deleting lines from a file recrsivly

[gfunk123]

Hi

Is there any way from a starting directory (say /usr) to search for files with lines of text that contain the string "hello" and delete the whole line from the file (and ONLY that line)

I have a whole load of html files through a directory structure that have an unwanted line which i need to remove

thanks and regards
Gary

 

[PHV]

Something like this ?
find /usr -name '*.html' | while read f
do sed 'g/hello/d' "$f" >"$f.tmp" && mv "$f.tmp" "$f"
done

Hope This Help, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884

 

[gfunk123]

could you just explain whats happening on that sed line please, im a bit confused where the delete line command is

 

[vgersh99]

or something like that [a variation on the same theme]:

#!/bin/ksh
find . -type f | while read file; do ex - ${file} <<EOF
g/^hello/d
wq!
EOF
done;


vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[PHV]

Sorry for the typo:
sed '/hello/d'

Hope This Help, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884

 

[fredericofonseca]

the delete is the "/d"

man sed will help you, and if you intend to do real work on Unix then you really need to know at least the basics.



Regards

Frederico Fonseca
SysSoft Integrated Ltd

http://www.syssoft-int.com

 

[gfunk123]

I tried this

find /usr -name '*.html' | while read f
do sed '/hello/d' "$f" >"$f.tmp" && mv "$f.tmp" "$f"
done

and i get the following error

sed: command garbled: /var/apache/htdocs/index.html
+ read f
+ /hello/d


any ideas where im going wrong ( i put a carriage return after the first line (read f)

 

[PHV]

You may try this:
find /usr -name '*.html' | while read f
do sed -e '/hello/d' <"$f" >"$f.tmp" && mv "$f.tmp" "$f"
done
BTW are you sure the sed command is '/hello/d' ?

Hope This Help, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884

 

[PHV]

As far as you don't use special shell character in the sed script you can use either single quotes (') or double quotes (") like this:
sed -e "/_files/d" </path/to/input >/path/to/output

Hope This Help, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884

 

[toolkit]

Or try this:

find /usr -type f -name "*.html" | xargs perl -i -n -e 'print unless /hello/;'

Note, try this out with [tt]-i.bak[/tt] first, which will create a backup file.
Cheers, Neil