Hi! I'm working on an update of my printtok.c module that will use a modified version of Infocom's ZSCII technique and tokenization to compress text. The current version only uses tokenization, is inferior and only helped a text adventure's text by about 13%. While I'm working on it, does anybody here have any ideas on how I can better compress text such that each string doesn't have to reference a previous string?
What is this module and how is it related to the Apple II?
printtok.c is a cc65 C module that prints strings with embedded tokens and expands those tokens to full strings. It currently only works okay--about 13% on one of my text adventures' strings--and requires manual compression. I have versions for CBM and Atari8 models, but the CBM one should work on an Apple2. It is at c65 additions - Manage /ui at SourceForge.net.
I am not sure this is the best forum for what is essentially a C programming question or request for code review. Isn’t there a cc65 forum?
Well, it's kind of on the edge. cc65 is a programming tool that can target Apple II, so it isn't completely irrelevant, however this question is generic enough it puts it into a grey area. Its applicability here depends on how specific to Apple II the issue is. That's why I didn't have a problem with questions related to cc65 generating binaries without the DOS3.3 header for use with ProDOS. That's clearly Apple II related. But if this is about a library that is multi-platform then it is less relevant.
You're both right. It is not Apple2-specific. However, it should work on an Apple2. That's why I posted it here.