Landon Curt Noll
81b4f460b7
The `display_as` field in an `.entry_json` file must match the regex `^[a-z][0-9a-z]*$` now. This is needed to reduce errors during the final round while converting a potential submission to entry form. Modify `1992/buzzard.2/.entry.json` and `1993/rince/.entry.json` to change `display_as` field to use lower case. Correct `2012/omoikane/.entry.json` to correctly reflect changes made by commit2955e5d07eand commit2955e5d07e. Change `bin/cvt-submission.sh` to verify that the `display_as` matches the above regex. Changed `bin/cvt-submission.sh` VERSION from "2.2.2 2025-06-30" to "2.3.0 2025-07-02". Rebuild all tarballs, both at the entry and the year level due recent changes to `var.mk`, `1337.mk`, and `ioccc.css`. Rebuild sitemap due to the changes above. Ran `make update` to test and perform the above.
To build:
make all
To use:
./cmills [starting_cash]
Try:
./cmills 5000
Judges' remarks:
The Makefile uses several -D defines. The following is is a list of the macros
that may be changed:
-DM=<max bet allowed>-DN=<52 * number of decks used>-DX=<how to seed the rand() generator>
The macro V should remain void.
The author noted that a compiler being used assumed that x=*p++ means x*=p++
but do you know why? This is not so strange as it might seem if you look at C's
history. Example code where
this occurs is in John Lions' Commentary on the Sixth Edition UNIX Operating System.
Author's remarks:
This is a blackjack program. It does splits, doubling down, insurance, etc.
As best I can figure, it should be ANSI-compliant, as well as portable to BSD &
SysV (as well as the irritating compiler I was using which assumes x=*p++
means x*=p++!). It is reasonably robust in its error handling---it won't let
you split, double, etc. unless you actually have the cash, etc. To terminate
the program, hit EOF or type 0 or a negative number for you wager.
The way it deals with splits is particularly cool (IMHO). You can split many times, and it will show all hands as they are played.