I/O operation on closed file #1664
Comments
Black fails with "I/O operation on closed file", see the reported issue on black: psf/black#1664
|
As another data point from another code base, I observed this issue when 17 of 188 files would be changed by black. Once those files were changed to satisfy black, the issue went away. When only 5 of 188 files do not satisfy black, the problem occurs sporadically for me. So it seems like the problem manifests when there are a sufficient number of diffs being generated. |
|
The problematic behavior only happens when the optional When wrapping the output stream with colorama, it is assumed that colorama returns a wrapped stream without its own I'll post a PR for this shortly. |
The wrap_stream_for_windows() function calls colorama.initialise.wrap_stream() function to apply colorama's magic to wrapper to the output stream. Except this wrapper is only applied on Windows platforms that need it, otherwise the original stream is returned as-is. The colorama wrapped stream lacks a detach() method, so a no-op lambda was being assigned to the wrapped stream. The problem is that the no-op lambda was being assigned unconditionally whether or not colorama actually returns a wrapped stream, thus replacing the original TextIOWrapper's detach() method. Replacing the detach() method with a no-op lambda is the root cause of the problem observed in psf#1664. The solution is to only assign the no-op detach() method if the stream lacks its own detach() method. Repairs psf#1664
The wrap_stream_for_windows() function calls colorama.initialise.wrap_stream() function to apply colorama's magic to wrapper to the output stream. Except this wrapper is only applied on Windows platforms that need it, otherwise the original stream is returned as-is. The colorama wrapped stream lacks a detach() method, so a no-op lambda was being assigned to the wrapped stream. The problem is that the no-op lambda was being assigned unconditionally whether or not colorama actually returns a wrapped stream, thus replacing the original TextIOWrapper's detach() method. Replacing the detach() method with a no-op lambda is the root cause of the problem observed in psf#1664. The solution is to only assign the no-op detach() method if the stream lacks its own detach() method. Repairs psf#1664
The wrap_stream_for_windows() function calls colorama.initialise.wrap_stream() function to apply colorama's magic to wrapper to the output stream. Except this wrapper is only applied on Windows platforms that need it, otherwise the original stream is returned as-is. The colorama wrapped stream lacks a detach() method, so a no-op lambda was being assigned to the wrapped stream. The problem is that the no-op lambda was being assigned unconditionally whether or not colorama actually returns a wrapped stream, thus replacing the original TextIOWrapper's detach() method. Replacing the detach() method with a no-op lambda is the root cause of the problem observed in psf#1664. The solution is to only assign the no-op detach() method if the stream lacks its own detach() method. Repairs psf#1664
* Repair colorama wrapping on non-Windows platforms The wrap_stream_for_windows() function calls colorama.initialise.wrap_stream() function to apply colorama's magic to wrapper to the output stream. Except this wrapper is only applied on Windows platforms that need it, otherwise the original stream is returned as-is. The colorama wrapped stream lacks a detach() method, so a no-op lambda was being assigned to the wrapped stream. The problem is that the no-op lambda was being assigned unconditionally whether or not colorama actually returns a wrapped stream, thus replacing the original TextIOWrapper's detach() method. Replacing the detach() method with a no-op lambda is the root cause of the problem observed in #1664. The solution is to only assign the no-op detach() method if the stream lacks its own detach() method. Repairs #1664
|
Running Before 4d71d74: 2 files would be reformatted, 56 files would be left unchanged, 86 files would fail to reformat.After 4d71d74: 88 files would be reformatted, 56 files would be left unchanged.My environment
I'll have to check that the patch also works on Windows but that will take a while - hopefully I don't have mess around with Windows Update :-) |
I procrastinated for way too long on getting back if this patch works on Windows or not but the answer is "yes it does" :D |
|
This was fixed quite a while ago, although it was only released in 21.4b0. |
* Repair colorama wrapping on non-Windows platforms The wrap_stream_for_windows() function calls colorama.initialise.wrap_stream() function to apply colorama's magic to wrapper to the output stream. Except this wrapper is only applied on Windows platforms that need it, otherwise the original stream is returned as-is. The colorama wrapped stream lacks a detach() method, so a no-op lambda was being assigned to the wrapped stream. The problem is that the no-op lambda was being assigned unconditionally whether or not colorama actually returns a wrapped stream, thus replacing the original TextIOWrapper's detach() method. Replacing the detach() method with a no-op lambda is the root cause of the problem observed in psf#1664. The solution is to only assign the no-op detach() method if the stream lacks its own detach() method. Repairs psf#1664
Running Black 20.8b1 with the
--diffoption on a tree (> 100 files) produces errors likeRunning
black --diffwith a single file does not produce this error.Linux/Python 3.7.8+
The text was updated successfully, but these errors were encountered: