Open GoogleCodeExporter opened 9 years ago
That's controversial as that's a non-standard, experimental and expired
(January 29, 2009) draft:
http://www.omz13.com/downloads/draft-somers-ftp-mfxx-04.html
...but on the other hand I see proftpd implements MFF and MFMT by default:
http://www.proftpd.org/docs/modules/mod_facts.html
AFAIK, there's no such thing as file creation time on Linux:
http://unix.stackexchange.com/questions/24441/get-file-created-creation-time
http://www.cyberciti.biz/tips/understanding-unixlinux-filesystem-inodes.html
http://stackoverflow.com/questions/5929419/how-to-get-file-creation-date-in-linu
x
File modification time is available; how would you change it though? You'd open
the file in append mode and then close it?
Also I'm wondering: why would anybody want to change file create/modification
time in the first place?
Original comment by g.rodola
on 21 Jan 2012 at 5:38
As well, FileZilla supports setting mtme using MFMT.
You are correct here of course. Changing the mtime is really the only
useful part of this spec.
You can set the time by using os.utime().
The common use-case is file synchronization. The sync software will
set the modification time on the server to match that of the client.
Then the sync software can compare times to see if local or remote
changes have occurred (the mtime will become out-of-sync if either
side is modified).
Original comment by btimby@gmail.com
on 21 Jan 2012 at 6:12
Since os.utime appears to be present on both POSIX and Windows, and given the
you case you described since reasonable, I think we should implement this (MFMT
/ edit file modification time).
I looking at the spec and it's not clear to me what MFF is for.
Original comment by g.rodola
on 21 Jan 2012 at 6:23
I think for modifying more than one fact at a time. However, since
MFCT is not useful on POSIX systems (or at least Linux) we are
probably not interested in it or MFF.
Original comment by btimby@gmail.com
on 21 Jan 2012 at 6:36
Initial patch implementing MFMT command is in attachement.
Have to look into into it further as the unittest fails though.
Original comment by g.rodola
on 29 Feb 2012 at 3:54
Attachments:
Original issue reported on code.google.com by
btimby@gmail.com
on 20 Jan 2012 at 9:47