Hi again ^^ On 27 June 2014 23:20, Leam Hall <leamh...@gmail.com> wrote: > On 06/27/14 17:40, Walter Prins wrote: >> On 26 June 2014 18:01, leam hall <leamh...@gmail.com> wrote: >>> On Thu, Jun 26, 2014 at 12:41 PM, Walter Prins <wpr...@gmail.com> wrote: >>>> On 26 June 2014 14:39, leam hall <leamh...@gmail.com> wrote: >>>>> Python 2.4.3 >>>>> Is there a better way to do this? >>>> I'd probably rather try Paramiko's SFTPClient and retrieve the file >>> Seem to not work on Python 2.4.3. >> What exactly did you try, and what error/output did you get? (I've >> had another look at Paramiko's homepage and it seems to support back >> to Python 2.2? [http://www.lag.net/paramiko/ ]) > Walter, I was cranky. Sorry.
Not a problem, know the feeling. :) > When I looked at ( http://www.paramiko.org ) it specified Python 2.6+ and > 3.3+. That's echoed in the README ( > https://github.com/paramiko/paramiko/blob/master/README ) I have not > actually tested to see what breaks on a 2.4 box. Hmmm... OK. Interesting. Perhaps Paramiko head doesn't support it anymore on head in github... but it might still be possible to get an older version for Python 2.4, but I guess (having read the rest of your post already) that maybe this is not the best course of action for your situation -- I'll stop pushing this idea shortly, I promise! Nevertheless, continuing on... > The other option was to copy in the files and stat them. Trying to build a > script that can be extended and not hit so much network bandwidth. OK -- just to note, I think perhaps another small misunderstanding has slipped in somewhere. My suggestion isn't to actually retrieve the file *contents*, but rather to retrieve information *about* the file on the remote system. That means only the remote stats about the file travel over the wire, not the file contents itself. Hence it should be approximate as light, or lighter (give or take) than running an ssh command and retrieving the output I'd guess. > The good > thing about the way this is currently going is that a co-worker gave me a > better Solaris solution that let me standardize the script for both Red Hat > Linux and Solaris. Still need to figure out AIX. OK, I get that would be attractive for your use case. :) > The possibility exists to build a python outside of the standard. The issue > is supportability. Like many others I'm a contractor and need to make sure > whatever I do is so boringly simple that I can train my replacement. On one > hand, that's a good thing. If people use my scripts after I'm gone then > hopefully they won't cuss too much. I get that as well. (A big complicated custom Python + Paramiko install is rather not the simplest/most maintainable solution when all you want is to know if a given file is of a certain age... especially if the in-house Python knowledge is limited and you need to hand it over to in-house staff.) > Since my exciting plans for the weekend have been canned I'll take a little > time and think about how best to do this. Some of it is "what should I do > with my career" type thinking. Some is just "hey, how does this work?". Good luck... need to do some of the same sometime (e.g. what should I do with my career), but I digress. :) > Thanks for not responding as poorly as I did. No worries, hope you have a good weekend despite your plans being canned. Walter _______________________________________________ Tutor maillist - Tutor@python.org To unsubscribe or change subscription options: https://mail.python.org/mailman/listinfo/tutor
