aboutsummaryrefslogtreecommitdiff
path: root/textcomp.c
diff options
context:
space:
mode:
Diffstat (limited to 'textcomp.c')
-rw-r--r--textcomp.c9
1 files changed, 4 insertions, 5 deletions
diff --git a/textcomp.c b/textcomp.c
index 2412f15..7139225 100644
--- a/textcomp.c
+++ b/textcomp.c
@@ -66,11 +66,10 @@
start of the next line, so no \r's are needed. Any \r sequences listed
in messages.pl are discarded before encoding is done.
- Actually, no prompts ever use capital X or Z. These should be used for
- dictionary lookups. Maybe X is followed by a 3-bit dict ID, for the 8
- most commonly repeated phrases (one of which will of course be "Taipan"),
- and Z is a 5- or 6-bit ID for 32 or 64 less common phrases. So far this
- isn't implemented because the decompressor isn't reentrant (yet).
+ Since no prompts ever use capital Z, it's used as an escape character
+ for dictionary lookups (e.g. Za = "Li Yuen", Zb = "Elder Brother").
+ This program doesn't do that, it's done by messages.pl, and textdecomp.s
+ does the dictionary extraction.
*/
#include <stdio.h>