Re: [Yade-dev] PotentialBlocks & Particles documentation on website?
Bruno Chareyre said: (by the date of Mon, 11 Feb 2019 18:41:19 +0100) > > That's why I think we could do to enable FEMxDEM. Plus a bit of > > stuff in documentation about how to run those example FEMxDEM > > scripts. Which I still want to write ;) > Wait we diverged from PB to FEMxDEM. uh, right sorry :) python-escript is not necessary for compilation. So we can safely say that it's recommended. > IIRC PB needs not just python packages, it needs something at compile > time and that's where we have an issue isn't it? yes, exactly. PB requires stuff at compile time: apt install coinor-clp coinor-libclp-dev Boon also wrote about those two: coinor-libclp1 coinor-libosi1v5, but from what I see they are pulled in as dependencies. Then also: cmake -DENABLE_POTENTIAL_BLOCKS=ON -DENABLE_POTENTIAL_PARTICLES=ON -DCMAKE_INSTALL_PREFIX=../install ../trunk Janek ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
Re: [Yade-dev] PotentialBlocks & Particles documentation on website?
On 2/11/19 6:19 PM, Janek Kozicki wrote: Bruno Chareyre said: (by the date of Mon, 11 Feb 2019 17:59:57 +0100) And this is standard in debian? I'm not sure how to interpret your question :) The question was: can I "sudo apt-get install python-escript"? Missing recommended packages do not break yade package. python-escript is not necessary for any other part of yade to work. Only if you actually want to start a FEMxDEM .py script you get errors without this package. Also if python-escript is missing for some reason in the linux distribution, then yade packaage would NOT be broken, because it does not depend on it (it only recommends it). Oh, yes, good point. Since its only used via python there is no runtime error at startup if not present. And in fact that is the case with new stable debian release, I don't know why but python-escript is not there. Yeah, that was my point. :( Maybe it will be added there later. So if we configure yade.deb to recommend python-escript, then it would be a hint, that it's nice to install it, but it is not entirely necessary. Ok. That's what I think we could do to enable FEMxDEM. Plus a bit of stuff in documentation about how to run those example FEMxDEM scripts. Which I still want to write ;) Wait we diverged from PB to FEMxDEM. IIRC PB needs not just python packages, it needs something at compile time and that's where we have an issue isn't it? Sorry if I'm a bit lost... Bruno ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
Re: [Yade-dev] PotentialBlocks & Particles documentation on website?
Bruno Chareyre said: (by the date of Mon, 11 Feb 2019 17:59:57 +0100) > On 2/8/19 7:37 PM, Janek Kozicki wrote: > > Enabling FEMxDEM means only adding python-escript to recommended > > packages. > And this is standard in debian? I'm not sure how to interpret your question :) In any debian package you can set "depends" and "recommends" information. If you install a package it automatically pulls all packages listed in "depends". The "recommends" are automatically pulled only if user has configured his package manager to also pull recommendations. Missing recommended packages do not break yade package. python-escript is not necessary for any other part of yade to work. Only if you actually want to start a FEMxDEM .py script you get errors without this package. Also if python-escript is missing for some reason in the linux distribution, then yade packaage would NOT be broken, because it does not depend on it (it only recommends it). And in fact that is the case with new stable debian release, I don't know why but python-escript is not there. Maybe it will be added there later. So if we configure yade.deb to recommend python-escript, then it would be a hint, that it's nice to install it, but it is not entirely necessary. That's what I think we could do to enable FEMxDEM. Plus a bit of stuff in documentation about how to run those example FEMxDEM scripts. Which I still want to write ;) -- Janek Kozicki ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
Re: [Yade-dev] how about bringing MyTetrahedron example back to documentation?
Agreed, I just had the same conclusion this morning. So I will add some code-blocks in documentation :) This way an interested new developer will add these files himself by hand and practice compiling etc. Janek Bruno Chareyre said: (by the date of Mon, 11 Feb 2019 18:01:22 +0100) > Hi Janek, > I would go for the code block in rst. > Bruno > > > > On 2/8/19 9:34 PM, Janek Kozicki wrote: > > Back in the day I wrote this simple example for Vaclav about how to > > write new geometry, collision detection and constitutive laws. > > I think that Anton also studied this example. You can see it here: > > > >git co -b old-tetra f2ddab567 > > > >find -name "*MyTetrahedron*" > > > > So I was going through documentation and I saw this empty section, > > > > https://yade-dev.gitlab.io/trunk/tutorial-advanced.html#new-law > > > > and I suppose that it would be a nice place to put it there. > > > > I am not sure if I should just write .. code-block: inside .rst with > > explanations of this code. And no files in the directory tree, or > > maybe put it in /examples/ directory? > > > > But /examples/ contain only *.py code, and that would be a C++ code > > example. So to have it working actually I would need to put it > > in /pkg/example maybe? > > > > what do you think? > > > > > > PS: yeah I still want to make those videos from some of those new 215 > > examples that I checked last week, but I prefer to go from one > > chapter to next one, without skipping :) > > > > ___ > Mailing list: https://launchpad.net/~yade-dev > Post to : yade-dev@lists.launchpad.net > Unsubscribe : https://launchpad.net/~yade-dev > More help : https://help.launchpad.net/ListHelp -- Janek Kozicki http://janek.kozicki.pl/ | ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
Re: [Yade-dev] how about bringing MyTetrahedron example back to documentation?
Hi Janek, I would go for the code block in rst. Bruno On 2/8/19 9:34 PM, Janek Kozicki wrote: Back in the day I wrote this simple example for Vaclav about how to write new geometry, collision detection and constitutive laws. I think that Anton also studied this example. You can see it here: git co -b old-tetra f2ddab567 find -name "*MyTetrahedron*" So I was going through documentation and I saw this empty section, https://yade-dev.gitlab.io/trunk/tutorial-advanced.html#new-law and I suppose that it would be a nice place to put it there. I am not sure if I should just write .. code-block: inside .rst with explanations of this code. And no files in the directory tree, or maybe put it in /examples/ directory? But /examples/ contain only *.py code, and that would be a C++ code example. So to have it working actually I would need to put it in /pkg/example maybe? what do you think? PS: yeah I still want to make those videos from some of those new 215 examples that I checked last week, but I prefer to go from one chapter to next one, without skipping :) ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
Re: [Yade-dev] PotentialBlocks & Particles documentation on website?
On 2/8/19 7:37 PM, Janek Kozicki wrote: Enabling FEMxDEM means only adding python-escript to recommended packages. And this is standard in debian? B ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp