| Optimizing a lookup table [message #52806] |
Fri, 02 March 2007 17:15  |
ianpaul.freeley
Messages: 18
Registered: March 2007
|
Junior Member |
|
|
I would like to write a function that uses a look-up table. Now the
easiest (coding-wise), is to just save the look-up table and have the
function restore it. However, I plan to be calling this function in a
loop, so this would result in numerous unnecessary disk-reads.
Anyway, what is the most efficient way to get a lookup table into an
IDL function?
My thoughts so far:
1) Just print the look-up table to the screen, copy/paste into the
function and add some brackets and commas, presto, variable is in the
function and will stay loaded as long as the function is compiled.
This will work for my present problem, but would be unwieldy for
really large look-up tables and I worry about double-precision getting
truncated on the print.
2) have a procedure that reads in the table and puts it in a common
block, then just start my function with an if-statement to see if the
common block exists and if not, call the reading procedure. My
question here is, how can I check to see if a common block has already
been created? I know I could call the common block maker outside the
loop, but that seems lame and makes it more complicated if I want to
share the code.
Unless someone comes up with something really witty, I'll just use
option 1. Just seems like this would be a common problem that
someone's solved before.
cheers,
IP Freeley
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
