RE: [PATCH] git-p4: close temporary file before removing

>31.07.2019, 09:53, "Philip McGraw" <philip.mcgraw@bentley.com>:
>>>  30.07.2019, 13:37, "Philip McGraw" <philip.mcgraw@bentley.com>:
>>>  > python os.remove() throws exceptions on Windows platform when attempting
>>>  > to remove file while it is still open. Need to grab filename while file open,
>>>  > close file handle, then remove by name. Apparently other platforms are more
>>>  > permissive of removing files while busy.
>>>  > reference: 
>>>  > ---
>>>  >  git-p4.py | 4 +++-
>>>  >  1 file changed, 3 insertions(+), 1 deletion(-)
>>>  >
>>>  > diff --git a/git-p4.py b/git-p4.py
>>>  > index c71a6832e2..6b9d2a8317 100755
>>>  > --- a/git-p4.py
>>>  > +++ b/git-p4.py
>>>  > @@ -1161,12 +1161,14 @@ def exceedsLargeFileThreshold(self, relPath, contents):
>>>  >                  return False
>>>  >              contentTempFile = self.generateTempFile(contents)
>>>  >              compressedContentFile = tempfile.NamedTemporaryFile(prefix='git-p4-large-file', delete=False)
>>>  > + compressedContentFileName = compressedContentFile.name
>>>  >              zf = zipfile.ZipFile(compressedContentFile.name, mode='w')
>>>  >              zf.write(contentTempFile, compress_type=zipfile.ZIP_DEFLATED)
>>>  >              zf.close()
>>>  >              compressedContentsSize = zf.infolist()[0].compress_size
>>>  >              os.remove(contentTempFile)
>>>  > - os.remove(compressedContentFile.name)
>>>  > + compressedContentFile.close()
>>>  > + os.remove(compressedContentFileName)
>>>
>>>  I'm not sure why NamedTemporaryFile() is called with delete=False above,
>>>  but it appears to me that it can have delete=True instead,
>>>  so that there is no need to call os.remove() explicitly
>>>  and thus worry about remove vs close ordering at all.
>>>
>>>  >              if compressedContentsSize > gitConfigInt('git-p4.largeFileCompressedThreshold'):
>>>  >                  return True
>>>  >          return False
>>>  > --
>>>  > 2.21.0.windows.1
>>>
>>>  Thank you,
>>>  Andrey.
>>
>> Thanks Andrey; simpler is certainly better! I will test and re-submit v2 of patch with that approach.
>
>Thank you, that would be great!
>
>-- 
>Andrey.