I previously wrote:
> The problem is that s3cmd uploads the files in subdirectories with a \
> instead of a / which means you have to access them like:
>
> http://s3/bucket/subdir\subdir\file.ext
I've solved my problem by slightly modifying S3uri.py around line 58:
self._object = unicodise(groups[1].replace('\\', '/'))
I realize this is not portable but I don't know any Python. However, I
think this is the right place to make the conversion in path separator
from Windows to Unix (S3).
Note while testing with --dry-run, I was still getting the wrong
behavior because it doesn't use the s3uri class but uses a shortcut. I
modified s3cmd.py around line 958 to use the S3Uri class:
uri = S3Uri(local_list[key]['remote_uri'])
output(u"upload: %s -> %s" % (local_list[key]['full_name_unicode'],
uri.uri()))
Dry-run output looked right and I'm running S3cmd sync now and it's
working great. Hope that helps,
Brian
------------------------------------------------------------------------------
_______________________________________________
S3tools-general mailing list
S3tools-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/s3tools-general
