Read and Analyze Giant Files

[andu]

--On Thursday, November 07, 2002 18:42:42 -1000 Sannyasin Sivakatirswami 
<katir at hindu.org> wrote:


> metacard, 13 lines
>
># !/usr/local/bin/mc
> on startup
>   put empty into the_message
>   put 0 into the_counter
>   read from stdin until empty
>   put it into the_message
>   repeat for each line this_line in the_message
>     if (this_line contains "mystic_mouse") then
>       put the_counter + 1 into the_counter
>     end if
>   end repeat
>   put the_counter
> end startup
>
>
>
> is there a more efficient way in Transcript to do this?
>

In Transcript maybe not but you can try metatalk;-).
First of all, I don't understand how you feed the file to the script in 
your script (read from stdin until empty?). Since the file is so huge you 
should probably use something like:

open file x for read
read from file x at <offset> for <count>
# so that you don't have to load 300MB into ram unless you want to.
# do a
repeat for each line x in it
 get offset("mystic_mouse",i)
if it is not 0 then add 1 to the_counter


Just an other possibility... Even better, have the server write a log each 
day to keep it shorter (unless it's google;-).


Regards, Andu Novac