From xemacs-m  Sat Sep 13 13:05:32 1997
Received: from punt-1.mail.demon.net (punt-1c.mail.demon.net [194.217.242.136])
	by xemacs.org (8.8.5/8.8.5) with SMTP id NAA24453
	for <xemacs-beta@xemacs.org>; Sat, 13 Sep 1997 13:05:31 -0500 (CDT)
Received: from haruspex.demon.co.uk ([158.152.162.249]) by punt-1.mail.demon.net
           id aa1115652; 13 Sep 97 18:56 BST
Received: (from ltb@localhost)
	by haruspex.demon.co.uk (8.8.7/8.8.7) id RAA04221;
	Sat, 13 Sep 1997 17:55:17 GMT
Mail-Copies-To: never
To: xemacs-beta@xemacs.org
Subject: Re: Odd quotes, font-lock, [^"]
References: <87bu1x46c3.fsf@bittersweet.inetarena.com>
X-url: http://www.haruspex.demon.co.uk/
X-Attribution: Len
X-Unexpected: The Spanish Inquisition
X-Mayan-date: Long count = 12.19.4.9.0; tzolkin = 1 Ahau; haab = 18 Mol
Mime-Version: 1.0 (generated by tm-edit 7.108)
Content-Type: text/plain; charset=US-ASCII
From: Leonard Blanks <ltb@haruspex.demon.co.uk>
Date: 13 Sep 1997 17:55:16 +0000
In-Reply-To: karlheg@inetarena.com's message of "13 Sep 1997 07:49:16 -0700"
Message-ID: <m2iuw514l7.fsf@haruspex.demon.co.uk>
Lines: 31
X-Mailer: Gnus v5.4.65/XEmacs 20.3(beta19) - "Kyiv"

>>>>> "Karl" == Karl M Hegbloom <karlheg@inetarena.com> writes:

 Karl> Is anyone else annoyed when editting perl, when a quote in a regex
 Karl> tricks font-lock, like this:

 Karl> [perl-locking.gif snipped]

 Karl> It would be good if the font-lock was aware of [^"] and \" type
 Karl> things.  Wish I knew how to fix it.

The problem is a general one in distinguishing escaped special
delimiters in a language and, I suspect, due to a tradeoff - either
add a good deal of complexity to the code checking the context for all
such characters and slowing all fontification, or put up with the
relatively rare annoyance of the misjudged meaning of a few
characters.

In editing LaTeX (and TeX) the special meaning of $ as a maths
delimiter is assumed even if the symbol occurs a context where it is
specified literally, as in

  \verb+$HOME/.bash_profile+

The result is the incorrect fontification of everything up to the next 
$ in the text.  This behaviour has been around long enough now that I
am used to it, but like Karl, would not mind a fix but suspect that
quite a slowdown would result.

-- 
Len

