---
- branch: MAIN
  date: Sat Feb  1 21:03:58 UTC 2020
  files:
  - new: '1.15'
    old: '1.14'
    path: pkgsrc/math/py-numba/Makefile
    pathrev: pkgsrc/math/py-numba/Makefile@1.15
    type: modified
  - new: '1.11'
    old: '1.10'
    path: pkgsrc/math/py-numba/PLIST
    pathrev: pkgsrc/math/py-numba/PLIST@1.11
    type: modified
  - new: '1.13'
    old: '1.12'
    path: pkgsrc/math/py-numba/distinfo
    pathrev: pkgsrc/math/py-numba/distinfo@1.13
    type: modified
  id: 20200201T210358Z.82a2196da2cfcb53abaad1f89b70310466b6fb96
  log: "py-numba: updated to 0.48.0\n\nVersion 0.48.0\n\nThis release is particularly
    small as it was present to catch anything that missed the 0.47.0 deadline (the
    deadline deliberately coincided with the end of support for Python 2.7). The next
    release will be considerably larger.\n\nThe core changes in this release are dominated
    by the start of the clean up needed for the end of Python 2.7 support, improvements
    to the CUDA target and support for numerous additional unicode string methods.\n\nEnhancements
    from user contributed PRs (with thanks!):\n\nBrian Wignall fixed more spelling
    typos in.\nDenis Smirnov added support for string methods capitalize, casefold,
    swapcase , rsplit , partition  and splitlines .\nElena Totmenina extended support
    for string methods startswith  and added endswith .\nEric Wieser made type_callable
    return the decorated function itself in\nEthan Pronovost added support for np.argwhere
    in\nGraham Markall contributed a large number of CUDA enhancements and fixes,
    namely:\n* Remove Python 3.4 backports from utils\n* Make device_array_like create
    contiguous arrays\n* Don窶å\x86² launch ForAll kernels with 0 elements\n* Fix
    various issues in CUDA library search\n* Enable use of records and bools for shared
    memory, remove ddt, add additional transpose tests\n* Fix: Add more appropriate
    typing for CUDA device arrays\n* test_consuming_strides: Keep dev array alive\n*
    State that CUDA Toolkit 8.0 required in docs\nJames Bourbeau added the Python
    3.8 classifier to setup.py in.\nJohn Kirkham added a clarification to the __cuda_array_interface__
    documentation in.\nLeo Fang Fixed an indexing problem in dummyarray in.\nMarcel
    Bargull fixed a build and test issue for Python 3.8 in.\nMaria Rubtsov added support
    for string methods isdecimal , isdigit , isnumeric  and replace .\n\nGeneral Enhancements:\n*
    Make type_callable return the decorated function\n* merge string prs\n\nThis merge
    PR included the following:\n* Implement str.capitalize() based on CPython\n* Implement
    str.casefold() based on CPython\n* Implement str.swapcase() based on CPython\n*
    Implement str.rsplit() based on CPython\n* Implement str.isdecimal\n* Implement
    str.isdigit\n* Implement str.isnumeric\n* Implement str.partition() based on CPython\n*
    Implement str.splitlines() based on CPython\n* Implement str.replace\n* Functionality
    extension str.startswith() based on CPython\n* Add functionality for str.endswith()\n*
    Disable help messages.\n* Add coverage for np.argwhere\n\nFixes:\n* Only use lives
    (and not aliases) to create post parfor live set.\n* Fix more spelling typos\n*
    Propagate semantic constants ahead of static rewrites.\n* Add Python 3.8 classifier
    to setup.py\n* Update setup.py and buildscripts for dependency requirements\n*
    Convert from arrays to names in define() and don窶å\x86² invalidate for multiple
    consistent defines.\n* Permit mixed int types in wrap_index\n* Catch the use of
    global typed-list in JITed functions\n* Fix, bug in bytecode analysis.\n\nCUDA
    Enhancements/Fixes:\n* Fix: Add more appropriate typing for CUDA device arrays\n*
    Make device_array_like create contiguous arrays\n* State that CUDA Toolkit 8.0
    required in docs\n* test_consuming_strides: Keep dev array alive\n* Fix IndexError
    when accessing the 窶�-1窶� element of dummyarray\n* Enable use of records
    and bools for shared memory, remove ddt, add additional transpose tests\n* Fix
    various issues in CUDA library search\n* Don窶å\x86² launch ForAll kernels with
    0 elements\n* Remove Python 3.4 backports from utils\nDocumentation Updates:\n\n*
    Clarify what dictionary means\n* Update docs for updated version requirements\n*
    Update deprecation notices for 0.48.0\nCI updates:\n\n* Install optional dependencies
    for Python 3.8 tests\n* Drop Py2.7 and Py3.5 from public CI\n* Fix CI py38\n"
  module: pkgsrc
  subject: 'CVS commit: pkgsrc/math/py-numba'
  unixtime: '1580591038'
  user: adam