Re: [Sbcl-bugs] Apparent bug in SB-IMPL::BACKQ-COMMA interpretation

[Stig H. <sti...@gm...>]

On Sat, Aug 25, 2012 at 12:56 PM, Denis Potapov <po...@ex...> wrote:
> * `(let ((a 1)) `(let ,a)) ;; Eval a form with backquote

The first backquote may be a normal quote without changing anything.
This is not a nested backquote issue.


> (LET ((A 1))
>   `(LET (SB-IMPL::BACKQ-COMMA A)
>      ))

I'd say the problem is here.  For some reason, PRINT messes up the
printing of backquote expressions.  It should either have printed
(LET ((A 1))
  `(let ,a))

or some expression with SB-IMPL:: that actually works.

What it currently does is just wrong.  And it stays wrong even if you
set *PRINT-READABLY*.

>
> * (eval *) ;; Eval the result --- OK
>
> (LET 1
>   )
* returns the internal expression without PRINT mindfuckery, so this works.



> * (LET ((A 1)) `(LET (SB-IMPL::BACKQ-COMMA A))) ;; if typed directly --- not OK
>...

Whether you retype or cut-n-paste the mess PRINT produced, the result
is still a mess.

I agree that there is a bug, but thinks the bug is with PRINT, not
with backquote.

Stig Hemmer