<feed xmlns='http://www.w3.org/2005/Atom'>
<title>delta/python-packages/numpy.git/numpy/testing, branch v1.10.0rc2</title>
<subtitle>github.com: numpy/numpy.git
</subtitle>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/'/>
<entry>
<title>MAINT: fix raise_warnings in np.test in py3</title>
<updated>2015-07-27T22:15:37+00:00</updated>
<author>
<name>Allan Haldane</name>
<email>allan.haldane@gmail.com</email>
</author>
<published>2015-07-18T15:34:11+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=ede15881846af7aa6058c89359b9f9190201bfb5'/>
<id>ede15881846af7aa6058c89359b9f9190201bfb5</id>
<content type='text'>
In python3 np.test would raise an error if raise_warnings is a list
since dict.keys() now returns a dict_keys instance. Eg,

    np.test('full', raise_warnings=[FutureWarning])
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In python3 np.test would raise an error if raise_warnings is a list
since dict.keys() now returns a dict_keys instance. Eg,

    np.test('full', raise_warnings=[FutureWarning])
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge pull request #6055 from seberg/test-warn-always</title>
<updated>2015-07-26T14:26:51+00:00</updated>
<author>
<name>Ralf Gommers</name>
<email>ralf.gommers@googlemail.com</email>
</author>
<published>2015-07-26T14:26:51+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=31f4f7903f9dea792dfe6ba213e00ad556a0285e'/>
<id>31f4f7903f9dea792dfe6ba213e00ad556a0285e</id>
<content type='text'>
TST: Make default for all warnings "always"</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
TST: Make default for all warnings "always"</pre>
</div>
</content>
</entry>
<entry>
<title>BUG: Fixed import error on Windows from not correctly defining a</title>
<updated>2015-07-13T18:54:50+00:00</updated>
<author>
<name>Ian Henriksen</name>
<email>insertinterestingnamehere@gmail.com</email>
</author>
<published>2015-07-13T18:54:02+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=cd9a6a2ddfb0653601865794f3539210b94955ce'/>
<id>cd9a6a2ddfb0653601865794f3539210b94955ce</id>
<content type='text'>
function listed in __all__ for numpy/testing/utils.py.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
function listed in __all__ for numpy/testing/utils.py.
</pre>
</div>
</content>
</entry>
<entry>
<title>TST: Make default for all warnings "always"</title>
<updated>2015-07-08T21:31:35+00:00</updated>
<author>
<name>Sebastian Berg</name>
<email>sebastian@sipsolutions.net</email>
</author>
<published>2015-07-08T21:31:35+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=64430fa20a7c6c97f906efded23c8c6e057c98e7'/>
<id>64430fa20a7c6c97f906efded23c8c6e057c98e7</id>
<content type='text'>
This should make it easier in some cases, since a warning
caught in some test (but not important/not causing failure),
will not shadow later tests.
In principle the best thing is probably to always check the number
of warnings raised, so that you notice when a new warning shows
up in a test, though there may be other corner cases here.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This should make it easier in some cases, since a warning
caught in some test (but not important/not causing failure),
will not shadow later tests.
In principle the best thing is probably to always check the number
of warnings raised, so that you notice when a new warning shows
up in a test, though there may be other corner cases here.
</pre>
</div>
</content>
</entry>
<entry>
<title>MAINT: pyflakes for numpy/testing, numpy/testing/tests.</title>
<updated>2015-07-05T16:35:01+00:00</updated>
<author>
<name>Charles Harris</name>
<email>charlesr.harris@gmail.com</email>
</author>
<published>2015-07-05T16:18:57+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=ad40c230c4da2ca336bed6b093e8efaba590eec3'/>
<id>ad40c230c4da2ca336bed6b093e8efaba590eec3</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>STY: PEP8 fixes for numpy/testing numpy/testing/tests.</title>
<updated>2015-07-05T15:59:07+00:00</updated>
<author>
<name>Charles Harris</name>
<email>charlesr.harris@gmail.com</email>
</author>
<published>2015-07-05T15:38:47+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=c80c609950fe5aeecf3082d397c7b5149a834fca'/>
<id>c80c609950fe5aeecf3082d397c7b5149a834fca</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>MAINT: filter out some warnings triggered by nose 1.3.7 + python 3.5b2</title>
<updated>2015-06-25T06:52:07+00:00</updated>
<author>
<name>Nathaniel J. Smith</name>
<email>njs@pobox.com</email>
</author>
<published>2015-06-25T06:52:07+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=e8c412453fa6c95a488e0ddfdb298be520b782d2'/>
<id>e8c412453fa6c95a488e0ddfdb298be520b782d2</id>
<content type='text'>
Filed upstream as:
    https://github.com/nose-devs/nose/issues/929
Not our problem.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Filed upstream as:
    https://github.com/nose-devs/nose/issues/929
Not our problem.
</pre>
</div>
</content>
</entry>
<entry>
<title>MAINT: move the special case for void comparison before the regular case</title>
<updated>2015-06-13T18:32:54+00:00</updated>
<author>
<name>Nathaniel J. Smith</name>
<email>njs@pobox.com</email>
</author>
<published>2015-05-08T03:24:06+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=1e3ab40493fadb5daa67c8a55c5360fd934cca7b'/>
<id>1e3ab40493fadb5daa67c8a55c5360fd934cca7b</id>
<content type='text'>
The ndarray richcompare function has special case code for handling
void dtypes (esp. structured dtypes), since there are no ufuncs for
this. Previously, we would attempt to call the relevant
ufunc (e.g. np.equal), and then when this failed (as signaled by the
ufunc returning NotImplemented), we would fall back on the special
case code. This commit moves the special case code to before the
regular code, so that it no longer requires ufuncs to return
NotImplemented.

Technically, it is possible to define ufunc loops for void dtypes
using PyUFunc_RegisterLoopForDescr, so technically I think this commit
changes behaviour: if someone had registered a ufunc loop for one of
these operations, then previously it might have been found and
pre-empted the special case fallback code; now, we use the
special-case code without even checking for any ufunc. But the only
possible use of this functionality would have been if someone wanted
to redefine what == or != meant for a particular structured dtype --
like, they decided that equality for 2-tuples of float32's should be
different from the obvious thing. This does not seem like an important
capability to preserve.

There were also several cases here where on error, an array comparison
would return a scalar instead of raising. This is supposedly
deprecated, but there were call paths that did this that had no
deprecation warning. I added those warnings.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The ndarray richcompare function has special case code for handling
void dtypes (esp. structured dtypes), since there are no ufuncs for
this. Previously, we would attempt to call the relevant
ufunc (e.g. np.equal), and then when this failed (as signaled by the
ufunc returning NotImplemented), we would fall back on the special
case code. This commit moves the special case code to before the
regular code, so that it no longer requires ufuncs to return
NotImplemented.

Technically, it is possible to define ufunc loops for void dtypes
using PyUFunc_RegisterLoopForDescr, so technically I think this commit
changes behaviour: if someone had registered a ufunc loop for one of
these operations, then previously it might have been found and
pre-empted the special case fallback code; now, we use the
special-case code without even checking for any ufunc. But the only
possible use of this functionality would have been if someone wanted
to redefine what == or != meant for a particular structured dtype --
like, they decided that equality for 2-tuples of float32's should be
different from the obvious thing. This does not seem like an important
capability to preserve.

There were also several cases here where on error, an array comparison
would return a scalar instead of raising. This is supposedly
deprecated, but there were call paths that did this that had no
deprecation warning. I added those warnings.
</pre>
</div>
</content>
</entry>
<entry>
<title>ENH: add clear_and_catch_warnings context manager</title>
<updated>2015-03-15T01:47:45+00:00</updated>
<author>
<name>Matthew Brett</name>
<email>matthew.brett@gmail.com</email>
</author>
<published>2015-03-15T01:32:01+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=28971cda67d22a107ceb50514eb9d53765c0258b'/>
<id>28971cda67d22a107ceb50514eb9d53765c0258b</id>
<content type='text'>
Add context manager for tests that, on entry, clears record of any
previous warnings from given modules, so that any warning raised in the
context block does get raised, rather than suppressed by a previous
warning causing an entry in the `__warningregistry_`` of the module.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add context manager for tests that, on entry, clears record of any
previous warnings from given modules, so that any warning raised in the
context block does get raised, rather than suppressed by a previous
warning causing an entry in the `__warningregistry_`` of the module.
</pre>
</div>
</content>
</entry>
<entry>
<title>DOC: Reconcile docstrings and function signatures where they disagree</title>
<updated>2015-03-02T20:00:18+00:00</updated>
<author>
<name>Robert McGibbon</name>
<email>rmcgibbo@gmail.com</email>
</author>
<published>2015-03-02T06:53:19+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/python-packages/numpy.git/commit/?id=6ca778d275ade854cfc81a6a975230c4ccd9402c'/>
<id>6ca778d275ade854cfc81a6a975230c4ccd9402c</id>
<content type='text'>
Fixed typos in docstrings were updated for functions where the parameter
names in the docstring didn't match the function signature.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Fixed typos in docstrings were updated for functions where the parameter
names in the docstring didn't match the function signature.
</pre>
</div>
</content>
</entry>
</feed>
