A couple of sed questions
in [Shell]
|
Prev: Fork Concept
Next: conditional grep-like command
From: nobody on 6 Oct 2009 18:40 Given the following sentence: This line has odd spaces in it How can I remove the extra spaces so the sentence reads: This line has odd spaces in it And secondly, how do you delete text BEFORE and AFTER a word? For example, This sentence has some words How would I delete everything before the word "has"? And delete everything after the word "has"? -Thanks
From: Chris F.A. Johnson on 6 Oct 2009 19:56 On 2009-10-06, nobody wrote: > Given the following sentence: > > This line has odd spaces in it > > How can I remove the extra spaces so the sentence reads: > > This line has odd spaces in it line="This line has odd spaces in it" set -f set -- $line newline=$* > And secondly, how do you delete text BEFORE and AFTER a word? > For example, > > This sentence has some words > > How would I delete everything before the word "has"? > And delete everything after the word "has"? line="This sentence has some words" left=${line%%has*} right=${line#*has} printf "%s\n" "${line#"$left"}" "${line%"$right"}" -- Chris F.A. Johnson, author <http://cfaj.freeshell.org/shell/> Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress) ===== My code in this post, if any, assumes the POSIX locale ===== and is released under the GNU General Public Licence
From: sharma__r on 6 Oct 2009 20:40 On Oct 7, 3:40 am, nobody <nob...(a)nowhere.com> wrote: > Given the following sentence: > > This line has odd spaces in it > > How can I remove the extra spaces so the sentence reads: > > This line has odd spaces in it > > And secondly, how do you delete text BEFORE and AFTER a word? > For example, > > This sentence has some words > > How would I delete everything before the word "has"? > And delete everything after the word "has"? > > -Thanks ## using the "tr" command to squeeze multiple spaces into one tr -s '\040' '\040' < yourfile ## using "sed" to squeeze multiple spaces into one sed -e 's/[ ][ ]*/ /g' yourfile ### deleting everything before a word: sed -e ' s/has/\ &/ s/.*\n// ' yourfile ### deleting everything after a word: sed -e ' s/has/&\ / s/\n.*// ' yourfile ### deleting everything before and after a word: sed -e ' s/has/\ &\ / s/.*\n\(.*\)\n.*/\1/ ' yourfile Note: Try these on a bourne shell command line. For a c-shell -based commandline the backslashes need to double. -- Rakesh
From: Richard C. Giles on 6 Oct 2009 22:28 On Oct 6, 8:40 pm, sharma...(a)hotmail.com wrote: > ### deleting everything before and after a word: > sed -e ' > s/has/\ > &\ > / > s/.*\n\(.*\)\n.*/\1/ > ' yourfile > > Note: Try these on a bourne shell command line. For a c-shell -based > commandline the backslashes need to double. > > -- Rakesh It would be a lot simpler to use: $ sed -e 's/.*\(has\).*/\1/' yourfile Why add tokens and additional statements when the base syntax will do the task. -Arcege
From: Ed Morton on 6 Oct 2009 23:01
Richard C. Giles wrote: > On Oct 6, 8:40 pm, sharma...(a)hotmail.com wrote: >> ### deleting everything before and after a word: >> sed -e ' >> s/has/\ >> &\ >> / >> s/.*\n\(.*\)\n.*/\1/ >> ' yourfile >> >> Note: Try these on a bourne shell command line. For a c-shell -based >> commandline the backslashes need to double. >> >> -- Rakesh > > It would be a lot simpler to use: > $ sed -e 's/.*\(has\).*/\1/' yourfile > > Why add tokens and additional statements when the base syntax will do > the task. Good question! Neither of them work, though, as they both look for a string "has" instead of a word "has": $ cat file lets chastise this effort $ sed -e ' s/has/\ &\ / s/.*\n\(.*\)\n.*/\1/ ' file has $ sed -e 's/.*\(has\).*/\1/' file has I believe GNU sed has some start/end of word delimiters but if all the OP actually wants is the word "has" printed there are simpler ways to do that! Regards, Ed. |