[bash / sed] remove all data between < >

Ljohan · 03-14-2006, 10:32 AM

Dear all,
question, I like to remove all data in a file which is between "<" and ">" and replace this with ||.

Example of original data:

fghgdhdfs<dhsj>hjsdhajk<dshj>dhsjk
sdsd<hh>dhsjyuiw<hjsd>hdjkshjkasjdks<dhsj>

this should become:
fghgdhdfs<||>hjsdhajk<||>dhsjk
sdsd<||>dhsjyuiw<||>hdjkshjkasjdks<||>

any good hint on how to do this?

Regards,
Johan.

Andrew Benton · 03-14-2006, 11:12 AM

If you put the text in a file called test.txt

Code:

sed -i 's,<[a-z]*>,<||>,g' test.txt

/bin/bash · 03-14-2006, 09:49 PM

http://sed.sourceforge.net/grabbag/tutorials/

Ljohan · 03-15-2006, 04:21 AM

Quote:

Originally Posted by Andrew Benton

If you put the text in a file called test.txt

Code:

sed -i 's,<[a-z]*>,<||>,g' test.txt

Thanks, seems to be working fine, however this is only working of I have data between < > which contains no special chars. If I have date end tags like </blabla> in the data it will mess up the entire data. I have tried some other options and tried to find a solution but the only thing I succeeded in was loosing a lot of sleep and I ended up with no solution.

Any good hints on how to handle this even in case in starts with </ or contains a "

Regards,
Johan.

jlliagre · 03-15-2006, 05:20 AM

Code:

sed 's/<[^>]*>/<||>/g' test.txt