slj@mtung.ATT.COM (S. Luke Jones) (12/06/88)
Here is my problem:
I have a file composed of multi-line records.
Continuation lines begin with arbitrary white space.
I want to extract *some* of the records.
for example:
In this file, the "Here is my..." and "for example" lines
begin multi-line records. Suppose the records I want
are those that begin with an uppercase letter ("Here...")
sed -n -e '/^[A-Z]/,/^[^(spc)(tab)]/p' file
will print each such record, but would also get the "for..."
line of the second record.
sed -n -e '/^[A-Z]/,/^[(spc)(tab)]/p' file
will not get "for..." but will only get the first following
line that begins with white space.
I've tried combinations and permutations of -n and '/pat/,/pat/!d'
and I give up.
Why oh why can't SED act like AWK does and *not* print this line???
--
S. Luke Jones
AT&T Infor#####Bell Labs
200 Laurel Avenue, Room MT 2E-337, Middletown, NJ 07748
slj@mtung.att.com -or- ...!att!mtung!slj (201)-957-2733maart@cs.vu.nl (Maarten Litmaath) (12/07/88)
slj@mtung.ATT.COM (S. Luke Jones) writes: \Here is my problem: \ I have a file composed of multi-line records. \ Continuation lines begin with arbitrary white space. \ I want to extract *some* of the records. \for example: \ In this file, the "Here is my..." and "for example" lines \ begin multi-line records. Suppose the records I want \ are those that begin with an uppercase letter ("Here...") \ sed -n -e '/^[A-Z]/,/^[^(spc)(tab)]/p' file \ will print each such record, but would also get the "for..." \ line of the second record. \ sed -n -e '/^[A-Z]/,/^[(spc)(tab)]/p' file \ will not get "for..." but will only get the first following \ line that begins with white space. #! /bin/sh exec sed -n -e ' /^[A-Z]/{ : loop p n /^[ ]/b loop /^$/ b loop } ' $* -- fcntl(fd, F_SETFL, FNDELAY): |Maarten Litmaath @ VU Amsterdam: let's go weepin' in the corner! |maart@cs.vu.nl, mcvax!botter!maart
pjh@otter.hpl.hp.com (Patrick Hyland) (12/08/88)
If you're willing to use '-f' rather than '-e' then, the following sedscript does what you want: #n : noprint s/^[A-Z]/&/ t print n b noprint : print p n s/^[ ]/&/ t print b noprint Patrick Hyland.
muller@mirsa.UUCP (Christophe Muller) (02/02/90)
That might be a stupid question, but does somebody know how to
do that job with "sed" ? I wanted to find all the lines matching a
pattern, delete these lines _and_ the line immediately following.
After half an hour of tests, man, Unix books, i gave up and used a
small awk script that worked well the first time ! I just thought that
sed might be faster, but is it possible ? (Note: we still don't have
perl...)
Cheers,
Christophe.
-- Important: e-mail address is: muller@phoenix.src.umd.edu --merlyn@iwarp.intel.com (Randal Schwartz) (02/02/90)
In article <520@mirsa.UUCP>, muller@mirsa (Christophe Muller) writes: | That might be a stupid question, but does somebody know how to | do that job with "sed" ? I wanted to find all the lines matching a | pattern, delete these lines _and_ the line immediately following. | | After half an hour of tests, man, Unix books, i gave up and used a | small awk script that worked well the first time ! I just thought that | sed might be faster, but is it possible ? (Note: we still don't have | perl...) Foreseeing a possible reply involving Perl means you've been following c.u.q for a while. Well, here's a sed solution: sed '/pattern/{N d }' <<EOF foo bar pattern bletch biz EOF prints 'foo' 'bar' and 'biz'. That should be a start. The important thing to remember is that 'd' makes you lose control of the current line, and the next line is restarted at the top of your script. So, get all your stuff together, and then do your 'd'. Just another sed hacker (when I'm not hacking Perl), -- /=Randal L. Schwartz, Stonehenge Consulting Services (503)777-0095 ==========\ | on contract to Intel's iWarp project, Beaverton, Oregon, USA, Sol III | | merlyn@iwarp.intel.com ...!any-MX-mailer-like-uunet!iwarp.intel.com!merlyn | \=Cute Quote: "Welcome to Portland, Oregon, home of the California Raisins!"=/
lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) (02/02/90)
In article <520@mirsa.UUCP> muller@mirsa.UUCP (Christophe Muller) writes: : I wanted to find all the lines matching a : pattern, delete these lines _and_ the line immediately following. : : After half an hour of tests, man, Unix books, i gave up and used a : small awk script that worked well the first time ! I just thought that : sed might be faster, but is it possible ? Certainly. /pattern/{ N d } If pattern matches, append the next line to the current line, and then discard the whole pattern buffer. : (Note: we still don't have perl...) It doesn't buy you much in this case, depending on the pattern. Searching for /window/ in /etc/termcap, perl beats sed by only .2 seconds, less than a 20% speedup. On a more complicated pattern, you might get more. Or less. That's using the perl command: time perl -pe '<>,$_ = "" if /window/;' /etc/termcap > /dev/null 1.0u 0.1s 0:02 40% 191+162k 23+5io 7pf+0w vs time sed -f window.sed /etc/termcap >/dev/null 1.2u 0.1s 0:02 56% 29+40k 24+0io 1pf+0w where window.sed contains the appropriate commands. awk weighs in at time awk '/window/{getline;next}{print}' /etc/termcap > /dev/null 2.3u 0.1s 0:03 72% 55+112k 22+0io 4pf+0w Larry Wall lwall@jpl-devvax.jpl.nasa.gov
maart@cs.vu.nl (Maarten Litmaath) (02/02/90)
In article <520@mirsa.UUCP>, muller@mirsa.UUCP (Christophe Muller) writes: \... I wanted to find all the lines matching a \pattern, delete these lines _and_ the line immediately following. [...] ----------8<----------8<----------8<----------8<----------8<---------- #!/bin/sh case $# in 0) echo "Usage: `basename $0` pattern [files]" >&2 exit 1 esac q=\\ pattern=`echo x"$1" | sed -e 's/.//' -e "s-/-$q$q/-g"` shift sed "/$pattern/,/.*/d" $* ----------8<----------8<----------8<----------8<----------8<---------- -- What do the following have in common: access(2), SysV echo, O_NONDELAY? | Maarten Litmaath @ VU Amsterdam: maart@cs.vu.nl, uunet!mcsun!botter!maart