RunRev compilation (was another "vs" thread)

[Stephen Barncard]

Hey, Richard, I'd give Tom Pittman many kudos for inventing CompilIt! 
- sure, it was convoluted concept, but it saved my ass many times in 
a working environment. In 1989 I did a lot with text string 
manipulation and hex-decimal conversion in software designed to go 
through a serial port to some hardware for control.

http://barncard.com/amstudios/htdoc/Pages/welcome_PP.html


  Hypertalk, even the improved versions later was still way too slow. 
So I used both Compilit! and Hyperbasic to make XCMDs that really 
sped up workflow. Hyperbasic was preferred because they had the 
better serial port code, which I had to use in a loop. Interperted 
hypertalk was way too slow. But Compilit was great for complex text 
manipulation.

I had never heard of the 'unreleased derivative' you speak of... interesting.

The 'funky syntax' you mentioned wasn't that hard - most of it was 
optimizing loops and helping the compiler recognize integers (by 
adding zero), the rest ran quite automatically. I thought it was 
brilliant.

sqb


>kee nethery wrote:
>>long ago in Hypercard land there was a product that would compile 
>>hypertalk into an XFCN or XCMD. I would not be surprised if someone 
>>on this list wrote it. Perhaps it is time for them to update and 
>>productize it?
>
>That was CompileIt!, written by Tom Pittman.  I enjoyed working with 
>an unreleased derivative of it made by Mark Hanrek, which automated 
>a lot of the symbol linking and stuff that made Pittman's interface 
>nearly impossible to work with.
>
>On top of that, since xTalk was never designed for compilation you 
>had to do a lot of funky syntax to get any reasonable compilation. 
>In retrospect, working with C is in many ways simpler than the odd 
>Symbol Table-HyperTalk mish-mash that CompileIt! required.
>
>
>--
>  Richard Gaskin