From: Leslie P. P. <sk...@vi...> - 2009-07-31 08:27:39
|
The behavior of unconditionally quitting on EOF can be annoying if you're working directly on your production machine's REPL. More than once I have inadvertently closed my image by hitting C-d. I could define my own REPL but that seems overkill for just a small change. Would anyone object against a patch introducing a common switch *TOPLEVEL-QUIT-ON-EOF* (on by default) and changes to the default REPL and SB-ACLREPL? Leslie -- http://www.linkedin.com/in/polzer |
From: Nikodemus S. <nik...@ra...> - 2009-07-31 09:55:48
|
2009/7/31 Leslie P. Polzer <sk...@vi...>: > Would anyone object against a patch introducing a common > switch *TOPLEVEL-QUIT-ON-EOF* (on by default) and changes > to the default REPL and SB-ACLREPL? I at least would be fine with something like that -- though possibly with a slight preference to *TOPLEVEL-EOF-HOOK* or similar. Cheers, -- Nikodemus |
From: Leslie P. P. <sk...@vi...> - 2009-08-05 07:37:04
|
It turns out that SB-ACLREPL already has a handy switch for control of EOF behavior, so I'll just use that and let the matter rest. Thanks for the feedback everyone. Leslie -- http://www.linkedin.com/in/polzer |
From: Gabriel D. R. <gd...@in...> - 2009-07-31 14:38:25
|
On Fri, Jul 31, 2009 at 3:27 AM, Leslie P. Polzer<sk...@vi...> wrote: > The behavior of unconditionally quitting on EOF can be > annoying if you're working directly on your production > machine's REPL. > > More than once I have inadvertently closed my image by > hitting C-d. > > I could define my own REPL but that seems overkill for > just a small change. > > Would anyone object against a patch introducing a common > switch *TOPLEVEL-QUIT-ON-EOF* (on by default) and changes > to the default REPL and SB-ACLREPL? All free Lisp systems I have access to support the following idiom at the shell level echo '(lisp-implmentation-version)' | lisp Whatever is done, that idiom should NOT be broken -- some configure/test programs use that for Lisp-based software out there. -- Gaby |
From: Leslie P. P. <sk...@vi...> - 2009-08-01 10:02:49
|
Gabriel Dos Reis wrote: > On Fri, Jul 31, 2009 at 3:27 AM, Leslie P. > Polzer<sk...@vi...> wrote: >> Would anyone object against a patch introducing a common >> switch *TOPLEVEL-QUIT-ON-EOF* (on by default) and changes >> to the default REPL and SB-ACLREPL? > > All free Lisp systems I have access to support the following idiom > at the shell level > > echo '(lisp-implmentation-version)' | lisp > > Whatever is done, that idiom should NOT be broken -- some > configure/test programs use that for Lisp-based software out there. Definitely. The idea is just to provide easy customization of this behavior in custom projects. Leslie -- http://www.linkedin.com/in/polzer |
From: Christophe R. <cs...@ca...> - 2009-08-02 19:56:12
|
"Leslie P. Polzer" <sk...@vi...> writes: > I could define my own REPL but that seems overkill for > just a small change. You could define your own function to bind to *REPL-READ-FORM-FUN* which does not quit on EOF, couldn't you? Cheers, Christophe |
From: Leslie P. P. <sk...@vi...> - 2009-08-02 19:29:00
|
Christophe Rhodes wrote: > "Leslie P. Polzer" <sk...@vi...> writes: > >> I could define my own REPL but that seems overkill for >> just a small change. > > You could define your own function to bind to *REPL-READ-FORM-FUN* > which does not quit on EOF, couldn't you? Yes. The nice thing about an explicit hook would be that all REPLs (e.g. the default and SB-ACLREPL) could be made to support it in a uniform way. Otherwise custom EOF handling would have to be added to a duplication of the original REPL code. Personally I'd still be most comfortable with the hook proposed by Nikodemus, but I'm not going to argue for this more than it is worth. Cheers, Leslie -- http://www.linkedin.com/in/polzer |