Xref: utzoo gnu.emacs.help:751 comp.emacs:9845
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!unipalm.co.uk!niall
From: niall@unipalm.co.uk (Niall Mansfield)
Newsgroups: gnu.emacs.help,comp.emacs
Subject: an exciting replace-regexp feature
Message-ID: <9101071057.AA00941@unipalm.co.uk>
Date: 7 Jan 91 10:57:02 GMT
Sender: daemon@tut.cis.ohio-state.edu
Reply-To: niall@unipalm.co.uk
Followup-To: gnu.emacs.help
Organization: Gatewayed from the GNU Project mailing list help-gnu-emacs@prep.ai.mit.edu
Lines: 37

In most editors which support regular expressions, 

	replace regular expression  "^."  ""

strips off the first character of each line.  In Emacs, the effect is
dramatically different: it removes everything from your file except
the newlines!

Any fixes?  


(I find this is a real problem when writing Emacs lisp
programs -- to strip off stuff from the beginning of lines, even when
you what's there, is dangerous. E.g. you might think that

	replace-regexp "^\t\t" ""

will safely remove two leading tabs, but it will damage lines which
begin with four tabs, six tabs, etc.  The messy way out is to use
a real newline character instead of the "^" metacharacter, as in

	replace-regexp "\n\t\t" "\n"

but then you need special-casing to handle the first line of the
file).

[ *** Please reply direct -- I'm not on the list. ***]

Regards,
	Niall

--------------------------------------------------------------------
			Unipalm XTech     
niall@unipalm.co.uk	145 St. Neots Road	tel: +44 954 211 797
	or		Hardwick          	fax: +44 954 211 244
niall@unipalm.uucp	Cambridge CB3 7QJ 
			England