Sept 25, 2013

[jhinebau]

James' Questions:

1) Who's dgupta? He/she seems legit. 2) Will IO functions ever need a specific home? If so keep it. Kill it if we can't think of a good example. 3) What classes currently have near-finished documentation that we can use for templates? 4) What .rst files should we all work on before 1.0 release? 5) Can minutes be recorded in "Issue" format for James' benefit? (6-month internship starting next week)

[putza]

dgupta: Jeffs undergraduate student He is going to work on the network stitching project. (Separate features in git for development release)

[jhinebau]

Minutes taken by James: a) dgupta is an undergrad working on network stitching with Jeff b) Do everything in SI, floats should be able to handle arbitrarily large or small numbers. When writing to text, use something like %e instead of %f (scipy lectures at http://scipy-lectures.github.io/ should have info on how to format strings) c) Maybe Base folder should be Utilities folder, and hold things like ImportMat and other utilities that you may or may not need. Jeff will rename. James will move ImportMat into it. d) Using numpy style documenation, Jeff's machine doesn't generate this properly. Andreas thinks that numpylib is a problem with Windows. e) Decided that current release (main branch) automatically loads on website. Andreas thinks that we should be able to fix Jeff's doc building problem. Will list findings on the "Documentation" issue that is currenlty open. f) James want's to know what everyone is doing before the release: i) Harold only GUI due to course work ii) Mahmoud doing permeability (also an example/demo of his code). iii) Andreas will do a few examples and a bit more documentation. iv) James is going to now work on documentation of his functions, and needs Jeff to tell him which parts of the primary rst files he should write. g) Andreas will work on the IOpickle/unpickle in the base class h) James wants preference file to be read in, that contains, among other things, the location of the Local Files folder. For now, he will put a global variable in OpenPNM.init that figures out the full path to the LocalFiles folder i) Andreas says there's a test that you can try different paths for files existing. j) Harold will make a new class of an "elegant dictionary" in the Network? folder that will have the ability to set data type and check data type, and convert if it is ever wrong (with a debug message). So, some list of keys and data types that can be used to check. k) James will wholesale replace all instances of np with sp, and we'll only use sp from now on. l) [combined with j] m) Andreas says to never specify float64, just float, and the system will use the largest it can handle. n) .gitignore should never be pushed. Also, never push anything that you are not trying to. It really messes up the repository. o) For the GUI, everyone needs to create blackbox functions for all algorithms they are using. Report to harold if you need a new widget, other than path or integer. p) This is the best way to generate a list of pore numbers from a face: mask = network.pore_poroperties['type']==face inlets = network.pore_properties['numbering'][mask] q) Harold needs to merge develop into gui, so we all can work on getting our functions into it. r) Harold is adding "Add property" inbetween "Add network" and "Add algorithm" s) loglevel should be added to all GUI black boxes, so that we can debug on the fly t) Generators should have the option to specify seed, so that exact network can be recreated. u) OpenPNM.IO.NetTVtp(net=pn) is being moved to OpenPNM.VIS.NetToVtp(net=pn) v) all things in the gui, the name you make, should be usable, and explorable with non-gui commands. Harold will create a way to save for now. But James thinks things should be able to be interacted with while the GUI is running

[jhinebau]

Feel free to create individual issues based on lines of these minutes.

[jgostick]

b) the correct string is indeed %e f-iv) @jhinebau should add his docuementation to /docs/userguide/algorithms. The file is just a skeleton now, James is welcome to flesh it out. q) done.