> Just what *is* the state of the run-time system?  

Heavily exercised on Linux x86.  Lightly exercised on Linux PPC and ARM.
Not ported to other architecures or operating systems.

 > I assume Chapter 8 is implemented.

Yes.

 > Is any of chapter 11 implemented? 

None.  We do have implementations of the thread stuff, but I'm pretty
sure it doesn't yet confirm to the proposal in Chapter 11.

 > I've noticed that some of Section 11.1 is just a way of generating code 
 > into a file in a different order, to make life easier for "a small, 
 > simple compiler".

Yes.  But I'm pretty sure we no longer believe in the value of all of
Chapter 11.

 > I've concocted a data structure that effectively solves some of these 
 > issues, without requiring a lot of specialized data structures to keep 
 > track of things, or new features in the C-- language.
 > 
 > It's what I call an insertable string.  (I desperately need *short*, 
 > convincing meaningful words for the types and operations, and I haven't 
 > found them  Suggestions eagerly sought.)
 > 
 > An insertable string is a sequence of characters and insertion-points.  
 > It starts out with no characters and just an insertion point.
 > 
 > The operations on it are:
 >  
 > * to insert a sequence of characters immediately before an insertion 
 > point. 
 > 
 > * to create a new insertion point immediately before an existing 
 > insertion point.
 > 
 > * To take the entire insertable-string and write it out to a file.

Looks nice actually.

 > It has really simplified the front-end code generator, at no cost in C-- 
 > language complexity.

It would be even better to have printf-like support for inserting
characters. 


Norman
_______________________________________________
Cminusminus mailing list
[email protected]
https://cminusminus.org/mailman/listinfo/cminusminus

Reply via email to