read_chunk() breaks in numpy 1.12.0 #383
Labels
bug
Issues that report (apparent) bugs.
Comments
|
This solved it for me. |
mbeacom
pushed a commit
that referenced
this issue
Feb 15, 2017
ensures int arguments to np.reshape; closes #383
tburrows13
pushed a commit
to tburrows13/moviepy
that referenced
this issue
Feb 27, 2017
* Fixed copy-paste typo Changed documentation for the color parameter to distinguish from bg_color * Fixed missing list (using python 3) * fixed module hierarchy for Trajectory * fixed addy * small recipe * on_color function docstring has wrong parameter The docstring for the on_color method has a parameter of bg_color but the function uses color as the parameter. * Update README.rst Changed Code Block to use Python Syntax Highlighting * fix deprecation message currently, the docstring is nonsense. ``` The function ``concatenate_videoclips`` is deprecated and is kept temporarily for backwards compatibility. Please use the new name, ``concatenate_videoclips``, instead. ``` * ImageSequenceClip: Check for fps and durations rather than fps and duration * Add a test case. * Add another test * ensures int arguments to np.reshape; closes Zulko#383 * fix issue Zulko#401 * fix issue Zulko#335 * Update maintainer section in README * make concatenate_videoclips Python 3 compatible.. fix issue Zulko#313 * Increment release version * CompositeVideoClip doesn't accept an argument of transparent * move PY3 variable to compat.py * fix movie => moviepy typo * fix issue Zulko#341 * Fixed typo Zulko#375 Zulko#375 fixed * fix issue Zulko#357, which makes real problem more obvious (media file does not exist * Revert "small recipe (mirroring a video)" * Fixed indentation * .gitignore ignore Mac-specific files, Jetbrains settings dir * README.rst make mention of Gitter, add PyPI and Gitter badge * README.rst wording, formatting * README.rst structure link targets, add co-maintainers (with @username) * README.rst move maintainers, contributing sections; change docs x-reference name * README.rst fix grammar * fix issue 145. raise Exception when concatenate method != chain or compose * make PEP8 compatible * fix PR Zulko#413 . (issue Zulko#357) * create test for issue Zulko#145 * add tests/media to .gitignore * fix Issue Zulko#385 , no DirectoryClip class (Zulko#434) * fix issue Zulko#385 , no DirectoryClip class * replace DirectoryClip with ImageSequenceClip * fix issue 417, unicode has no attribute 'shape' error. * add test for issue 417 * add test for issue 417 * Fixed resize documentation issue Zulko#319 (Zulko#346) * Handle bytes when listing fonts in VideoClip.py (Zulko#306) Handle bytes when listing fonts in VideoClip.py * add test for PR306 * add test for PR306 (Zulko#440) * create test file for pull requests (Zulko#433) * Test issue 407 (video has a valid fps after concatenate function) (Zulko#443) * Move PR test to test_PR.py file (Zulko#444) * move PR test from test_issues.py to test_PR.py * add code to download python_logo.png * remove duplicate test_issue_417 function * add testing with travis-ci (Zulko#447) added testing via travis-ci * put DEVNULL into compat.py (Zulko#432) * add travis-ci badge to readme file * pick highest fps when concatenating (Zulko#416) * readers.py cast chunksize from float to int * choose highest fps of clips when concatenating * pick highest fps when concatenating * pick highest fps when concatenating * fps either max or none * remove resolve markers removed resolve markers such as HEAD, etc so that the file will compile correctly. Remove some double blank lines, etc * update concatenate.py; add c.fps is not None * add test for issue 416 * fix test_issue_416
tburrows13
pushed a commit
to tburrows13/moviepy
that referenced
this issue
Feb 27, 2017
* Fixed copy-paste typo Changed documentation for the color parameter to distinguish from bg_color * Fixed missing list (using python 3) * fixed module hierarchy for Trajectory * fixed addy * small recipe * on_color function docstring has wrong parameter The docstring for the on_color method has a parameter of bg_color but the function uses color as the parameter. * Update README.rst Changed Code Block to use Python Syntax Highlighting * fix deprecation message currently, the docstring is nonsense. ``` The function ``concatenate_videoclips`` is deprecated and is kept temporarily for backwards compatibility. Please use the new name, ``concatenate_videoclips``, instead. ``` * ImageSequenceClip: Check for fps and durations rather than fps and duration * Add a test case. * Add another test * ensures int arguments to np.reshape; closes Zulko#383 * fix issue Zulko#401 * fix issue Zulko#335 * Update maintainer section in README * make concatenate_videoclips Python 3 compatible.. fix issue Zulko#313 * Increment release version * CompositeVideoClip doesn't accept an argument of transparent * move PY3 variable to compat.py * fix movie => moviepy typo * fix issue Zulko#341 * Fixed typo Zulko#375 Zulko#375 fixed * fix issue Zulko#357, which makes real problem more obvious (media file does not exist * Revert "small recipe (mirroring a video)" * Fixed indentation * .gitignore ignore Mac-specific files, Jetbrains settings dir * README.rst make mention of Gitter, add PyPI and Gitter badge * README.rst wording, formatting * README.rst structure link targets, add co-maintainers (with @username) * README.rst move maintainers, contributing sections; change docs x-reference name * README.rst fix grammar * fix issue 145. raise Exception when concatenate method != chain or compose * make PEP8 compatible * fix PR Zulko#413 . (issue Zulko#357) * create test for issue Zulko#145 * add tests/media to .gitignore * fix Issue Zulko#385 , no DirectoryClip class (Zulko#434) * fix issue Zulko#385 , no DirectoryClip class * replace DirectoryClip with ImageSequenceClip * fix issue 417, unicode has no attribute 'shape' error. * add test for issue 417 * add test for issue 417 * Fixed resize documentation issue Zulko#319 (Zulko#346) * Handle bytes when listing fonts in VideoClip.py (Zulko#306) Handle bytes when listing fonts in VideoClip.py * add test for PR306 * add test for PR306 (Zulko#440) * create test file for pull requests (Zulko#433) * Test issue 407 (video has a valid fps after concatenate function) (Zulko#443) * Move PR test to test_PR.py file (Zulko#444) * move PR test from test_issues.py to test_PR.py * add code to download python_logo.png * remove duplicate test_issue_417 function * add testing with travis-ci (Zulko#447) added testing via travis-ci * put DEVNULL into compat.py (Zulko#432) * add travis-ci badge to readme file * pick highest fps when concatenating (Zulko#416) * readers.py cast chunksize from float to int * choose highest fps of clips when concatenating * pick highest fps when concatenating * pick highest fps when concatenating * fps either max or none * remove resolve markers removed resolve markers such as HEAD, etc so that the file will compile correctly. Remove some double blank lines, etc * update concatenate.py; add c.fps is not None * add test for issue 416 * fix test_issue_416
tburrows13
pushed a commit
that referenced
this issue
Mar 2, 2017
* Changed the `col` paramater of `ColorClip` to `color` and added a DeprecationWarning if the users tries to use `col` * Update VideoFileClip.py * Added warnings * Update (#3) * Fixed copy-paste typo Changed documentation for the color parameter to distinguish from bg_color * Fixed missing list (using python 3) * fixed module hierarchy for Trajectory * fixed addy * small recipe * on_color function docstring has wrong parameter The docstring for the on_color method has a parameter of bg_color but the function uses color as the parameter. * Update README.rst Changed Code Block to use Python Syntax Highlighting * fix deprecation message currently, the docstring is nonsense. ``` The function ``concatenate_videoclips`` is deprecated and is kept temporarily for backwards compatibility. Please use the new name, ``concatenate_videoclips``, instead. ``` * ImageSequenceClip: Check for fps and durations rather than fps and duration * Add a test case. * Add another test * ensures int arguments to np.reshape; closes #383 * fix issue #401 * fix issue #335 * Update maintainer section in README * make concatenate_videoclips Python 3 compatible.. fix issue #313 * Increment release version * CompositeVideoClip doesn't accept an argument of transparent * move PY3 variable to compat.py * fix movie => moviepy typo * fix issue #341 * Fixed typo #375 #375 fixed * fix issue #357, which makes real problem more obvious (media file does not exist * Revert "small recipe (mirroring a video)" * Fixed indentation * .gitignore ignore Mac-specific files, Jetbrains settings dir * README.rst make mention of Gitter, add PyPI and Gitter badge * README.rst wording, formatting * README.rst structure link targets, add co-maintainers (with @username) * README.rst move maintainers, contributing sections; change docs x-reference name * README.rst fix grammar * fix issue 145. raise Exception when concatenate method != chain or compose * make PEP8 compatible * fix PR #413 . (issue #357) * create test for issue #145 * add tests/media to .gitignore * fix Issue #385 , no DirectoryClip class (#434) * fix issue #385 , no DirectoryClip class * replace DirectoryClip with ImageSequenceClip * fix issue 417, unicode has no attribute 'shape' error. * add test for issue 417 * add test for issue 417 * Fixed resize documentation issue #319 (#346) * Handle bytes when listing fonts in VideoClip.py (#306) Handle bytes when listing fonts in VideoClip.py * add test for PR306 * add test for PR306 (#440) * create test file for pull requests (#433) * Test issue 407 (video has a valid fps after concatenate function) (#443) * Move PR test to test_PR.py file (#444) * move PR test from test_issues.py to test_PR.py * add code to download python_logo.png * remove duplicate test_issue_417 function * add testing with travis-ci (#447) added testing via travis-ci * put DEVNULL into compat.py (#432) * add travis-ci badge to readme file * pick highest fps when concatenating (#416) * readers.py cast chunksize from float to int * choose highest fps of clips when concatenating * pick highest fps when concatenating * pick highest fps when concatenating * fps either max or none * remove resolve markers removed resolve markers such as HEAD, etc so that the file will compile correctly. Remove some double blank lines, etc * update concatenate.py; add c.fps is not None * add test for issue 416 * fix test_issue_416 * Update Gloin (#4) * Fixed copy-paste typo Changed documentation for the color parameter to distinguish from bg_color * Fixed missing list (using python 3) * fixed module hierarchy for Trajectory * fixed addy * small recipe * on_color function docstring has wrong parameter The docstring for the on_color method has a parameter of bg_color but the function uses color as the parameter. * Update README.rst Changed Code Block to use Python Syntax Highlighting * fix deprecation message currently, the docstring is nonsense. ``` The function ``concatenate_videoclips`` is deprecated and is kept temporarily for backwards compatibility. Please use the new name, ``concatenate_videoclips``, instead. ``` * ImageSequenceClip: Check for fps and durations rather than fps and duration * Add a test case. * Add another test * ensures int arguments to np.reshape; closes #383 * fix issue #401 * fix issue #335 * Update maintainer section in README * make concatenate_videoclips Python 3 compatible.. fix issue #313 * Increment release version * CompositeVideoClip doesn't accept an argument of transparent * move PY3 variable to compat.py * fix movie => moviepy typo * fix issue #341 * Fixed typo #375 #375 fixed * fix issue #357, which makes real problem more obvious (media file does not exist * Revert "small recipe (mirroring a video)" * Fixed indentation * .gitignore ignore Mac-specific files, Jetbrains settings dir * README.rst make mention of Gitter, add PyPI and Gitter badge * README.rst wording, formatting * README.rst structure link targets, add co-maintainers (with @username) * README.rst move maintainers, contributing sections; change docs x-reference name * README.rst fix grammar * fix issue 145. raise Exception when concatenate method != chain or compose * make PEP8 compatible * fix PR #413 . (issue #357) * create test for issue #145 * add tests/media to .gitignore * fix Issue #385 , no DirectoryClip class (#434) * fix issue #385 , no DirectoryClip class * replace DirectoryClip with ImageSequenceClip * fix issue 417, unicode has no attribute 'shape' error. * add test for issue 417 * add test for issue 417 * Fixed resize documentation issue #319 (#346) * Handle bytes when listing fonts in VideoClip.py (#306) Handle bytes when listing fonts in VideoClip.py * add test for PR306 * add test for PR306 (#440) * create test file for pull requests (#433) * Test issue 407 (video has a valid fps after concatenate function) (#443) * Move PR test to test_PR.py file (#444) * move PR test from test_issues.py to test_PR.py * add code to download python_logo.png * remove duplicate test_issue_417 function * add testing with travis-ci (#447) added testing via travis-ci * put DEVNULL into compat.py (#432) * add travis-ci badge to readme file * pick highest fps when concatenating (#416) * readers.py cast chunksize from float to int * choose highest fps of clips when concatenating * pick highest fps when concatenating * pick highest fps when concatenating * fps either max or none * remove resolve markers removed resolve markers such as HEAD, etc so that the file will compile correctly. Remove some double blank lines, etc * update concatenate.py; add c.fps is not None * add test for issue 416 * fix test_issue_416 * Update VideoFileClip.py * Added test * Added return for PEP 8 compliance * Change
crhan
added a commit
to crhan/pyannote-video
that referenced
this issue
Aug 5, 2017
This was referenced Aug 5, 2017
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
In
audio.io.readers.FFMPEG_AudioReader, theread_chunk()method breaks as of numpy 1.12.0 (released a few days ago). Numpy no longer accepts floats as index values, so theresult.reshape((len(result)/self.nchannels, self.nchannels))call raises an exception. In practice, this means working with audio files is basically impossible with the latest numpy. Patch is trivial; PR coming in a moment.The text was updated successfully, but these errors were encountered: