[CLUG] He SED she SED we SED that they SED.....

[Adam Lock]

Assuming this is a one-off task, a method I often use is to open XEmacs
(or Emacs), and record a macro that does one line and replay that for
the remainder.

e.g.

1. Start recording - C-X (
2. Go to start of line - C-A
3. Delete the first bunch of words - ESC-D, ESC-D, ESC-D, C-D
4. Search for the next quote - C-S "
5. Kill the rest of the line and clean up spaces and quotes - C-K, DEL,
DEL
6. Down to next line (C-N)
7. Stop recording - C-X )

Run the macro for the remaining lines & save. Obviously if you need to
keep running the script it's better off using a command-line tool.


Baldwin_James at emc.com wrote:
> 
> Lads,
> 
> Quick question on Sed..I am brain-dead today...
> 
> I have a file for eg. below and want to keep only 1 column of data and
> delete the rest before and after that column;
> HOW?  Using SED?  How about a pre-filter of Grep before SED?
> (so, I want to make a file list and cut the rest of the data out)
> 
> BEFORE FILE:
> 
> Cannot Process File /usr/bin/12765 "The file is completely arsed. Can't
> access"
> Cannot Process File /usr/bin/123244 "The file is completely arsed. Can't
> access"
> Cannot Process File /usr/bin/134 "The file is completely arsed. Can't
> access"
> Cannot Process File /usr/bin/4334345 "The file is completely arsed. Can't
> access"
> Cannot Process File /usr/bin/12733 "The file is completely arsed. Can't
> access"
> Cannot Process File /usr/bin/1245981115 "The file is completely arsed. Can't
> access"
> 
> AFTER FILE:
> /usr/bin/12765
> /usr/bin/123244
> /usr/bin/134
> ....etc
> ....etc
> ....etc....
> 
> THANKS...............J.!
> 
> _______________________________________________
> Cork maillist  -  Cork at linux.ie
> http://www.linux.ie/mailman/listinfo/cork


-- 
Adam Lock - locka at iol.ie