| |
| |
| |
|
Page: 1 2
Comments:
<0> Any good suggestions for dealing with elapsed time stamps (ddd-HH:MM:SS format) and converting to something standard and parseable (elapsed seconds or elapsed minutes would be good).
<1> Well, the lazy thing to do is to make all of the seperators the same.
<1> So that is what I would do first.
<0> newmanbe: Point. At the same time I could just parse it out.
<0> Hrm.
<1> Give an example input and output.
<2> karsten: date -d "22 Fri Oct 22 16:15:08 2004" +%s
<0> root 1 0 S 0.0 00:00:40 90-00:28:25 476 3636 ? init
<0> Column 7
<0> I'm tracking down long-lived processes on a grid.
<0> For a given list of users, there shouldn't be any processes without a seperate (batch submission system) job entry.
<0> ... which is timestamped eg: "Feb 8 16:14"
<0> ... I need to compare timestamp w/ elapsed time for any processes for which the user _does_ have a current job entry.
<0> First output si from 'ps' with a format string supplied. I've got an existing shell script which ensures output is largely uniform across different hosts (varying 'Nix OSs, architectures).
<3> karsten: Easy.
<3> karsten: How you doin'?
<0> pr3d4t0r: Better than Vickie Lynn Hogan
<0> pr3d4t0r: Better known as Anna Nichole.
<4> ah
<4> Gold Digger
<0> goldfish: Dead digger
<4> She died?
<0> http://news.google.com/?ncl=1113457815&hl=en
<4> haha
<4> Deadly.
<4> I wonder who will get her money.
<0> The guy who was with her when she died, perhaps.
<4> Interesting.
<0> ... who was also with her when her son died in the bahamas.
<0> Howard K. Stern.
<0> No, not _that_ Howard Stern.
<0> Lawyer, 37.
<4> I don't know who that one is :)
<4> Name is familair though.
<0> Wikipedia article.
<4> It would be very ironic if the son of her old husband got all the money.
<0> http://en.wikipedia.org/wiki/Howard_K._Stern
<4> :-)
<0> Particularly as he's dead.
<4> haha
<4> I should keep up more with america.
<0> Truth to tell, there's bits of this I've only become aware of in the past hour.
<0> Weird part is I got a call this morning for a Vickie Lynn H. Different last name though. Wrong number.
<4> lol
<0> But right after I typed in the response to pr3d4t0r above I had a freak moment. Just replayed the voicemail, it's not the same name.
<4> :)
<0> Still pretty random though.
<3> Such a tragic life.
<4> lol.
<5> hehe
<0> Innit jes?
<3> http://img172.imageshack.us/img172/6959/1157960476967zi8.jpg
<4> :-)
<0> SFW?
<0> Needs a "I"m Unix"...
<6> pr3d4t0r: you might want to add some .htaccess mod_rewrite stuff. so you could access the post with http://www.eugeneciurana.com/pastebin/2 http://www.eugeneciurana.com/pastebin/pastebin.php?show=2 or however?
<6> It woudn't be a big deal. well just in case
<6> ah well, it's not that much shorter so
<3> calmar: I'll check on that.
<3> calmar: Maybe a link to cime.net/p or something (another one of my domains).
<6> pr3d4t0r: http://corz.org/serv/tricks/htaccess2.php there are some examples
<6> the one where he redirects the blog thing maybe
<7> awk '/Iowa/,/Montana/' prints the lines between those 2 expressions - how would I print everything but that pattern space ?
<7> I thought something like: '!/Iowa/,/Montana/' but obviously doesn't work
<4> The pattern1, pattern2 form of an expression is called a range pattern. It matches all input records starting with a record that matches pattern1, and continuing until a record
<4> that matches pattern2, inclusive. It does not combine with any other sort of pattern expression.
<4> Does it have to be awk?
<4> awk '/Iowa/{i++}/Montana/{i--;next}(!i){print}'
<4> that will break if Iowa occours more than once
<4> it was just a simple example
<4> bleh
<4> nm
<8> L|NUX
<9> hey
<8> yo join #xmb
<9> ok
<10> L|NUX there
<11> how in awk can i print lines starting from the last 4?
<12> Zap-W: always keep trcak of the last four lines and print them when you hit the end
<11> sounds complicated
<12> Zap-W: you know about the tail command, right?
<11> hey
<11> back
<11> yes, i want to use all in awk without tail
<4> awk '{i[j++]=$0} END{k=j-4;for(k;k<j;k++){print i[k]}}'
<6> printing $2...$NF , would anybody have an idea about that? I use a loop actually. hm http://tinyurl.com/25mwlm
<4> nothing at that url
<4> for loop, yeah
<4> or use cut.
<6> http://www.eugeneciurana.com/pastebin/pastebin.php?show=2171 (that one)
<4> nope
<4> cut(1) :)
<4> cut -f2-
<6> oh, I see. ah ok, then the loop probably is ok :) thanks
<4> yeah im not getting anything at that url
<4> awk '{print substr($0,length($1)+2)}'
<4> ***uming your field seperator is 1 char
<6> http://pastebin.ca/346615 ;)
<6> I see. that would be an option. I do that (the second time where I need it), thx a lot
<13> hi!
<6> goldfish: works nice, thanks!
<6> hi LZMA
<13> If I do a diff of two files, theres always the ">" infront of the important part . eg: "> sda1" how can i remove the ">"?
<6> LZMA: what do you want to do actually?
<11> bah gotta go workout
<6> sed 's/^>//' but well
<13> diff /bla.txt /blaa.txt has the output "> sda1" Then I want to use awk to remove the > and the space in front of sda1
<13> so that the output is "sda1"
<6> awk '{ sub(/^> /,""); print;}'
<6> (or only those files beginning with > maybe?
<6> )
<13> calmar, perfect! that's what I needed! Thank you
<6> hm, all right
<13> bye
<8> goldfish, awk '{ sub(/^/, "\x01", $4 ) ; sub(/.*\x01/, "" ) ; print }'
<4> What's this?!
<8> to print from $4+
<3> sayajin-god: HeH.
<14> gg
<3> sayajin-god: That looks over-engineered.
<14> its the fastest best way
<3> sayajin-god: Do you have a resume for me in PDF format that I can shop around? I have my monthly call with CS on Tue or Wed next week.
<14> mirc ****s, doesnt even highlight.. its crap written by noobs for serval reasons
<14> a friend should have made one, but he went
<3> Anyway... time to leave. Business demands my attention.
<3> Cheers!
<14> gg, laters
<4> fastest? :-)
<14> maybe, try benchmarking
<14> you need printf "%s%s", $field, OFS
<14> and removing the last OFS, is difficult
<14> so you need to build a string within the loop, then print it
<14> try, good experience for ya
<14> so you need a substr op or sub op after the loop
<6> or maybe printing up the next to last and printing the last manually?
<6> awk '{ sub(/^/, "\x01", $4 ) ; sub(/.*\x01/, "" ) ; print }' real 0m1.385s
<6> aeh, wrong
<6> file: 1 2 3 4 ...20
<6> time awk '{for (i=10;i< (NF-1);i++){ printf ("%s%s",$i, OFS);}print $NF}' dddd real: 0m1.385s
<6> time awk '{ sub(/^/, "\x01", $10 ) ; sub(/.*\x01/, "" ) ; print }' dddd real: 0m0.604s
<6> file contains 32000 lines so
<4> awk '{print substr($0,length($1)+2)}'
<4> try that way
<6> ok
<4> replace $1 with whatever
<6> time awk '{print substr($0,length($1)+2)}' dddd real: 0.610s
Return to
#awk
or
Go to some related
logs:
#web
kmilo ubuntu nvram
#fedora
opencubicplayer ubuntu
perl -e fork while fork
#lisp
postfix chroot sasl unknown password verifier
#fedora
Xorg 7.2 could not find default font
perl shell escape
|
|
