Activity for Ankur Sinha

  • Ankur Sinha Ankur Sinha created ticket #36

    Citing generateDS

  • Ankur Sinha Ankur Sinha posted a comment on ticket #1

    +1 for this feature (although I don't know how much work this is to implement). At the moment, our classes all look like this: Izhikevich2007Cell(neuro_lex_id=None, id=None, metaid=None, notes=None, properties=None, annotation=None, C=None, v0=None, k=None, vr=None, vt=None, vpeak=None, a=None, b=None, c=None, d=None, gds_collector_=None, **kwargs_) which means users don't really get any information about what each argument to the constructor should be---they need to go look at our schema documentation...

  • Ankur Sinha Ankur Sinha posted a comment on ticket #28

    Hi Dave, I think I figured out what was going on. In the generated nml.py file, in the __str__ method of GeneratedsSuper, we're initialising a new output variable as an object of the StringIO class. At the moment, we simply from io import StringIO, which while works for Python 3, isn't the right StringIO to use for Python 2. In Python 2, for strings, we must use StringIO.StringIO because there, io.StringIO is for unicode strings. So, I think the generateds.py code should be tweaked as per this diff:...

  • Ankur Sinha Ankur Sinha posted a comment on ticket #28

    Hi Dave, I haven't been able to pin point the cause of the error yet but I did find a workaround that fixes the issue for us: https://github.com/NeuralEnsemble/libNeuroML/pull/110/commits/aa025c608a678697d14acaca1cf8d7ed0a0a7fd6 The error is caused when we use nml.py with other bits, like reading from HDF5 files, so I'm thinking it's caused by some code in our HDF5 related methods where we haven't been careful enough to take both Python 2 and 3 into account. So we'll just carry this tweak for the...

  • Ankur Sinha Ankur Sinha posted a comment on ticket #28

    Hi Dave, Thanks for that. I'll go through the failing test and see if I can narrow it down. If you're not seeing the error, it could be something else in the library that's using nml.py perhaps. I'll report back once I have something. Thanks again, Ankur

  • Ankur Sinha Ankur Sinha created ticket #28

    Python2.7: TypeError: unicode argument expected, got 'str'

  • Ankur Sinha Ankur Sinha posted a comment on ticket #20

    Thanks very much, Dave. I can confirm that this is now fixed. Please close this ticket as "fixed" :)

  • Ankur Sinha Ankur Sinha posted a comment on ticket #20

    Thanks very much, and sorry for the late reply. I've attached the XSD and other files we use for generation to the issue. You can also see them in our repository here: https://github.com/NeuralEnsemble/libNeuroML/tree/development/neuroml/nml The schema is currently at version 2.2

  • Ankur Sinha Ankur Sinha posted a comment on ticket #20

    The attached tweaks in the generated nml.py file "fixes" this error.

  • Ankur Sinha Ankur Sinha posted a comment on ticket #20

    Here's an example error that we get while running our unit tests: 'Annotation' object has no attribute 'add_anytypeobjs_'", AttributeError("'Annotation' object has no attribute 'add_anytypeobjs_'"))

  • Ankur Sinha Ankur Sinha created ticket #20

    Using `--use-getter-setter=none` does not add `add_anytypeobjs_`, but this method is still used in places in the generated python

  • Ankur Sinha Ankur Sinha posted a comment on ticket #13

    We're still seeing this on generateDS.py version 2.39.9 with Python 3.9. Modifying the generated file to include the tweak mentioned above still fixes it. Could this please be looked at (and if the tweak is correct, included in the next release)?

  • Ankur Sinha Ankur Sinha created ticket #13

    "TypeError: %d format: a number is required, not str"

  • Ankur Sinha Ankur Sinha created ticket #498

    Build failure on s390x

  • Ankur Sinha Ankur Sinha posted a comment on ticket #487

    We're carrying a patch that fixes this issue also for 3.0.1. Attached.

  • Ankur Sinha Ankur Sinha created ticket #487

    Build failures on all architectures other than x86_64 and i686

  • Ankur Sinha Ankur Sinha posted a comment on ticket #483

    Hello! Any updates on this please? If there's anything I can do help here, please do let me know. Cheers! Ankur

  • Ankur Sinha Ankur Sinha created ticket #483

    GDCM install does not export correct CMake variables

  • Ankur Sinha Ankur Sinha posted a comment on ticket #22

    Hi Eric, Sure, a soname bump would be OK. I'll rebuild the packages that use xmedcon. Well, the xmedcon-config seems to be doing what standard tools like pkgconf already do - provide the required flags when needed. So, for example, pkgconfig puts arch specific files "pc" files in /usr/lib/pkgconfig and /usr/lib64/pkgconfig to separate the arches, and so there isn't a conflict. Maybe that'd worth looking into? I'll see if I can find some time to write one up too. It doesn't look too hard :D https://people.freedesktop.org/~dbn/pkg-config-guide.html...

  • Ankur Sinha Ankur Sinha posted a comment on ticket #22

    Hi Erik, Yes, that looks good! I've built a scratch build here to test it out. The builds do not have conflicting headers anymore. They still produce two different xmedcon-config scripts, but I can rename them based on the arch which is probably the correct way to go about it anyway. I see there's no soname bump/API change here, so do you think it's OK to also update the Fedora stable releases from 0.14.x to 0.15.0? Thanks much! Ankur

  • Ankur Sinha Ankur Sinha created ticket #22

    Compilation produces arch specific header files

  • Ankur Sinha Ankur Sinha created ticket #403

    Some tests failing

  • Ankur Sinha Ankur Sinha created ticket #402

    Incorrect cmake command in Utilities/doxygen/CMakeLists.txt

  • Ankur Sinha Ankur Sinha created ticket #18

    Appdata file for arcem

1