320.camel@zlarin> Le samedi 07 novembre 2009 ? 15:22 +0100, s.celles at gmail.com a ?crit : > On Sat, Nov 7, 2009 at 3:12 PM, Sylvestre Ledru > wrote: > > Hello Sebastien, > > > > You have a few solutions: > > * Try: > > rm scilab-5.1-alpha-1.app/Contents/MacOS/lib/scilab/libstdc++.* > > or > > scilab-5.1-alpha-1.app/Contents/MacOS/lib/thirdparty/*libstdc++* > > * download the beta1 of Scilab 5.2.0 which is pretty stable and fixes some other issues under Mac OS X (like > > keyword shortcuts, docking system, etc). > > > > Sylvestre > > Thanks > But you shouldn't make the 5.1-alpha-1 for Mac OS X still available on > the website It works fine with Mac OS X 10.5 and it is specified on the website that it is only for this version. > because I don't think having an absolute path to Mr "pmarecha" /Users > directory is a good idea. It is just some buildtime information. It is not causing any issue. > because the server is very slow (about 40 minutes to get > scilab-5.1-alpha-1... and > 40 minutes again to get scilab-5.2-beta-2 !!! Where are you downloading it from ? It is usually very fast. Sylvestre > Kind regards > > Sebastien -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From s.celles at gmail.com Sat Nov 7 16:48:08 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sat, 7 Nov 2009 16:48:08 +0100 Subject: [Scilab-Dev] Scilab with Mac OS X 10.6.1 In-Reply-To: <1257606841.7028.5320.camel@zlarin> References: <8e9379c0911070607te733c88ja46c61e40c3c2738@mail.gmail.com> <1257603126.7028.5132.camel@zlarin> <8e9379c0911070622x4988a50dg1cff5d86dacce4cf@mail.gmail.com> <1257606841.7028.5320.camel@zlarin> Message-ID: <8e9379c0911070748l1dcce69cie0bae5e6507cba93@mail.gmail.com> >> because the server is very slow (about 40 minutes to get >> scilab-5.1-alpha-1... and >> 40 minutes again to get scilab-5.2-beta-2 !!! > Where are you downloading it from ? It is usually very fast. I'm downloading from http://www.scilab.org/download/5.2.0-beta-1/scilab-5.2.0-beta-1.dmg at arround 40kB/s with a 8Mb/s connection for myself it's very slow ! From sebastien at celles.net Sat Nov 7 18:45:17 2009 From: sebastien at celles.net (sebastien at celles.net) Date: Sat, 7 Nov 2009 18:45:17 +0100 Subject: Matrix viewer/editor Message-ID: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> Hello, I'd like to know if the Scilab dev team could do something that could be very usefull : a matrix viewer/editor. The key features could be : in the scilab command line writing view M (or viewmat M or show M) will lauch a GUI that could be usefull for viewing the M matrix (instead of using the text only user interface) moreover using an other command line (for example editmat M will lauch a GUI that could be usefull for editing the M matrix (instead of using the text only user interface) I know that I can live without such a feature but it will be easier to quickly have a look at a quite big matrix and editing it. An other idea could be to add a menu entry in "Applications" "Edit List or Matrix" will show a list of list or matrix variables (with size) currently used User could click on one variable to edit it. Some other features could also be useful such as being able of showing/editing several list or matrix next to each other Kind regards Sebastien From fvogelnew1 at free.fr Sat Nov 7 19:04:32 2009 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Sat, 07 Nov 2009 19:04:32 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> Message-ID: <4AF5B6B0.40209@free.fr> Try this: M=rand(10,5);editvar("M") F. sebastien at celles.net said on 07/11/2009 18:45: > Hello, > > I'd like to know if the Scilab dev team could do something that could be > very usefull : a matrix viewer/editor. > > The key features could be : > in the scilab command line writing > > view M > (or viewmat M or show M) > will lauch a GUI that could be usefull for viewing the M matrix > (instead of using > the text only user interface) > > moreover using an other command line (for example > editmat M > will lauch a GUI that could be usefull for editing the M matrix > (instead of using > the text only user interface) > > I know that I can live without such a feature but it will be easier to > quickly have > a look at a quite big matrix and editing it. > > An other idea could be to add a menu entry in "Applications" > "Edit List or Matrix" > will show a list of list or matrix variables (with size) currently used > User could click on one variable to edit it. > > Some other features could also be useful such as being able of showing/editing > several list or matrix next to each other > > Kind regards > > Sebastien > > From s.celles at gmail.com Sat Nov 7 20:05:15 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sat, 7 Nov 2009 20:05:15 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <4AF5B6B0.40209@free.fr> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> Message-ID: <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> On Sat, Nov 7, 2009 at 7:04 PM, Fran?ois Vogel wrote: > Try this: > > M=rand(10,5);editvar("M") >M=rand(10,5);editvar("M") !--error 4 Variable non d?finie: editvar From fvogelnew1 at free.fr Sun Nov 8 00:16:06 2009 From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=) Date: Sun, 08 Nov 2009 00:16:06 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> Message-ID: <4AF5FFB6.3040709@free.fr> s.celles at gmail.com said on 07/11/2009 20:05: >> M=rand(10,5);editvar("M") > !--error 4 > Variable non d?finie: editvar WFM on master ~6 weeks ago. What's your version? F. From vincent.lejeune at u-psud.fr Sat Nov 7 23:38:08 2009 From: vincent.lejeune at u-psud.fr (Vincent Lejeune) Date: Sat, 7 Nov 2009 23:38:08 +0100 Subject: Issue with CheckLhs/CheckRhs ? Message-ID: Hi, with the following code, I have a strange behavior (on Scilab 5.2 beta 1 windows)?: int sci_gpuInit(char* fname) { ?? CheckRhs(0,0); ?? CheckLhs(0,0); ? return 0; } when I called the function, I have the following errormessage : gInit() gInit : Mauvais nombre d'argument(s) de sortie : 0 attendu. A=gInit() gInit : Mauvais nombre d'argument(s) de sortie : 0 attendu. The last one is expectable...but the following code works : A=gInit without parentheses...But with only : gInit I have the same error as previously. Is it a bug ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From sylvestre.ledru at scilab.org Sun Nov 8 02:43:02 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Sun, 08 Nov 2009 02:43:02 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> Message-ID: <1257644582.7028.7217.camel@zlarin> Le samedi 07 novembre 2009 ? 20:05 +0100, s.celles at gmail.com a ?crit : > On Sat, Nov 7, 2009 at 7:04 PM, Fran?ois Vogel wrote: > > Try this: > > > > M=rand(10,5);editvar("M") > > >M=rand(10,5);editvar("M") > !--error 4 > Variable non d?finie: editvar It is probably because you are running Scilab under Mac OS X and Tcl/Tk support in Scilab does not work with this operating system and editvar is based on Tk :/ Sylvestre -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From s.celles at gmail.com Sun Nov 8 06:38:32 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sun, 8 Nov 2009 06:38:32 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <1257644582.7028.7217.camel@zlarin> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> Message-ID: <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> I thought that all the UI of Scilab 5 was made using Java. (no more Tcl/Tk because of its ugly look and feel) From yann.collette at scilab.org Sun Nov 8 08:36:39 2009 From: yann.collette at scilab.org (Yann Collette) Date: Sun, 08 Nov 2009 08:36:39 +0100 Subject: [Scilab-Dev] Issue with CheckLhs/CheckRhs ? In-Reply-To: References: Message-ID: <4AF67507.3010801@scilab.org> Hi, I thinks that the problem comes from the fact that it is not possible to have 0 Lhs. gInit() is equivalent to ans = gInit(). add sciprint('Lhs = %d Rhs = %s\n', Lhs, Rhs) To test and replace CheckLhs(0,0) by CheckLhs(0,1) YC Vincent Lejeune a ?crit : > Hi, > > with the following code, I have a strange behavior (on Scilab 5.2 beta > 1 windows) : > > int sci_gpuInit(char* fname) > { > CheckRhs(0,0); > CheckLhs(0,0); > > return 0; > } > > when I called the function, I have the following errormessage : > gInit() > gInit : Mauvais nombre d'argument(s) de sortie : 0 attendu. > A=gInit() > gInit : Mauvais nombre d'argument(s) de sortie : 0 attendu. > > The last one is expectable...but the following code works : > A=gInit > > without parentheses...But with only : > gInit > I have the same error as previously. > > Is it a bug ? > > > From stephane.mottelet at utc.fr Sun Nov 8 10:37:11 2009 From: stephane.mottelet at utc.fr (=?UTF-8?B?U3TCjsOpcGhhbmUgTW90dGVsZXQ=?=) Date: Sun, 08 Nov 2009 10:37:11 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> Message-ID: <4AF69147.6080204@utc.fr> s.celles at gmail.com a ?crit : > I thought that all the UI of Scilab 5 was made using Java. > (no more Tcl/Tk because of its ugly look and feel) > Latests versions of Tcl/Tk use Tile as theme engine, so the ugly look and feel is an old story now... Moreover, one of the advantages of Tk is the packer. There will never be such a widget placement algorithm with uicontrols, since Scilab has chosen the Matlab way; too bad :-( S. From s.celles at gmail.com Sun Nov 8 10:39:41 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sun, 8 Nov 2009 10:39:41 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <4AF69147.6080204@utc.fr> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> <4AF69147.6080204@utc.fr> Message-ID: <8e9379c0911080139o4dacccc1jb09d48a80e8146bd@mail.gmail.com> All right if Tk is not so ugly... but it should works "out of the box" for Mac OS X too... 2009/11/8 St??phane Mottelet : > s.celles at gmail.com a ?crit : >> >> I thought that all the UI of Scilab 5 was made using Java. >> (no more Tcl/Tk because of its ugly look and feel) >> > > Latests versions of Tcl/Tk use Tile as theme engine, so the ugly > look and feel is an old story now... Moreover, one of the advantages > of Tk is the packer. There will never be such a widget placement > algorithm with uicontrols, since Scilab has chosen the Matlab way; > too bad :-( > > S. > > From stephane.mottelet at utc.fr Sun Nov 8 10:46:06 2009 From: stephane.mottelet at utc.fr (=?UTF-8?B?U3TCjsOpcGhhbmUgTW90dGVsZXQ=?=) Date: Sun, 08 Nov 2009 10:46:06 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <8e9379c0911080139o4dacccc1jb09d48a80e8146bd@mail.gmail.com> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> <4AF69147.6080204@utc.fr> <8e9379c0911080139o4dacccc1jb09d48a80e8146bd@mail.gmail.com> Message-ID: <4AF6935E.20900@utc.fr> Sylvestre has already explained the problem : > There are some conflicts between the way the JVM and Tk are instantiated > in Scilab. > The JVM needs to be launched into a dedicated thread by the > CoreFoundation (Apple API). [1] This way blocks the Tk startup. > We spend some time on this issue but we don't know any fix for this. > Maybe there is... I will be happy to apply a patch for it. > > Now that on OSX Scilab does not user X11 any more, the problem does not seem so simple ! S. s.celles at gmail.com a ?crit : > All right if Tk is not so ugly... > but it should works "out of the box" for Mac OS X too... > > 2009/11/8 St??phane Mottelet : > >> s.celles at gmail.com a ?crit : >> >>> I thought that all the UI of Scilab 5 was made using Java. >>> (no more Tcl/Tk because of its ugly look and feel) >>> >>> >> Latests versions of Tcl/Tk use Tile as theme engine, so the ugly >> look and feel is an old story now... Moreover, one of the advantages >> of Tk is the packer. There will never be such a widget placement >> algorithm with uicontrols, since Scilab has chosen the Matlab way; >> too bad :-( >> >> S. >> >> >> From sylvestre.ledru at scilab.org Sun Nov 8 12:51:56 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Sun, 08 Nov 2009 12:51:56 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <4AF69147.6080204@utc.fr> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> <4AF69147.6080204@utc.fr> Message-ID: <1257681116.7028.7838.camel@zlarin> Le dimanche 08 novembre 2009 ? 10:37 +0100, St??phane Mottelet a ?crit : > s.celles at gmail.com a ?crit : > > I thought that all the UI of Scilab 5 was made using Java. > > (no more Tcl/Tk because of its ugly look and feel) > > > Latests versions of Tcl/Tk use Tile as theme engine, so the ugly > look and feel is an old story now... I though that using Tile involved a redevelopment of the Tk part of code ?! Sylvestre > > S. > -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From stephane.mottelet at utc.fr Sun Nov 8 13:10:41 2009 From: stephane.mottelet at utc.fr (=?UTF-8?B?U3TCjsOpcGhhbmUgTW90dGVsZXQ=?=) Date: Sun, 08 Nov 2009 13:10:41 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <1257681116.7028.7838.camel@zlarin> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> <4AF69147.6080204@utc.fr> <1257681116.7028.7838.camel@zlarin> Message-ID: <4AF6B541.1050406@utc.fr> Sylvestre Ledru a ?crit : > Le dimanche 08 novembre 2009 ? 10:37 +0100, St??phane Mottelet a ?crit : > >> s.celles at gmail.com a ?crit : >> >>> I thought that all the UI of Scilab 5 was made using Java. >>> (no more Tcl/Tk because of its ugly look and feel) >>> >>> >> Latests versions of Tcl/Tk use Tile as theme engine, so the ugly >> look and feel is an old story now... >> > I though that using Tile involved a redevelopment of the Tk part of > code ?! > Not sure. The Scicoslab team did it, I mean, Scicoslab uses the latest version of Tk, which uses Tile (at least for the OSX version). S. > Sylvestre > > >> S. >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.celles at gmail.com Sun Nov 8 14:14:45 2009 From: s.celles at gmail.com (s.celles at gmail.com) Date: Sun, 8 Nov 2009 14:14:45 +0100 Subject: [Scilab-Dev] Matrix viewer/editor In-Reply-To: <4AF6B541.1050406@utc.fr> References: <8e9379c0911070945g2773f5dak1b58e5c1eca58525@mail.gmail.com> <4AF5B6B0.40209@free.fr> <8e9379c0911071105u49908eaey5deaa226339082ef@mail.gmail.com> <1257644582.7028.7217.camel@zlarin> <8e9379c0911072138q6630e005nd378798e688b53c2@mail.gmail.com> <4AF69147.6080204@utc.fr> <1257681116.7028.7838.camel@zlarin> <4AF6B541.1050406@utc.fr> Message-ID: <8e9379c0911080514u5e5a3a72h75e127019b30d718@mail.gmail.com> > Not sure. The Scicoslab team did it, I mean, Scicoslab uses the latest > version of Tk, which > uses Tile (at least for the OSX version). I would like to add some comments about the matrix viewer/editor (I've reboot my Mac under Linux to have a look at "editvar") As I said previously, it could be usefull to be able to edit several matrix for example : a=rand(4,3) b=rand(4,1) editvar a b moreover it could be usefull to provide a way to edit hypermatrix (showing only two given dimensions) Kind regards From Samuel.Gougeon at univ-lemans.fr Mon Nov 9 19:04:06 2009 From: Samuel.Gougeon at univ-lemans.fr (Samuel Gougeon) Date: Mon, 09 Nov 2009 19:04:06 +0100 Subject: ATOMS categories Message-ID: <20091109190406.4872474zqrvbbc74@webmail2.univ-lemans.fr> Hello, Beside the compilation chain, one of the great new features of the new online toolboxes management system ATOMS is the ability to tag each toolbox with more than one category (see the "New toolbox" interface), and to use subcategories. However, they are a number of things that, IMO, would be worthwhile to tune for initializing the system before it gets crowdy. I would like to discuss them here, before any possible bug/request report. First, the "New toolbox" interface allows /any user/ to create a new category or sub-category. This feature sounds very hasardous. IMO, it should be a webmaster privilege, after discusion of categories here on users@ and with internal Consortium agreement. One may fear that indexation of modules in categories turns quite quickly messy ; because it would then be complicated to correct it afterwards, we should avoid this situation without waiting. A messy tool is no tool, so ignoring past efforts of everyone for keeping clear categories. Moereover, i do not feel really convinced by the new list of 17 defined categories and 9 subcategories (see list on page http://atoms.scilab.org/. Former page: http://www.scilab.org/contrib/index_contrib.php?page=download) An alternative is proposed herebelow. Leading numbers are references for the following discussion. (1) * Xcos (2) * Visualisation (graphics) (3) * User Interfaces (UI) - Editor styles - Graphical UI - Browsers - Other (4) * Instruments control - Data acquisition (drivers) - Remote control (drivers) - Real-time & misc. (5) * Data handling & processing - General - Strings & texts - Image - Sound - Databases (6) * Mathematics (7) * Advanced tools - Optimization - Modeling & control - Finite elements - Graphs & Networks - Advanced processing - Other (8) * Applied tools - Physics - Biology - Economy - Other (9) * System & dev - Operating Systems - File management - InterSoft - Parallelization - Other (10)* Education (11)* Documentation - Manuals - Tutorials & demos - Tools for docs ----- (1) Xcos: Since multi-tagging is possible, a "Manual" sub-category is useless (tagging Xcos + Manual) (2) Visualisation instead of Graphics, since GUI is also related to Graphics, while a specific box may be created for UIs. So, for GUI, Visualisation AND UI will be tagged. (3) UI. Not sure that sub-categories would be useful here. List given for mind mapping. (4) Instruments control: For uploading drivers for detectors & actuators. Tools ror instrumental communication protocol shall go to Realtime & misc box (5) * Data handling & processing: Sound, Strings & text processing shoult not be forgotten. (6) * Mathematics: Obviously, for general maths tools: Statistics, linear algebra, differential calculus, symbolic, geometry, topology, etc. Sub-categories might be defined afterwards, or not. (7) * Advanced tools: Neural networks => Advanced processing. Genetic algorithms & Simul. annealing => Optimization. (Why are there presently 5 sub-categories for Optim box ??? Will long stay with only few items for each one). (8) * Applied tools: Additional sub-categories might be defined later, with respect to the number and color of the Other-box' population. e.g., thinking about "Humanities" (Automatic language & speach processing => Other). (9) * System & dev: OS category shall not specifically target Windows, and shall become a sub-category here. "Parallelization" and "Benchmarks & Tests" might be skipped to join "Other". (10)* Education: Alternative: Could be considered as a sub-category of the "Applied tools" main box. (11)* Documentation: Not only manuals! Sub-categories are compulsory, because of language diversity! Filter on the language could be implemented later. What about the compilation chain here ? In the "New toolbox" interface, a language(s) field should be proposed, at least for modules of this box. Discussion and comments are welcome. Cheers Samuel From pierre.marechal at scilab.org Mon Nov 9 20:47:58 2009 From: pierre.marechal at scilab.org (Pierre MARECHAL) Date: Mon, 09 Nov 2009 20:47:58 +0100 Subject: [Scilab-Dev] ATOMS categories In-Reply-To: <20091109190406.4872474zqrvbbc74@webmail2.univ-lemans.fr> References: <20091109190406.4872474zqrvbbc74@webmail2.univ-lemans.fr> Message-ID: <4AF871EE.5010602@scilab.org> Hello, Samuel Gougeon wrote : > Hello, > > Beside the compilation chain, one of the great new features > of the new online toolboxes management system ATOMS is the ability > to tag each toolbox with more than one category (see the "New toolbox" > interface), and to use subcategories. > However, they are a number of things that, IMO, would be worthwhile > to tune for initializing the system before it gets crowdy. I would > like to discuss them here, before any possible bug/request report. ATOMS portal is not totally completed : - lack of documentation - Category list not totally established - Some ergonomic points ... We are aware of it. But obviously, a discussion about categories can't be useless > First, the "New toolbox" interface allows /any user/ to create a > new category or sub-category. This feature sounds very hasardous. > IMO, it should be a webmaster privilege, Totally agree with you. Actually, we have already planned to disable this feature. > after discusion of categories > here on users@ and with internal Consortium agreement. > One may fear that indexation of modules in categories turns > quite quickly messy ; because it would then be complicated to > correct it afterwards, No panic Samuel :-) : - A module can belongs to several categories - A module can be moved from a category to another - Categories can be renamed, moved or deleted ... without any consequence on the Atoms client fortunately ! > we should avoid this situation without waiting. > A messy tool is no tool, so ignoring past efforts of everyone > for keeping clear categories. > Moereover, i do not feel really convinced by the new list of 17 > defined categories and 9 subcategories > (see list on page http://atoms.scilab.org/. > Former page: > http://www.scilab.org/contrib/index_contrib.php?page=download) > An alternative is proposed herebelow. Leading numbers are references > for the following discussion. Don't forget : - The description of each categories - The priority : The alphabetic order is not always the best choice. Pierre -- ============================================== Pierre MARECHAL Ing?nieur Support et D?veloppements ------ Consortium Scilab Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex ------ T?l : +33.1.39.63.56.81 ============================================== From florian.de-vuyst at ecp.fr Tue Nov 10 10:06:55 2009 From: florian.de-vuyst at ecp.fr (Florian De Vuyst) Date: Tue, 10 Nov 2009 10:06:55 +0100 Subject: Delaunay 2D and 3D in scilab Message-ID: <4AF92D2F.5080606@ecp.fr> Hi, native delaunay 2D and 3D functions in scilab would be of great utility ! Of course, there is CGLAB but it is apparently no more maintained and doesn't work anymore under Windows XP, Cheers, florian de vuyst From c.fuenfzig at gmx.de Tue Nov 10 11:39:52 2009 From: c.fuenfzig at gmx.de (=?iso-8859-1?Q?=22Christoph_F=FCnfzig=22?=) Date: Tue, 10 Nov 2009 11:39:52 +0100 Subject: Multivariate Polynomials Message-ID: <20091110103952.108290@gmx.net> Hi all, I am new to Scilab and I would like to find out the best way to use or represent multivariate polynomials in Scilab :-) But I am a bit lost .. I found the site http://wiki.scilab.org/Memory_representation_of_variables#head-86f7f0f2b926076ca13eeb77d875f64c33036035 detailing Scilab's memory representation of variables "Matrix of polynomials". How is a single (univariate) polynomial stored ? What is a reasonable way to represent multivariate polynomials ? Thanks a lot for your help, Christoph -- Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 - sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser From Samuel.Gougeon at univ-lemans.fr Tue Nov 10 13:00:16 2009 From: Samuel.Gougeon at univ-lemans.fr (Samuel Gougeon) Date: Tue, 10 Nov 2009 13:00:16 +0100 Subject: [Scilab-Dev] Delaunay 2D and 3D in scilab In-Reply-To: <4AF92D2F.5080606@ecp.fr> References: <4AF92D2F.5080606@ecp.fr> Message-ID: <4AF955D0.90103@univ-lemans.fr> ----- Message d'origine ----- De : Florian De Vuyst Date : 10/11/2009 10:06: > Hi, > > native delaunay 2D and 3D functions in scilab would be of great utility ! > Of course, there is CGLAB but it is apparently no more maintained and > doesn't work anymore under Windows XP, > > Cheers, florian de vuyst Have you checked mesh2d(), genfac3d() and eval3dp() in Scilab ? Regards Samuel From stephane.mottelet at utc.fr Fri Nov 13 07:50:06 2009 From: stephane.mottelet at utc.fr (=?UTF-8?B?U3TCjsOpcGhhbmUgTW90dGVsZXQ=?=) Date: Fri, 13 Nov 2009 07:50:06 +0100 Subject: potential problems with Scilab's uicontrol Message-ID: <4AFD019E.4000703@utc.fr> Hi all, Since I am currently evaluating if it is worth uising Scilab's uicontrols (instead of Tk widgets), I would like to warn the Scilab team that these uicontrols have a strange non-standard behavior (with respect to any gui system) : once an uiconcontrol has been created, i.e. : f=gcf(); h=uicontrol(f,'style','frame'); the "position" field of the uicontrol changes each time the figure is resized. I am sorry to be so crude, but I cannot imagine that people who have decided that uicontrols should behave this way have ever programmed any serious gui :-) The problem is very simple : the origin of the screen coordinate system (the one for uicontrols) should be on the upper-left corner of the window (this is the case in Matlab). In Scilab, it is located in the lower-left corner : this makes all the stuff very tricky and/or unusable as an alternative to Tk. hth S. From sylvestre.ledru at scilab.org Wed Nov 18 17:37:44 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Wed, 18 Nov 2009 17:37:44 +0100 Subject: SEP #36: Pretty print Message-ID: <1258562264.5418.1575.camel@zlarin> Hello, This SEP details the evolution of texprint and the introduction of a prettyprint function. Don't hesitate if you have comments/remarks. Sylvestre -- ------------------------- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France -------------- next part -------------- A non-text attachment was scrubbed... Name: SEP_36_prettyprint.pdf Type: application/pdf Size: 385860 bytes Desc: not available URL: From dhirajmagare at gmail.com Thu Nov 19 19:34:29 2009 From: dhirajmagare at gmail.com (DHIRAJ MAGARE) Date: Fri, 20 Nov 2009 00:04:29 +0530 Subject: Regarding toolbox creation Message-ID: <5eeada680911191034j61238d08qe0f82e6db954eb79@mail.gmail.com> Hello Sir/Madam, My self is DHIRAJ, wants to create our own toolbox. With the help of given toolbox guide ( http://www.scilab.org/contrib/index_contrib.php?page=toolbox_guide ), I have created all files directories. But I could not get proper visualization about my toolbox. I stuck at very basic thing. Might be for you is very less valuable or minor. My sincere request to you please help me out into this development of my toolbox. Query: Before sending my toolbox zip file under 'SCILAB contribution' procedure, if I want to check whether my toolbox is properly working or not, by running at my end. How can complile my toolbox? At present, can I get my toolbox in the list of existing toolbox? How? Egerosly wating for your vital response for my progress. -- Regards DHIRAJ B. MAGARE -------------- next part -------------- An HTML attachment was scrubbed... URL: From sylvestre.ledru at scilab.org Tue Nov 24 13:47:38 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Tue, 24 Nov 2009 13:47:38 +0100 Subject: [Scilab-Dev] jogl library problems. In-Reply-To: <1256887618.4859.9089.camel@zlarin> References: <1256887618.4859.9089.camel@zlarin> Message-ID: <1259066858.10460.7.camel@korcula.inria.fr> Le vendredi 30 octobre 2009 ? 08:26 +0100, Sylvestre Ledru a ?crit : > Le jeudi 29 octobre 2009 ? 22:13 -0300, Jonathan Blanchard a ?crit : > > checking for glTexParameterf in -ljogl... no > > configure: error: libjogl: Library missing (Cannot find symbol > > glTexParameterf). Check if libjogl - C/Java (JNI) interface for JOGL - > > is installed and if the version is correct. Note that you might have > > to update etc/librarypath.xml to provide the actual path the the JNI > > libraries. > > > > Considering that it was working before did Scilab changed the version > > of JOGL required? I tried both 1.1.1a and the 2.0 beta (or something > > since their website is confusing). > It is just that, on startup, I am checking also the jni library of jogl. > libjogl should be available to your linker. Jonathan, I don't know if you fixed the issue but I just encounter a similar problem. When libjogl.so is not explicitly linked against libGL.so, the configure fails to detect libjogl.so. I fixed in the 5.2 branch by adding -lGL to the LDFLAGS when checking the usability of libjogl.so Sylvestre -- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France From BlanchardJ at ieee.org Tue Nov 24 14:07:28 2009 From: BlanchardJ at ieee.org (Jonathan Blanchard) Date: Tue, 24 Nov 2009 09:07:28 -0400 Subject: [Scilab-Dev] jogl library problems. In-Reply-To: <1259066858.10460.7.camel@korcula.inria.fr> References: <1256887618.4859.9089.camel@zlarin> <1259066858.10460.7.camel@korcula.inria.fr> Message-ID: On Tue, Nov 24, 2009 at 8:47 AM, Sylvestre Ledru wrote: > Le vendredi 30 octobre 2009 ? 08:26 +0100, Sylvestre Ledru a ?crit : > >> Le jeudi 29 octobre 2009 ? 22:13 -0300, Jonathan Blanchard a ?crit : > >> > checking for glTexParameterf in -ljogl... no >> > configure: error: libjogl: Library missing (Cannot find symbol >> > glTexParameterf). Check if libjogl - C/Java (JNI) interface for JOGL - >> > is installed and if the version is correct. Note that you might have >> > to update etc/librarypath.xml to provide the actual path the the JNI >> > libraries. >> > >> > Considering that it was working before did Scilab changed the version >> > of JOGL required? I tried both 1.1.1a and the 2.0 beta (or something >> > since their website is confusing). >> It is just that, on startup, I am checking also the jni library of jogl. >> libjogl should be available to your linker. > Jonathan, I don't know if you fixed the issue but I just encounter a > similar problem. > When libjogl.so is not explicitly linked against libGL.so, the configure > fails to detect libjogl.so. I fixed in the 5.2 branch by adding -lGL to > the LDFLAGS when checking the usability of libjogl.so > > Sylvestre > > > -- > Sylvestre Ledru > ------------------------- > The Scilab Consortium > Digiteo > Domaine de Voluceau > Rocquencourt - B.P. 105 > 78153 Le Chesnay Cedex > France > > > > Oh, Well I initially thought that it was a Solaris specific issue so I patched configure.ac to include -lGL on my local copy. I also modified the result of the test to include -lGL at build time but I am not sure it is necessary nor desired. Excellent timing on that one though since I made that fix just last night. Jonathan Blanchard From vincent.couvert at scilab.org Tue Nov 24 15:43:54 2009 From: vincent.couvert at scilab.org (Vincent COUVERT) Date: Tue, 24 Nov 2009 15:43:54 +0100 Subject: [Scilab-Dev] potential problems with Scilab's uicontrol In-Reply-To: <4AFD019E.4000703@utc.fr> References: <4AFD019E.4000703@utc.fr> Message-ID: <4B0BF12A.2040602@scilab.org> Hi St?phane, Let's take an example: f = figure(); uicontrol('Parent',f,'Position',[10 10 40 20],'string','test') This code gives the same result under Scilab 4.1.2, Scilab 5.1.1 and Matlab (the reference for uicontrol position is the bottom left corner of the figure). The problem occurs when the user resizes the figure: - Matlab: the uicontrol position does not change - Scilab 4.1.2: the uicontrol position changes on screen but not in its properties (too bad!) - Scilab 5.1.1 the uicontrol position changes on screen and in its properties The problem we have since the beginning of uicontrols in Scilab in this change of position according to the figure resize. We should perhaps consider to change this behavior to be consistent with Matlab one but we have first to check if this does not produce other problems. I just reported a bug (http://bugzilla.scilab.org/show_bug.cgi?id=5477) so that we investigate the problem. Vincent St??phane Mottelet a ?crit : > Hi all, > > Since I am currently evaluating if it is worth uising Scilab's > uicontrols (instead of Tk widgets), I would like to warn the > Scilab team that these uicontrols have a strange non-standard > behavior (with respect to any gui system) : once an > uiconcontrol has been created, i.e. : > > f=gcf(); > h=uicontrol(f,'style','frame'); > > the "position" field of the uicontrol changes each time the figure is > resized. > I am sorry to be so crude, but I cannot imagine that people who have > decided that uicontrols should behave this way have ever programmed > any serious gui :-) > > The problem is very simple : the origin of the screen coordinate system > (the one for uicontrols) should be on the upper-left corner of the window > (this is the case in Matlab). > In Scilab, it is located in the lower-left corner : this makes all the > stuff very tricky and/or unusable as an alternative to Tk. > > hth > > S. > From sylvestre.ledru at scilab.org Tue Nov 24 16:24:34 2009 From: sylvestre.ledru at scilab.org (Sylvestre Ledru) Date: Tue, 24 Nov 2009 16:24:34 +0100 Subject: [Scilab-Dev] SEP #36: Pretty print In-Reply-To: <1258562264.5418.1575.camel@zlarin> References: <1258562264.5418.1575.camel@zlarin> Message-ID: <1259076274.10460.72.camel@korcula.inria.fr> > This SEP details the evolution of texprint and the introduction of a prettyprint function. > Don't hesitate if you have comments/remarks. Here is a new version (1.1) of the SEP. Two changes: * Change in the order of the input argument: delim to becomes the third * screenshot added Sylvestre -- Sylvestre Ledru ------------------------- The Scilab Consortium Digiteo Domaine de Voluceau Rocquencourt - B.P. 105 78153 Le Chesnay Cedex France -------------- next part -------------- A non-text attachment was scrubbed... Name: SEP_36_prettyprint_1.1.pdf Type: application/pdf Size: 397910 bytes Desc: not available URL: