| |
| |
| |
|
Comments:
<0> in terms of regular expression support, is grep and sed same? :)
<1> i have two columns of texts, with lines like 100 dolphi-100, other line like 200 text-200.. i am looking for a way to transform this into 100 dolphi-x, 200 text-x .. i want to substitute the first word in second word, anyone knows how?
<2> and the "x" is always the same?
<1> yes
<1> but not the 100, 200..
<2> yep
<2> echo 100 dolphi-100 | sed 's/-.*$/-blahblah/'e
<2> echo 100 dolphi-100 | sed 's/-.*$/-blahblah/'
<1> but the 100 might be in any part of word..
<2> if i understood the point
<2> so you only want to substitute the part after - which is the same with first column?
<1> i may get a line like 100 www.test.com/100.jpg
<2> and the result
<2> like 100 www.test.com/x.jpg
<2> ?
<1> no, the - generally will not be present. maybe i used a bad example
<1> yes
<1> i may do it in shell scripting..
<2> echo 100 dolphi-100.jpg | sed 's/^\([^[:blank:]]*\)\([[:blank:]]* [^[:blank:]]*\)\1\([^[:blank:]]*\)$/\1\2x\3/'
<1> hmmmmm...
<1> thanks o__o
<2> nz
<0> helloman: do you need -r for [:blank:] ?
<2> no, i have GNU sed 4.1.4
<0> ok
<2> I need -r for \|
<1> how can i delet unmatched lines? currently i am doing a grep before the sed, to sed only matched lines..
<1> s/delet/delete/
<1> for unmatched lines, i meant.. unmatched in a substituition. lines that doesn't suffered a substituition.. hmm..
<2> /^How it looks like$/d
<0> !d
<2> :)) thats true
<2> /^How it doesnt look like$/d
<0> !d ?
<0> ops :)
<2> /^How it looks like$/!d
<2> well
<2> :)
<1> like.. sed 's/regex/sub; /regex/!d' ? hmmmm..
<0> dark_light: like, forget the s
<0> dark_light: like, only use what helloman showed.
<1> but i want to substitute the regex if encountered
<1> and delete lines that it isn't present
<0> rite. do it in two steps then.
<2> so delete the unmatched lines before
<0> first get rid of bad lines, then do a s/this/that/g ;)
<0> hmm, did -r support non-greedy flags?
<1> you mean, sed '/reged/!d' | sed 's/regex/sub/' ?
<2> no, sed '/regex/!d;s/regex/sub/'
<1> but it's what i am doing with grep, and it's ugly.. why you are recommending this?
<0> why is it ugly ;)
<2> or
<2> no, sed -n 's/regex/sub/p'
<1> helloman, ah
<1> o.o
<2> this prints only substituted lines
<1> hmm, nice
<0> sed be fantastico. ;)
<2> everyone loves it!
<0> new star of the town ;)
<2> :)
<0> hulloh redduck666
<0> hmm, is the sed faq updated?
<0> regarding gnu sed 3.02 3.02.80 "The N command no longer discards the contents of the pattern space upon reaching the end of file. This is not a bug, it's a feature. However, it breaks certain scripts which relied on the older behavior of N."
<0> was this "fix" only for those two versions?
<0> nevermind
<0> http://pastebin.ca/99404 <-- althought it's rather oxymoronic script, why does it not work? ;)
<2> what is it supposed to do?
<0> put every line in hold space, when reaching the end of file print holdspace
<0> but it appears to become and infinite loop
<0> sedsed claims that it cannot find the label to jump to
<2> it looks for me as if H adds always the first line
<2> this works http://pastebin.ca/99419
<0> strange
<0> the n command is indeed a little bit confusing
<3> Astinus-: read the next line of input (which will change the line number also -> = )
<0> m4n: well...
<4> Hiho. I have a problem trying to use sed for search and replace on an IP address. Do I need to escape the . in the "replacement" string as well?
<5> yes
<4> gnubien: Is there an easy (commandline way) how I can get an IP address escaped or do I need to do some substring stuff?
<5> swissmade: post an example of what you want sed to do
<4> something like: car myfile | sed \s/196\.168\.0\.5/$dynamically_acquired_ip/g > new_myfile
<4> oops s/car/cat
<5> ok
<4> gnubien: I get $dynamically_acquired_ip as a normal string, but it's a normally formatted IP address. Like 192.168.0.8 for example.
<6> hum...how to output the line,the next 2 lines and the previous 2 lines which match the expression ?
<6> thx!
<5> swissmade: dynamically_acquired_ip="1.1.1.1"; echo "196.168.0.5" |sed 's/196\.168\.0\.5/'"$dynamically_acquired_ip"'/g' #1.1.1.1
<5> salamanders http://www.student.northpark.edu/pemente/sed/sed1line.txt AND sed.sf.net
<5> salamanders: grep grep in sed1line
<6> have greped...:(
<5> salamanders: sed -n -e '/regexp/{=;x;1!p;g;$!N;p;D;}' -e h # print 1 line of context before and after regexp, with line number indicating where the regexp occurred (similar to "grep -A1 -B1")
<4> gnubien: Will try that. Thx!
<6> gnubien:Will try that. Thx! :D
<4> gnubien: Hmm.. I tried all sorts of using double quotes, single quotes and ` but nothing works.
<5> swissmade: does my example work for you?
<4> I think I got it.
<4> Yes, the example works.
<5> '"$dynamically_acquired_ip"' note the quotes
<4> Yes.. I did note that.
<5> ok, still have a problem?
<4> The script has to be run as root, but I experimented not being root. And there's a permission error in the log output.
<5> run the script as root?
<4> Ah.. nope. Now it run fine, but it has put in $dynamically_acquired_ip literally.
<5> making progress...
<4> http://rafb.net/paste/results/C1eqvS31.html
<4> Want the full script?
<5> whats the problem now?
<4> It replaces the old IP with $2 literally.
<5> ok, no problems now?
<4> No.. it doesn't put in 212.96.136.197 it put's in $2 (the string dollar-2 not the variable value).
<4> I want to replace the IP address 212.96.128.171 in the file with the second argument from calling the script.
<5> swissmade: ask in #bash ?
<4> gnubien: OK :)
<5> swissmade: greycat in #bash means but $2 in quotes = "$2"
<4> So remove the ' around "$2"?
<5> swissmade: maybe cat /sites/"$1"/zope/etc/zope.conf
<5> try putting all the $1's in "$1" = quotes
<4> Ha! removing the ' seems to work.
<5> test it many times to see
<4> I am. But it works now. (4 tests so far)
<5> cool
<4> Thanks! :)
<5> have fun ;)
<4> To be honest, bash scripting is not exactly my idea of fun. ;)
<5> do what you like to do, more fun that way ;)
<4> gnubien: That won't pay the bills
<5> do what you love to do and find a way to get paid for it and you get paid twice ;)
<4> gnubien: I tired that. Didn't work out. Possibly I'm not that good at what I love to do that people would pay enough.
<4> (then again, I'm not that good at what I get paid for either. ;)
<4> Sad existence, hu?
<5> swissmade: follow your heart and do what you love cause the payback is better than money
<4> gnubien: How old are you? :)
<5> old enough to know better but young enough to be tempted to try it ;)
<4> gnubien: I haven't lost all idealism (I'm trying to build my own company with my wife) ..
<4> but my experience following that ideal has been rather sobering.
<5> swissmade: yes, steep learning curve in life; simple life is the best imho
<4> gnubien: Have you tried it? I have a vegetable path. Nice for stress relieve but I would not want to have my survival depend on it. ;)
<4> s/path/patch
<5> sure, tried it, liked it, live it
<5> swissmade: our bodys need more than vegetables like fish oil, vitamin b, etc
<4> I know.. not a vegetarian myself. ;
<4> Gotta run! Nice meeting you and thanks again!
<5> bye
Return to
#sed
or
Go to some related
logs:
#perl
javascript SyntaxError: missing ) in parenthetical
#php
#php
tzap error while parsing inversion
#linux
#math
#gimp
startfluxbox terminate called after aborted fluxbox
acer 5670 lirc
|
|