Create your own GitHub profile
Sign up for your own profile on GitHub, the best place to host code, manage projects, and build software alongside 40 million developers.
Sign up-
University of Cambridge
- Cambridge UK
- http://ericwieser.me
Pinned
2,194 contributions in the last year
Activity overview
Contribution activity
January 2020
Created a pull request in numpy/numpy that received 9 comments
MAINT: Refactor dtype conversion functions to be more similar
Before there were three signatures:
PyArray_Descr*(...)
PyArray_Descr*(..., int* err_flag)
int(..., PyArray_Descr** ret)
This unifies them all into
+76
−67
•
9
comments
- BUG: Add some missing C error handling
- MAINT: Avoid escaping unicode in error messages
- MAINT: Ensure `_convert_from_*` functions set errors
- BUG: Add missing error handling to _convert_from_list
- MAINT: Push down declarations in _convert_from_*
- MAINT: Express PyArray_DescrAlignConverter in terms of _convert_from_any
- MAINT: Use a simpler return convention for internal functions
- BUG: Fix PyArray_DescrAlignConverter2 on tuples
- MAINT: remove duplicated if statements between DescrConverters
- MAINT: Tidy PyArray_DescrConverter
- BUG: Use PyDict_GetItemWithError() instead of PyDict_GetItem()
- MAINT: Work with unicode strings in `dtype('i8,i8')`
- MAINT: Remove FIXME comments introduced in the previous commit
- MAINT: Improve const-correctness
- MAINT: Implement keyword-only arguments as syntax
- MAINT: Delete numpy.distutils.compat
- MAINT: Improve const-correctness of string arguments
- MAINT: Improve const-correctness of shapes and strides
- MAINT: Eliminate some calls to `eval`
- MAINT: cleanup use of sys.exc_info
- DEP: records: Deprecate treating shape=0 as shape=None
- Refactor differential operators to avoid duplication
- Use NotImplemented to defer MV operations to Dop
- Remove trailing empty return statements
- Cleanup ReciprocalFrame
- Use apply_function_list in Simp.apply
- Remove the now-unused `test_init_slots`
- Add documentation for `metric.symbol_list`
- Drop python 2.7
- ENH: Add cyclic garbage collection to NumPy
- BUG: `np.resize` negative shape and subclasses edge case fixes
- MAINT: Remove sys.version checks
- MAINT: Remove sys.version checks in tests
- MAINT: cleanup sys.version dependant code
- MAINT: fix typos
- NEP, ENH: implement NEP34 (deprecate ragged object array creation)
- MAINT: Ragged cleanup
- MAINT: Eliminate some calls to `eval`
- ENH: Merge empty hist bins
- MAINT: linalg: use symbol suffix in fallback lapack_lite
- MAINT: Ensure `_convert_from_*` functions set errors
- DOC: Updating f2py docs to python 3 and fixing some typos
- BUG: Add missing error handling to _convert_from_list
- MAINT: Remove NPY_PY3K constant
- MAINT: Refactor dtype conversion functions to be more similar
- DOC: Added tutorial about linear algebra on multidimensional arrays
- MAINT: Express PyArray_DescrAlignConverter in terms of _convert_from_any
- MAINT: travis-ci: Update CI scripts.
- MAINT: C code simplifications
- Return data types of np.average and np.ma.average behave the same. Weights for np.ma.average can be masked.
- MAINT: Use a simpler return convention for internal functions
- MAINT: remove duplicated if statements between DescrConverters
- MAINT: Remove Python 2 divide
- MAINT: Remove Python2 ndarray.__unicode__
- Some pull request reviews not shown.
- Install swig on linux ci
- Make Colin maintainer
- Increase version number to 1.4.0 pre-release
- Allow dropping into interactive pdb shell when hitting exception. Ad…
- Allow users to load into any Python code from GPI
- New example and improve BinaryValue performance
- Adjust path for sim makefiles
- Differentiate set_signal_value for logic string from ASCII string
- Use lock on read transaction
- Remove trailing whitespace
- Do not ignore failures during test initialization
- Fix for handle class equality operators.
- Conditional waiting on events.
- Improve documentation; fix typos.
- Fix with_timeout trigger
- Documentation improvements: add markup; tweak wording; fix typos
Created an issue in numpy/numpy that received 6 comments
np.dtype hangs on self-referential dtypes
Malicious, but: >>> l = [] >>> l.append(('a', l)) >>> np.dtype(l) # hangs, probably also stackoverflow Should raise a RecursionError.
6
comments
30
contributions
in private repositories
Jan 7 – Jan 14