Hi!
Is the following a bug in CMUCL or am I missing something? It
works on Allegro Trial...
* (code-char 28)
#\^\
* (prin1 (code-char 28))
#\^\
#\^\
* (read-from-string (prin1-to-string (code-char 28)))
Unexpected EOF on #<String-Input Stream> after escape character.
Restarts:
0: [abort] Return to Top-Level.
Debug (type H for help)
(common-lisp::reader-eof-error #<String-Input Stream> "after escape character")
0]
Regards,
--
Nils Goesche
Ask not for whom the <CONTROL-G> tolls.
From: ······@corporate-world.lisp.de
Subject: Re: character printing
Date:
Message-ID: <92pvsa$t2v$1@nnrp1.deja.com>
In article <··············@darkstar.cartan>,
Nils Goesche <······@t-online.de> wrote:
> Hi!
>
> Is the following a bug in CMUCL or am I missing something? It
> works on Allegro Trial...
>
> * (code-char 28)
> #\^\
#\^\ does not make much sense as a character name... Why
is there a second "\" ?
> #\^\
> * (read-from-string (prin1-to-string (code-char 28)))
Does this work:
(read-from-string "#\^\\")
?
Anyway, this looks like a question to the CMUCL maintainers.
Sent via Deja.com
http://www.deja.com/
* Nils Goesche <······@t-online.de>
| Is the following a bug in CMUCL or am I missing something?
It looks like a bug in the printer. (char-name (code-char 28)) should
return a string that has one final backslash and prints with two, and so
the same should be true for the printed representation of the character.
* ······@corporate-world.lisp.de
| #\^\ does not make much sense as a character name...
Well, the _name_ should be the string "^\\".
| Why is there a second "\"?
Because ^ is a common modifier character that turns the following
character into a control character. ^A is ctrl-A, and ^\ is perfectly
acceptable as a control character in a context where \ is not escaping
the following character -- it is, however, in Common Lisp.
#:Erik
--
Performance is the last refuge of the miserable programmer.
······@corporate-world.lisp.de writes:
> In article <··············@darkstar.cartan>,
> Nils Goesche <······@t-online.de> wrote:
> > Hi!
> >
> > Is the following a bug in CMUCL or am I missing something? It
> > works on Allegro Trial...
> >
> > * (code-char 28)
> > #\^\
>
> #\^\ does not make much sense as a character name... Why
> is there a second "\" ?
It's stands for `Ctrl-\'. When you hit Ctrl-\ on a Unix terminal
with default settings, you send a SIGQUIT to the foreground
process, which usually causes it to terminate and dump core, for
debugging purposes. (You appear to be using a Mac)
> > #\^\
> > * (read-from-string (prin1-to-string (code-char 28)))
>
> Does this work:
>
> (read-from-string "#\^\\")
>
> ?
No, but this does:
(read-from-string "#\\^\\\\")
> Anyway, this looks like a question to the CMUCL maintainers.
Yes, as Erik Naggum said, it is a bug in the printer, as
expected. Some kind soul from cmucl.cons.org already sent me a
patch. I was trying to print out and read back in some character
tables, and was surprised that it didn't work. I thought that
maybe I was missing some special variable like *print-readably*,
so I'd rather ask here before reporting a bug.
Regards,
--
Nils Goesche
Ask not for whom the <CONTROL-G> tolls.