Lisp reader bug

Lisp reader bug

[Kenichi Handa]

Emacs of CVS HEAD has a bug in Lisp reader.  For instance,

(read (format "'%c" (decode-char 'ucs #x102)))
=> (quote )

I'm not sure, but perhaps this change is the cause.

2003-02-18  Kim F. Storm  <storm@cua.dk>

	* lread.c (read1): Fix last change.
	"`" is not always special.  Allow "?" after a character constant.

This change adds codes something like this (total 4 places):

	  || index ("\"';([#?", next_next_char)

but, it seems that `index' doesn't work well if the second
arg is greater than 0xFF.  Actually, if I change such lines
to:

	  || (next_next_char < 0x100 && index ("\"';([#?", next_next_char))

the bug disappears.

Kim, could you take a look at this problem?

---
Ken'ichi HANDA
handa@m17n.org

Re: Lisp reader bug

[Andreas Schwab]

storm@cua.dk (Kim F. Storm) writes:

|> Kenichi Handa <handa@m17n.org> writes:
|> 
|> > Emacs of CVS HEAD has a bug in Lisp reader.  For instance,
|> > 
|> > (read (format "'%c" (decode-char 'ucs #x102)))
|> > => (quote )
|> > 
|> > I'm not sure, but perhaps this change is the cause.
|> > 
|> > 2003-02-18  Kim F. Storm  <storm@cua.dk>
|> > 
|> > 	* lread.c (read1): Fix last change.
|> > 	"`" is not always special.  Allow "?" after a character constant.
|> > 
|> > This change adds codes something like this (total 4 places):
|> > 
|> > 	  || index ("\"';([#?", next_next_char)
|> > 
|> > but, it seems that `index' doesn't work well if the second
|> > arg is greater than 0xFF.  Actually, if I change such lines
|> > to:
|> > 
|> > 	  || (next_next_char < 0x100 && index ("\"';([#?", next_next_char))
|> > 
|> > the bug disappears.
|> > 
|> > Kim, could you take a look at this problem?
|> 
|> I'm not aware of that limitation in `index' (it is not documented),

The POSIX standard says that index shall be equivalen to strchr, and the C
standard says that strchr converts its second argument to char before the
comparison.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux AG, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Lisp reader bug

[Kim F. Storm]

Kenichi Handa <handa@m17n.org> writes:

> Emacs of CVS HEAD has a bug in Lisp reader.  For instance,
> 
> (read (format "'%c" (decode-char 'ucs #x102)))
> => (quote )
> 
> I'm not sure, but perhaps this change is the cause.
> 
> 2003-02-18  Kim F. Storm  <storm@cua.dk>
> 
> 	* lread.c (read1): Fix last change.
> 	"`" is not always special.  Allow "?" after a character constant.
> 
> This change adds codes something like this (total 4 places):
> 
> 	  || index ("\"';([#?", next_next_char)
> 
> but, it seems that `index' doesn't work well if the second
> arg is greater than 0xFF.  Actually, if I change such lines
> to:
> 
> 	  || (next_next_char < 0x100 && index ("\"';([#?", next_next_char))
> 
> the bug disappears.
> 
> Kim, could you take a look at this problem?

I'm not aware of that limitation in `index' (it is not documented),
but if that's the case, your fix is appropriate...  Please install it
if not already done.

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Lisp reader bug

[Richard Stallman]

    Emacs of CVS HEAD has a bug in Lisp reader.  For instance,

    (read (format "'%c" (decode-char 'ucs #x102)))
    => (quote )

Is that a bug?  I am not sure.  What do you think correct behavior
would be?

Re: Lisp reader bug

[Kenichi Handa]

In article <5xwuiab37d.fsf@kfs2.cua.dk>, storm@cua.dk (Kim F. Storm) writes:
> I'm not aware of that limitation in `index' (it is not documented),
> but if that's the case, your fix is appropriate...  Please install it
> if not already done.

Just done.

Andreas Schwab <schwab@suse.de> writes:
> The POSIX standard says that index shall be equivalen to strchr, and the C
> standard says that strchr converts its second argument to char before the
> comparison.

Thank you for confirming it.

---
Ken'ichi HANDA
handa@m17n.org

Re: Lisp reader bug

[Kenichi Handa]

In article <E191imp-0005UD-00@fencepost.gnu.org>, Richard Stallman <rms@gnu.org> writes:
>     Emacs of CVS HEAD has a bug in Lisp reader.  For instance,
>     (read (format "'%c" (decode-char 'ucs #x102)))
>     => (quote )

> Is that a bug?  I am not sure.  What do you think correct behavior
> would be?

The correct behaviour is that the 2nd element of the
returned list a symbol whose name is the same as:
  (string (decode-char 'ucs #x102))
That is also the behaviour of the previous Emacs.

---
Ken'ichi HANDA
handa@m17n.org