DataMuseum.dk

top - metrics - download

    Length: 1386 (0x56a)
    Types: TextFile
    Names: »README«

└─⟦b20c6495f⟧ Bits:30007238 EUUGD18: Wien-båndet, efterår 1987
    └─⟦this⟧ »EUUGD18/General/Ncode/README« 

TextFile

You should now have README, ncode.c, ncode.1 ncode.sample.1 and ncode.sample.2.

The program will now handle arbitrarily large files (you can even #include
/usr/dict/words) and runs much quicker than the first one I posted.

This source is known to compile and run on 4.3 BSD and on Sun 3.0.  I
understand that it compiles fine on sysV if you replace the calls to random()
and srandom() with calls to rand() and srand().  Non 4.3 users will probably
also have to replace the innards of a_random_number() with something that
works on your system.  You basically want some number to seed the random
number generator.  A getpid() doesn't change very quickly but will work;
calls to the real-time clock work better.  This version of a_random_number()
was written by pixar!brighton who insists that all the baroque bit twiddling
helps.  I just know that it works.

Compile it "cc ncode.c -O -o ncode" and then try

	ncode ncode.sample.1

For more fun, try

	ncode -n 5 ncode.sample.1

And for a good education say

	ncode -g "I had a date with FOX\" ncode.sample.1

Then you might want to read the man page, ncode.1, to learn about how it
works.  You can make it pretty with "nroff -man ncode.1".  If all of this
leaves you hopelessly confused, send mail to {sun,ucbvax}!pixar!good and
I'll send you more confusing sample files.  If I feel like it.


		--Craig
		...{ucbvax,sun}!pixar!good