> How about
> cat log|grep -v -E [[:alnum]]'{2096}'>> log.bak
OK< but you can miss the cat out
grep -v -E [[:alnum]]'{2096}' log >> log.bak
But I confess I've no idea how that works, I've never
seen that notation in a grep before! Checking man reveals
an "extended regex" which I interpret as:
any alphanumeric repeated 2096 times?
But I'm not sure about my interpretation...
> The issue is - will unix shell command be any more
> efficient than a python script??
Usually, if you only use a single process like grep
because they are written in C. But when you pipeline
with cat it might slow it down enough to make Python
competitive. Process startup tends to be the limiting
factor in shell pipelines.
> Also i used append because i gathered that the user
> will not want to erase the previous logs. He is free
> to use a single > if he does.
Interesting, I assumed he would want to delete the
old logs... As you say, whichever is appropriate is
easily controlled.
Alan G.
_______________________________________________
Tutor maillist - Tutor@python.org
http://mail.python.org/mailman/listinfo/tutor
