As with all buffer overflows, it is hard to make a test that
consistently triggers tests the behavior. This test does not work for me
through the test_slic.py file, but it works in isolation.
Implement Appveyor builds
initial test
Updated appveyor.yml
New nosetest call
With conda update conda
with pillow
without pillow
TST: Change case sensitivive JPG extension
Revert "TST: Change case sensitivive JPG extension"
This reverts commit 2deed7cc63736f7c6f0387bd37df4c6643c32847.
Trying with Python 2.7
Trying with version env
Trying again with variables
Testing on all Python versions
don't allow failures
Allow failure
Do not actually use tests
Ignore failing tests
Removed Python 2.6 test
Testing only Python 2.6
Testing out more sklearn like AppVeyor CI
Added install to pip
Add artifacts
Enabled all permutations
Disable 2.6, add PIL
Python 2.6, 2.7 only with PIL
Testing with variable dependencies
Allow failure with IF ELSE
With Python 3.4
Scikit-learn like appveyor CI
Fixing paths
Undo path fix
path fix single line
path fix single line 2
Using Miniconda
More path fixes
New wheelhouse link
Added pillow to requirements.txt
Added networkx to requirements.txt
Add testing of 32/64-bit Python 2.7 and 3.4 to matrix
Debugging Cython compile
Retry with all 4 builds
Updated install.ps1 file
Updated based on latest python-appveyor-demo
Debugging pip install
Specify numpy 1.8.1 until whl is uploaded to rackspace
Use skimage-wide requirements.txt file
Minor comment change to trigger build
Install wheel and then install from WHEELHOUSE
Install six from pip
Install networkx from pip
Install pyparsing from pip
Install pytz from pip
Try using just find-links
Install the binary dependencies first, then the rest
Add pillow to the install list
Fix appveyor.yml syntax
Fix requirements.txt syntax
Fix requirements.txt syntax again
Fix appveyor call to initial install
Fix appveyor call to initial install again
Fix appveyor call to initial install yet again
Install wheel
Install wheel first
Install wheel and nose in the appveyor requirements.txt
Fix Python3 version to match python ftp site
Only use cleanup decorator if available
Add debug info to multiimage test
More debugging information
Fix handling of path separators on Windows
Add another warning guard
Fix warning handling for non-windows
Do not use TkAgg as it may be causing alloc error
Clean up echo command
Allow for unclosed file warning
Fix spacing in echo command
It still defaults to `True` but only when the last dimension of the
input array could be construed as RGB.
Also, update ValueError description in docstring.
Fix handling of multiple warnings
Update all test __init__ files
Update segmentation pkg
Update the color pkg
Update the exposure pkg
Update the filters pkg
Update the io pkg
Update the measure pkg
Update morphology package
Restructure test setup function
Add expected_warnings to __all__
Update restoration pkg.
Remove explicit filter check since it is done elsewhere
Fix the image test helpers
Update the transform pkg
Fix util pkg
Update viewer pkg
Add a function to set up a skimage test
Switch to new test helper function
Import local packages that raise warnings in test setup function
More fixes to doctests
Fix regionprops doc test
Try and fix the test_rank failure.
Remove no longer needed RectangleSelector shim
Skip more doctests in _regionprops
Try importing another scipy subpackage
Reset plugins prior to running collections test
Handle warnings in morphology pkg
Add __init__ for morpohology tests
Handle warnings for novice pkg
Handle warnings for restoration pkg
Handle warnings for segmentation pkg
Handle warnings for _shared pkg
Handle warnings for transform pkg
Handle warnings for util pkg
Handle warnings in viewer module
Also:
* New tests to cover these new checks
* Improvements to docstrings and user warnings
* Generalize handling of `sampling` in accordance with docstring
* Some extra whitespace to improve readability
Previously, having a different `sigma` for different dimensions
required an array input. This allows the user to use a simple list,
which gets converted to an array internally.
Importantly, it removes a very unhelpful error:
```python
>>> im = np.random.rand(10, 20)
>>> from skimage import segmentation as seg
Exception AttributeError: "'UmfpackContext' object has no attribute '_symbolic'" in <bound method UmfpackContext.__del__ of <scipy.sparse.linalg.dsolve.umfpack.umfpack.UmfpackContext object at 0x1045ff5d0>> ignored
>>> s = seg.slic(im, 2, sigma=[2, 1])
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-3-689b36a2f0ef> in <module>()
----> 1 s = seg.slic(im, 2, sigma=[2, 1])
/Users/nuneziglesiasj/venv/skimdev2/lib/python2.7/site-packages/scikit_image-0.9dev-py2.7-macosx-10.5-x86_64.egg/skimage/segmentation/slic_superpixels.pyc in slic(image, n_segments, compactness, max_iter, sigma, multichannel, convert2lab, ratio)
106 if not isinstance(sigma, coll.Iterable):
107 sigma = np.array([sigma, sigma, sigma])
--> 108 if (sigma > 0).any():
109 sigma = list(sigma) + [0]
110 image = ndimage.gaussian_filter(image, sigma)
AttributeError: 'bool' object has no attribute 'any'
```
`np.atleast_3d` will add a singleton dimension at the end of an array
if needed. This is not the correct thing to do if `multichannel=False`
based on the subsequent lines. If the input image was 2D with shape
`(40, 50)` and `multichannel=False`, then `np.atleast_3d` gives it
shape `(40, 50, 1)`, and then, because `multichannel=False`, the rest
of the code gives it shape `(40, 50, 1, 1)`. This results in the final
returned array having shape `(40, 50, 1)` instead of the desired
`(40, 50)`.
This commit fixes that and updates the test to detect this failure.
Stefan van der Walt