Re: [Axiom-developer] vmlisp

[Gabriel Dos Reis]

On Mon, 14 May 2007, Waldek Hebisch wrote:

| Another candidate is the following:
| 
| --- pp2/wh-20070421/src/interp/postpar.boot.pamphlet    Sun Apr 22 00:24:35 
2007+++ wh-20070421/src/interp/postpar.boot.pamphlet        Sat May  5 21:30:02 
2007@@ -259,15 +259,7 @@
| 
|  postForm (u is [op,:argl]) ==
|    x:=
| -    atom op =>
| -      argl':= postTranList argl
| -      op':=
| -        true=> op
| -        $BOOT => op
| -        GETL(op,'Led) or GETL(op,'Nud) or op = 'IN => op
| -        numOfArgs:= (argl' is [['Tuple,:l]] => #l; 1)
| -        INTERNL("*",STRINGIMAGE numOfArgs,PNAME op)
| -      [op',:argl']
| +    atom op => [op,:postTranList argl]
|      op is ['Scripts,:.] => [:postTran op,:postTranList argl]
|      u:= postTranList u
|      if u is [['Tuple,:.],:.] then
| 
| 
| The true branch means that other parts are never executed.

Yes.  Clearly, it has always looked to me as a temporary hack.
(Otherwise, it would have been entirely removed or commented out).
Before, removing the dead code, we should understand what was originally
intented and why it was not working and prompted the hack.

-- Gaby