af.features.Features crashes when its __del__ method is called

Open
#268 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
38/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Stale
Tech stack
python
Domain
api

Research direction

The reported entry points are af.orb, Features.get_xpos(), Features.get_ypos(), and Features.del; first run the provided minimum working example on Windows with ArrayFire 3.8.0. Trace the object lifetime when the position arrays fall out of scope, then verify that deleting features no longer crashes silently and that the returned-position case still works.

Written by the indexing model from the issue text.

Description

When using a Features() object, arrayfire crashes silently when calling the object's __del__() method if the object's get_xpos() method has already been deleted (or has fallen out of scope).

Potentially due to a similar root case as mentioned in #169 (or due to the changes made as a result of this in commit a9031fb5e0db371bc88c958f0b8d43efd7fe8f45)

Running on Windows; using arrayfire 3.8.0.

Minimum working example:

import arrayfire as af

def get_feature_locations(features):
    # Get x and y positions of features
    x_pos = features.get_xpos()
    y_pos = features.get_ypos()
    
    # If this is uncommented, the code works as intended
    # return x_pos, y_pos

img = af.randn(d0=500, d1=500)
features, descriptors = af.orb(img, threshold=1, max_features=10, num_levels=1)
feature_locations = get_feature_locations(features)

print("Before delete")
del features
print("After delete")

Which outputs:

Before delete

Uncommenting line 10 (return x_pos, y_pos) outputs:

Before delete
After delete

Thanks!

Dominant language
Python
Stars
422
Forks
63
PR merge metrics
No merged PRs in 30d

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from arrayfire/arrayfire-python

All issues in arrayfire/arrayfire-python

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.