Simple bash programming question

[ggelln]

I am trying to make a shell script that will manage
a plethora of cocktail recipes I have.

I have a favorites file which contains a list of absolute pathnames to cocktail recipes that the user has chosen for a given day.

The user enters a name of the cocktail that they would like to remove and the program then searches (grep) each recipe in his list for a match. If a match is found I would like the pathname to be removed from the file.

I'm not sure how to do this as I need to erase a line only if the file it points to contains the given condition . . .

Any tips or comments (even a different way of thinking about the solution, though no perl please ) would be greatly appreciated.

the code so far looks as follows (the sed line does not work)

[dannym]

Hi!

Instead of
sed 's/$r//' $LFAVORITES/${USER}.${DATE}
use:

grep -vi "^${r}$" $LFAVORITES/${USER}.${DATE}.tmp && mv $LFAVORITES/${USER}.${DATE}.tmp $LFAVORITES/${USER}.${DATE}

If you want to remove the string from $r on a line on its own from the given file.

[ggelln]

Wow thanks for your quick response

I used the code (modified slightly, 2nd argument to
grep had a .tmp expension which I don't think can work. I imagine it was just a typo, tell me if I'm wrong . . .)

I really like the using the inverse grep listing, I had always wondered what that would be usefull for, now I know!

If you have the inkling can you tell me what the
"^$r$" regex does exactly I know the ^ matches the beginning of the line, but what does the last $ do?

Thanks a lot.

Gabriel

[cram]