Re: [Yade-dev] Python 2 removal

Hi guys, for inspiration, you can check out commit https://github.com/woodem/woo/commit/e25623325d8fd0a5dcc9f94a7b78249e36bff449 and quite a few afterwards (between Oct 18 2015 and Nov 10 2015, some of them). The hardest part definitely was str/bytes distinction, plus testing. The more tests

Re: [Yade-dev] Fwd: [Yade-users] Critical bugfix for periodic boundaries (part 2)

> Hello Vaclav (Hi yade-devs), > I was wondering one thing when I checked again the periodic sorting. > I realized that there was a possibility for all bounds to have a > drift toward larger and larger positive values (not sure how it is in > Woo, I'm speaking of your implementation in yade). >

[Yade-dev] [Bug 1604266] Re: yade not working prpoperly on Ubuntu/Kubuntu 16.04

I am correcting what I said about IPython 5.0. The first version supporting qt5 is 3.x. 4.x will probably work with Yade, but with some warnings. 5.0 is not compatible due to API changes (custom prompt, shortcuts, eventloop integration). That is all unrelated to importing incorrect qt version in

[Yade-dev] [Bug 1604266] Re: yade not working prpoperly on Ubuntu/Kubuntu 16.04

FYI IPython packaged in 16.04 does not support qt5 yet, you need to install version >=5 from pip. -- You received this bug notification because you are a member of Yade developers, which is subscribed to Yade. https://bugs.launchpad.net/bugs/1604266 Title: yade not working prpoperly on

Re: [Yade-dev] Yade release, new versioning?

On 20.4.2016 08:35, Jan Stránský wrote: I am ok with .MM . What about just YY.MM , 16.04, like Ubuntu? That will fail being chronological at the turn of the century. v. ___ Mailing list:

[Yade-dev] [Bug 1544022] [NEW] ETA shows garbage like 4043:242618:08

Public bug reported: When I run a simulation which sets max number of steps, the UI shows garbage ETA (estimated time of arrival), as shown in the attachment. The script is fairly large, so I can't isolate MWE, but I guess it will apply to any simulation setting the number of steps beforehand.

Re: [Yade-dev] [Branch ~yade-pkg/yade/git-trunk] Rev 3709: add function has() to O.interactions to check if (id1, id2) exists + fix doc of O.interactions

revno: 3709 committer: Bruno Chareyre bruno.chare...@hmg.inpg.fr timestamp: Thu 2015-07-30 18:08:08 +0200 message: add function has() to O.interactions to check if (id1,id2) exists + fix doc of O.interactions You can also call it

Re: [Yade-dev] [Bug 1475844] Re: Swap problem in InteractionLoop

Yes, I understand the idea on performance. table[i][j]- is nearly as fast a p-. There an extra check that table element is something, but, as you say, it is very quick. It makes sense but it is pitty, it was quite a bit of work to implement the cache thing everywhere wasn't it? :) It is

[Yade-dev] [Bug 1475844] Re: Swap problem in InteractionLoop

Hi Bruno, I think the two-step solution is right as far as I can see. For the performance, I apologize for my incorrect statement; you're right about this. The logic is a bit different in Woo (https://github.com/eudoxos/woodem/blob/master/pkg/dem/ContactLoop.cpp) and I will check again what is

Re: [Yade-dev] [Branch ~yade-pkg/yade/git-trunk] Rev 3683: Non invasive refactoring of InsertionSortCollider

Remove unneeded struct, replace by a simple variables to escape overhead. Just a note, the struct removed is a bit field (https://en.wikipedia.org/wiki/Bit_field) so both booleans were stored in one byte; if they are replaced by two bools, they occupy 2 bytes, thus the overhead is

[Yade-dev] [Bug 1455621] Re: documentation bug: getting class name fails

It might have been removed. Those things are in lib/serialization/Serializable.cpp; I grepped the source for getClassName everywhere and it is not exposed to Python. gui/qt4/SerializableEditor.py uses __class__.__name__ everywhere. You can always define an alias by saying

Re: [Yade-dev] [Bug 1455621] [NEW] documentation bug: getting class name fails

Public bug reported: In the example code of section [1], class.name fails. I couldn't find the solution myself, still searching... [1] https://yade-dem.org/doc/prog.html?highlight=classname#run-time-type-identification-rtti use Sphere().__class__.__name__ v. -- You received this bug

Re: [Yade-dev] [Yade-users] Yade 1.14.0 released

On 21/04/15 22:46, Anton Gladky wrote: - Provide Python3 interface - Prepare for migration into Qt5 Python2 and Qt4 will probably soon removed from main distribution archives. Hi Anton, it seems that the libqglviewer is not co-installable for both qt4 and qt5, and is always built only

[Yade-dev] [Bug 1440887] Re: Buggy data in capillary files

Hi Jerome, is the code to generate those files in the github repo yet? Cheers, v. -- You received this bug notification because you are a member of Yade developers, which is subscribed to Yade. https://bugs.launchpad.net/bugs/1440887 Title: Buggy data in capillary files Status in Yet Another

Re: [Yade-dev] results scatter with identical initial simulation

Thanks everybody for opinions. Christian, you were right: I ran a non-trivial simulation (65k steps, perhaps 100k particles) and indeed *the results are exactly the same with OpenMP-less or OpenMP/1thread, and differ with 2 threads (and presumably more).* For illustration of the scatter,

[Yade-dev] results scatter with identical initial simulation

Hi there, I'm having an issue with results differing between initially identical simulations. I know we have two stock explanations: * the fairytale, authored by myself back then, that (force, ...) summation order is non-deterministic due to OpenMP (threads switching controlled by the

Re: [Yade-dev] results scatter with identical initial simulation

(sorry for PM) I assume you are running your simulation in parallel. If not identical simulations should give identical results (at least in Yade). That's what everybody says, but did you try to run the same sim and see? OK, I guess I go do that right now :) Will post back. v.

Re: [Yade-dev] (moved to yade-dev) Interaction of FrictMat and a new class of material

Hi, I did not read the whole thread, but in general, one has to use DECLARE_FUNCTOR_ORDER so that goReverse is ever used. Ig2_ functors never call goReverse, since the contact is reordered (swapped) in InteractionLoop so that it has geometries in the order handled by the functor; but

[Yade-dev] [Bug 1423130] Re: impossible to unselect bodies - blincking forever

Did you try to hit escape to unselect? Woo has that, and I think I added it to Yade already. -- You received this bug notification because you are a member of Yade developers, which is subscribed to Yade. https://bugs.launchpad.net/bugs/1423130 Title: impossible to unselect bodies - blincking

[Yade-dev] Sphinx's objects.inv not readable on the server

Hi there, I am using the intersphinx extension to link to Yade docs from within Woo docs. For that to work, the extension downloads inventory of Sphinx objects from http://www.yade-dem.org/doc/objects.inv . Recently, though, I am getting 403 Forbidden error. Coud I ask to have that file

[Yade-dev] [Bug 1400075] Re: O.forces.addMove does not work

Maybe an intermediate state of the tensile tests, where groups of particles had to be moved as rigid blocks - now done with constant velocity but maybe with constant disp/iter previously? Yes, that must have been it! It was for sure at the time where blocked DOFs would make the particle simply

[Yade-dev] [Bug 1400075] Re: O.forces.addMove does not work

Hi, the code was always working, and still should be, in NewtonIntegrator.pp:208 (and 2 other lines for rotations) I read if (scene-forces.getMoveRotUsed()) state-pos+=scene-forces.getMove(id); It was supposed to impose displacement or rotation change directly, and to do it in a way which is

[Yade-dev] [Bug 1398326] Re: boost/math/nonfinite_num_facets.hpp No such file or directory

The default boost version in precise was 1.46 though boost 1.48 was provided in the repositories with boost packages suffixed with that version. -- You received this bug notification because you are a member of Yade developers, which is subscribed to Yade. https://bugs.launchpad.net/bugs/1398326

Re: [Yade-dev] Parallel loops over interactions to sum

Hi Jerome, this has nothing to do with indeterminism (that has only minor effect, due to rounding), what you have is that the same variable gets written to from different threads, so they are overwriting each others values, something like this: 1. thread 1 reads the value A from memory,

Re: [Yade-dev] Editing values in inspect was annoying.

The proper fix would be to use isHot. The way you fixed this is that as long as the value in question is focused (even without being edited), it will not be updated. v. Hi, editing values in inspect was quite annoying - the value would reset for no reason while the cursor was in it. I think

[Yade-dev] slides from the conference?

Hey everybody, maybe I just missed those -- are presentations from the dev conference online already? Cheers, Václav ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe :

Re: [Yade-dev] (Anecdotic ?) Changes in Law2_.._Capillarity : H-M / C-S

One solution would be to use double inheritance, maybe. Not sure how boost::python would handle that. Boost::python is fine with that, but what would not work is all the YADE_CLASS_BASE_DOC_*** macros, serialization; that is both fixable, but most importantly, class indices and dispatching

[Yade-dev] [Bug 1041084] Re: memory leak while assigning boost::python attributes

@Bruno: unable to reproduce #11, tried this with Woo: from woo.dem import * e=FlowAnalysis() import sys for k in range(1): sys.stderr.write('.') e.flags=range(0,100) (FlowAnalysis.flags is declared as vectorint). Memory usage is sitting at 0.6% with no change at all. --

[Yade-dev] [Bug 1041084] Re: memory leak while assigning boost::python attributes

PS the python definition is probably [1] (that is, passing by value and not by reference, read-write and without the postLoad hook when modified), if you want to check it out. v. [1] http://bazaar.launchpad.net/~eudoxos/woo/trunk/view/head:/lib/object/Object.hpp#L161 -- You received this bug

[Yade-dev] [Bug 1349880] Re: Segfault on O.saveTmp() and O.loadTmp() if GUI is enabled.

Might be related to see https://svn.boost.org/trac/boost/ticket/8290. Basically if any python-instantiated object is destroyed from c++, there is a potential problem with crashing python by messing with its stuff without obtaining the global interpreter lock (GIL). You have to protect all those

[Yade-dev] EIGEN_NO_ALIGN

Hi Anton, you mentioned that you did some experimentation with vectorized (aligned) Eigen builds. I am interested to hear what you observed. I tried that some time ago and my findings are the following: 1. supporting aligned types with boost::python (in minieigen) is hard or impossible, see

[Yade-dev] [Bug 1267164] Re: Fails to build from yade-daily sources

Source package from the repo fails to build locally just the same as git: $ apt-get source --build yadedaily /home/eudoxos/build/yade-daily-8+3772+64~ubuntu13.10.1/py/wrapper/customConverters.cpp: In static member function ‘static PyObject* custom_OpenMPAccumulator_to_int::convert(const

[Yade-dev] [Bug 1267131] [NEW] FTBFS in _pulyhedra_utils.cpp

Public bug reported: Yade trunk (from github f9694ae7ecc8aa9b62a2735f86b0aa4bee1a69e9) fails to build from source on Ubuntu 13.10: [ 3%] Building CXX object gui/CMakeFiles/_GLViewer.dir/qt4/OpenGLManager.cpp.o /home/eudoxos/yade/trunk/py/_polyhedra_utils.cpp: In function ‘void

[Yade-dev] [Bug 1267164] [NEW] Fails to build from yade-daily sources

Public bug reported: I am using the yade-daily ppa with Ubuntu 13.10. When I run $ apt-get source --build yade-daily the build process continues normally but it fails towards the end: [ 96%] Building CXX object py/CMakeFiles/_customConverters.dir/wrapper/customConverters.cpp.o cd

[Yade-dev] [Bug 1267164] Re: Fails to build from yade-daily sources

Thanks, I tried this (changing the distribution to saucy in the sources.list) but it fails with The following packages have unmet dependencies: yadedaily : Depends: python-yadedaily (= 8+3750+64~ubuntu13.10.1) -- when investigate further, I need libyade, which is unavailable. I try with apt-get

[Yade-dev] [Bug 599743] Re: static class attributes are not documented by sphinx

** Changed in: yade Assignee: Václav Šmilauer (eudoxos) = (unassigned) -- You received this bug notification because you are a member of Yade developers, which is the registrant for Yade. https://bugs.launchpad.net/bugs/599743 Title: static class attributes are not documented by sphinx

Re: [Yade-dev] farewell

Sorry, but seems, you are going to make Yade as it was far in the past. No? Come on... I am not reverting all my commits. Is not the most promising pairing of different packages as, for example, LIGGGHTS-OpenFOAM, EDEM-FLUENT and so on? I am sure other people on the list know better -- and

Re: [Yade-dev] farewell

Some of your last developments are still largely unknown to me (namely the L3Geom series), how far are they validated and functional? They were mostly sketches. I will certainly build on them myself in the fork, because I consider tham as a step forward, but cannot recommend them for wide

Re: [Yade-dev] farewell

I hope, yade will survive without your contribution. http://www.youtube.com/watch?v=XM1fDkC4c0Q Are we having mailing-list funeral?! Cheer up, it is just a piece of code! Also it would be interesting to know about your new project. I will let you know on the list, but (if ever) it will not be

[Yade-dev] farewell

Dear colleagues, most open-source projects are either abandoned, or developed forever, without ever getting to the done state. Yade is widely used now and, in my eyes, fully functional and completed. I am happy about that. I decided to retire from the development, not wanting to stay in some

Re: [Yade-dev] perfomance benchmark

Cause is a non-parallelised InsertionSortCollider, who need about 80% time with 4 threads. The cause is not (only) that it is not parallelized, but that the algorithm scales worse than O(n). You see that for 1 thread collider's percentage gets up with increasing n. Parallelizing the collider

Re: [Yade-dev] perfomance benchmark

But scaling O(NlogN) is not so bad.. And perfomance can be significantly optimized now by playing with verletDist, I think. But it need some tries. You are between two influences; if verletDist is too big, then many potential interactions will exist, which slows down the InteractionLoop (many

[Yade-dev] [Bug 724396] Re: crash after add particles in parallel mode

I don't think so. I think sizeof(vectorT) is always sizeof(size_t)+sizeof(void*), the array is allocated on the heap dynamically. For the independent resizing: sounds fine, though sizes should be re-synced in sync() before the summation loop, otherwise it will run slower due to in-loop size

[Yade-dev] [Bug 721107] Re: InsertionSortCollider memory leaks

Anton, whatever Yade lets you do is a proper way of using it. Why is it not proper? -- You received this bug notification because you are a member of Yade developers, which is the registrant for Yade. https://bugs.launchpad.net/bugs/721107 Title: InsertionSortCollider memory leaks Status in

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2741: 1. LOG_INFO and LOG_WARN are disabled when log4cxx feature is disabled.

1. LOG_INFO and LOG_WARN are disabled when log4cxx feature is disabled. Anton, I would prefer if you keep those enabled. Or at least eplain why the log message you mentioned elsewhere makes you so unhappy? ___ Mailing list:

Re: [Yade-dev] buildbot

https://bugs.launchpad.net/yade/+bug/410250 (should be fine without log4cxx) I disabled automatic builds. Indeed, seems all tests are passed, but exit code is not 0... ___ Mailing list: https://launchpad.net/~yade-dev Post to :

Re: [Yade-dev] reversedForces and Co.

I was thinking about the meaning of bools like reversedForces, compressivePositive and so on. Recently I was quite confused about that and my point is: say that in our contact law we decide that compressive forces are positive. Then, why not be consistent and take compressive stresses also

Re: [Yade-dev] reversedForces and Co.

Goal stresses don't have to be changed, compressive stresses are always negative. The only thing that needs to be changed in scripts (still too much, I agree) is you need PeriTriaxController.revertSign=false with Dem3DofGeom laws, and PeriTriaxController.revertSign=true with ScGeom laws (L3

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2710: 1. Add plot.autoAddData with docs. A limitation is that = should not be in names of data column...

pprint(plot.data) #doctest: +ELLIPSIS + {'gravWork': [0.0, -25.13274...], +'i': [0, 1], +'kinRot': [0.0, 0.0], +'kinTrans': [0.0, 7.5398...], +'nonviscDamp': [0.0, 10.0530...], +'total energy': [0.0, -7.5398...]} As it is failed with last

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2710: 1. Add plot.autoAddData with docs. A limitation is that = should not be in names of data column...

Can you try 2716? Not tested, but could make a difference. If not, I don't know. v. ___ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help :

Re: [Yade-dev] weak/strong contacts

I hope Vaclav's function can Chiara's :-) ___ 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] read-only attribute

Hi (Vaclav) How to add a read-only attrubute to a class in order to I can read it from python? See https://www.yade-dem.org/sphinx/prog.html#yade-class-base-doc-macro-family and attribute flags. You can specify that the normal will be (Attr::readonly|Attr::noSave). That way, it will be

Re: [Yade-dev] read-only attribute

yes, I know. But if I create a surface using utils or geom module(or import it from stl or other place), I can not assure it. Seems, need to add a specialized function for this in yade.utils. You can fix those functions to orient facets consistently, although I believe that most of them do so

Re: [Yade-dev] strange behaviour

Who is working on dynamic problems? Could not you explain why it occurs (see attached script)? And do test the configuration with own laws? The interactions are not erased obviously, otherwise it works as expected. Since your law just tests penetrationDepth, then it could be a bug in ScGeom

Re: [Yade-dev] Google Summer of Code 2011

I am really interested in implementation of this feature as I was surprised of ESyS productivity, which uses the same technology. If I have enough time/resources I would try to do it. IMHO Esys-particles is so efficient because they are much less flexible. The downside of MPI is that it would

Re: [Yade-dev] Google Summer of Code 2011

4 is a prerequisite for 2, no? Practically yes, I think (though the current collider could also work per-domain efficiently, with a few additional lines). v. ___ Mailing list: https://launchpad.net/~yade-dev Post to :

Re: [Yade-dev] Google Summer of Code 2011

Well, I think it would be easier to start from very simple yade-mpi version. Without UI, OpenGL (only VTK) and python interface. Just to make it work. Then I would say it is easier to just download ESyS-particle...? What is Yade without Python? ___

Re: [Yade-dev] periodic boundary consolidation

Come on, dude... Did something change between in since midnight, such that your last 2 replies are so different? The first one was nice, but I have no intention to reply to your flames in the latter. I am glad that we more or less agreed about Cell, no need to spoil it.

Re: [Yade-dev] periodic boundary consolidation

1/ refH vs. trsf : do you agree that it is better to make them independant? 2/ refH can be kept in Cell if you prefer. Yes, yes. I was thinking that I take care of the first one, since I sort-of maintain the rendering part, but if you do it, no problem. 3/ I'll remove H0. It is just a

[Yade-dev] Google Summer of Code 2011

Hi there, is anyone interested in paid ($3500 IIRC) work on Yade during the summer? This should especially concern PhD students (who work over the summer anyway) with a topic that could be judged generally useful for all other users of Yade and for development of the project.

[Yade-dev] [Bug 707966] Re: Yade does not start on Natty 11.04

I have no idea what is it doing, perhaps post build-log so that we see what is being linked. Also, the output of ldd home/gladk/dem/yade/cleanComp/build/lib/yade-cleanComp/lib /lib_gts__python-module.so would help. I don't have natty, so I cannot test. -- You received this bug notification

Re: [Yade-dev] periodic boundary consolidation

Hi there, an explanation to the last commit: 1. I don't think size should be settable to set lengths of hSize. If you really need something like that, you're probably smart enough to change hSize directly. setBox was meant as name that explicitly sets axis-aligned box. 2. why removing

Re: [Yade-dev] SpherePadder

Sorry also for the late reply. Concerning SpherePadder, the code is into the attic folder !!??, is it a problem if it stays in this folder as the lattice ? Lattice has been in Hades for quite long time now. There is no attic. I made SpherePadder compile only with the sphere-padder feature

Re: [Yade-dev] periodic boundary consolidation

I don't see how setBox can help. Is somebody fool enough to (1) start by defining a non-aligned cell, then (2) make it aligned again? I don't think so. I don't know what you think, but I guess it is better to call it setBox, if it creates box... setBox defines the whole geometry, whereas

Re: [Yade-dev] periodic boundary consolidation

What I proposed previously was unnecessarily complicated, since reference hSize (Cell::getHSize0) can be computed from trsf and hSize, there is no need to keep track of it. The display refHSize is in the code, but it will not be used in scripts most the time. Commited in r2693, further

[Yade-dev] makeing yade single-threaded by default

Hi there, due to numerous performance problems when Yade uses the default number of threads (one thread per core), I think it would be better to make run single-threaded by default, so that unknowledgeable people are not hit. Those who know assign the right number of threads anyway. It

[Yade-dev] periodic boundary consolidation

Hi there (especially Bruno), I see that the current approach with hSize is more flexible and that we can do without refSize in most cases. Also, it is true that is makes more sense if trsf is just accumulator of velGrad which does not change geometry and can be changed by the user freely,

Re: [Yade-dev] making yade single-threaded by default

Thanks for feedback, it was done in r2692. V. Hi, In most of my simulations I use 1 thread/core and find it the most optimal solution. Also it helps to get completely identical results excluding openmp-instability. And only on a very large simulations (10^5 particles) it is reasonable to use

[Yade-dev] [Bug 707966] Re: Yade does not start on Natty 11.04

I adjusted the linking (yade-opengl is merged in yade-support), it should be solved now. I can't verify on natty, though, so I let you close if it is solved now. ** Changed in: yade Status: New = In Progress -- You received this bug notification because you are a member of Yade

[Yade-dev] [Bug 668329] Re: gts imports surfaces incorrectly (regression from 0.50)

Fixed in r2691 ** Changed in: yade Status: New = Fix Released ** Changed in: yade Assignee: (unassigned) = Václav Šmilauer (eudoxos) -- You received this bug notification because you are a member of Yade developers, which is the registrant for Yade. https://bugs.launchpad.net/bugs

Re: [Yade-dev] can't compile rev 2689

Hi Giulia, you should delete execCheck = ... line from scons.profile-default, then it will work. Sorry, I don't know why it got saved in that file in the first place, it should not. It is used b ythe --rebuild option, but that one should not save the profile. HTH, v. Hi all, I was working

Re: [Yade-dev] NozzleFactory

ps. What is SpheresPadder? An algorithm by J-F Jerier for generating stressfree packing based on tetrahedral mesh. I think we should remove it from the code, since there is no documentation, not a single example how to use it, and no-one using it. v

Re: [Yade-dev] NozzleFactory

Vaclav, do you have any thoughts on packing strategy? Well, I think the current state is satisfactory, no? There are (dense) packing generators in the pack module (regularHexa, regularOrtho, randomDensePack, randomPeriPack), which generate a SpherePack object with spheres, given a spatial

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2675: 1. Debian-directory cleaning and modifying (checked with lintian -IE --pedantic, no warnings at a...

Hi Anton, I was not following for a few days. Are we still packaging the -dbg things, as you removed the templates? I think they should be definitely in a separate package (since those files are huge). I know that debian has a special way of packaging debugging symbols, is that what we do

[Yade-dev] [Bug 706937] Re: StepDisplacer assumes use of NewtonIntegrator

Hi Jerome, you are welcome to update the documentation. Since non- dynamic bodies don't exist anymore, that distinction should disappear from the doc. The necessity of using NewotnIntegrator is connected to that, since it is only the intergrator that changes positions (based on velocities). You

[Yade-dev] [Bug 707966] Re: Yade does not start on Natty 11.04

I was compiling with 4.5 on maverick, and never had problems. I think the problem can be that yade-support is not linked with boost libs at all (...?). You can try to go to lib/SConscript:18, and change LIBS=['dl','m']+[] to LIBS=['dl','m']+[l for l in env['LIBS'] if l!='yade-support']

Re: [Yade-dev] NozzleFactory

I don't know much about SpheresPadder but it is used by Frédéric Donze and co-workers for generating rock mass IIRC (how do you evaluate what is used and what is not btw?). I guess based on the fact that I modified the class quite heavily in May (IIRC), and no-one complained. I tried to use it a

Re: [Yade-dev] Periodic simulations (continued)

(1) It is believed that the first method is generally more convenient, since it will let :yref:`Cell.trsf` reflect only the transformation produced during the simulation, independently of the initial period geometry. it is believed: who? We believe perhaps would be better, no? Or just that We

Re: [Yade-dev] Cell deformation handling

Hi Bruno, I think it makes sense what you write and it is good that you did the implementaiton that works. The fact that flipCell can be encapsulated inside the collider is nice, for instance. I would like to just ask for 2 things: 1. Make sure that the old interface (setting refSize etc)

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2653: - Add incremental formulation for the moment rotation (bending only) to HM law (it is very simila...

You are computing bending, while Plassiard tried to isolate rolling. The difference is not big, and I can't personaly states that one or the other is better. They should give very similar results, I'm sure. I am not familiar with this terminology, what should be the difference between rolling

Re: [Yade-dev] [Yade-users] problem with PyRunner

Are you saying that running yade-* scripts/test/law-test.py fails? Or do you run it in a different way (such as loading a simulation from the disk -- if you save simulation and load it, only the c++ data are preserved, but no python variables (including defined functions))? Cheers, v. It

Re: [Yade-dev] TriaxialTest regression tests.

Ok, sorry, I checked now that it is not part of regression tests; how you show the result is up to you then. For the files, you can require to run the script from the same directory as where it is, like we do for examples which have their data files in their own directory. v. Thanks for

Re: [Yade-dev] TriaxialTest regression tests.

A path problem... It works here only because I have the files in data duplicated in yade/bin. I'll try and fix that. In the meantime you could move data files in bin too. Thanks for reporting! Bruno, you have to include all data in the script itself (including initial sphere positions, and

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] 1 revision removed

I am not bzr expert, but I am sure you are doing something wrong. Are you merging the right way? trunk to your local copy or vice versa? I would recommend that you use rebase instead of merging, that tries to make your changes originating from the current upstream version (see docs, I think it

Re: [Yade-dev] Trunk is not compiling

I just made a checkout of the actual trunk (bzr2640) and tried to compile it. It doesn't work. To me it looks like there are some problems in respect to some includ files. I tried to fix it but once one error is solved I get a new one just related to another include. Hi Klaus, sorry for

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2621: 1. Make middle-click paste path to variable in the ui

(TSController is changing position AND velocity in trunk, not good with new Newton, even though it is still working correctly for spheres). Hi Bruno, would it be OK for you if we remove the line that is changing wall position directly (TriaxialStressController.cpp:64)? Some people are using

Re: [Yade-dev] [Yade-pkg] Yade versioning for packages.

But one of mentors wishes to see the package called just yade version 0.60. http://lists.debian.org/debian-mentors/2010/12/msg00472.html What do you think if we do it? We already have the yade package: yade-0.60 (version 0.60-*) package provides (or should provide) virtual package yade in

Re: [Yade-dev] energy conservation? - more results

Going back to the dynamic case (energy.py) : It can be verified that there is an unbalanced energy evolution independant of friction and damping. To make plastic dissipation 0, it is enough to set frictionAngle close to pi/2. For damping, make it 0. The results are attached. ScGeom and L3 give

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2626: 1. Clean up NewtonIntegrator following the discussion on the list; adds 1st order damping for asp...

It seems to me that this kind of loop parallelizes badly since one thread will access a member directly while another one will be updating it from inside the clump inner loop. So 1 loop would be better than 2. I think that is not relevant, since 1. yade parallellizes very badly in general

Re: [Yade-dev] builbot vs. cgal

I asked Remi to add cgal in buildbot features so that TesselationWrapper and others get documentation. I remember Vaclav speaking of packaging/licensing issue though. Should we have different builds for packaging and documentation then (we proly don't need to build doc after each commit)? Ok for

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2626: 1. Clean up NewtonIntegrator following the discussion on the list; adds 1st order damping for asp...

FOREACH(Clump::MemberMap::value_type mm..) { moveMember(mm); saveMaximaVelocity(mm);} Heh... You will not remove one loop in Newton actually, you will remove it in moveMember! I think I will keep the loop in Clump. It is functionality that could be also used elsewhere

[Yade-dev] [Bug 692759] Re: report yade errors during sphinx generation

Oh, that is a very nice idea. There are a few things that you might run into: 1. There are many spurious warnings by sphinx, which cannot be avoided unless upstream integrates patches (https://bitbucket.org/birkenfeld/sphinx/issue/407/patch-do-not-inspect- boost-python) 2. The :yref: directive

Re: [Yade-dev] r2610 and later

I would prefer to test such changes in separate branch. If we continue to commit in trunk without testing, it will be very difficult to find the root of the problem. The last working revision for me is about r2608. Hi Anton, I think everything should work as it used to now. Do you still see

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2621: 1. Make middle-click paste path to variable in the ui

I see. It fixed clumps and DOF-wise blocking, right? Nice! I merged with r2615 so that non-dynamic bodies are skipped again. ? Non-dynamic (which means (see Body::isDynamic) blockedDOFs==DOF_ALL) bodies should not be skipped in the integrator, they will just not have acceleration applied on

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2621: 1. Make middle-click paste path to variable in the ui

Non-dynamic (which means (see Body::isDynamic) blockedDOFs==DOF_ALL) bodies should not be skipped in the integrator, they will just not have acceleration applied on the per-DoF basis. Too many people are modifying the same code without being coordinated. Sorry for not writing clearly. This

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2621: 1. Make middle-click paste path to variable in the ui

The velocity integration (i.e. v+=accel*dt) should be skipped, same for individual DOF blocking and damping. It was the case before 2621 and in 2622. If !dynamic, Newton is only doing two operations : pos+=vel, ori*=rotation. Gee, we are having serious communication issues. Why did you

Re: [Yade-dev] [Branch ~yade-dev/yade/trunk] Rev 2621: 1. Make middle-click paste path to variable in the ui

I modified NewtonIntegrator again in 2625. Please give me a christmas present and don't change NewtonIntegrator before discussing on the list what does not work and how you are going to fix that. There is apparently interplay of many things that must be kept in mind when changing something.

Re: [Yade-dev] added`legendPosition` and `legendPositionSecondary`

plot.py: added `legendPosition` and `legendPositionSecondary` so we can decide where the legend should be placed, by default it is 'upper left' and 'upper right' Do you know that the legends are draggable with mouse (depending on matplotlib version, to be more exact)? Perhaps that is

[Yade-dev] [Bug 690091] Re: r2613 packs.py is not working

Hi Anton, sorry for late reply. The Translation engine (and everything similar) should be modified just like Sega did for RotationEngine in last commits: remove (not just comment, they are really useless now) all blocks that are for non-dynamic bodies (i.e. setting new positions), that will be

Re: [Yade-dev] r2610 and later

Sorry to be annoying, but I think we do not test last changes very well. It is not good. I feel some instability in results, strange behavior (see packs.py and flying away particles). I would prefer to test such changes in separate branch. If we continue to commit in trunk without testing,

  1   2   3   4   5   6   7   8   9   10   >