From: Nuralanur@...
Date: 2006-02-01T03:56:09+09:00
Subject: Re: Premature end of regular expression with non-ascii chara

-------------------------------1138733736
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

Dear Nick,
=20
I'm glad things finally worked out for you.
Actually, I things like '\352' do not to split any further (at least on my =20
system (cygwin 6.7.0.0-6) on Windows XP),
this seems to be some  encoding for "=EA" etc. that is  Windows-specific.
In particular, it is not equivalent to a string composed of '\',  '3','5','2=
'
(you would have to double the '\' in the regexp if you were searching
for a string '\352' in a text).
=20
But the line=20
=20
splitted_text=3Dtext.split(/(?=3D.)/)
=20
should produce an Array with the individual letters in the string.
That involves a concept about regexps called zero-width positive
lookahead (see _http://www.regular-expressions.info/lookaround.html_=20
(http://www.regular-expressions.info/lookaround.html) ).
=20
Please let us all know if you still encounter  problems.
=20
Best regards,
=20
Axel=20

-------------------------------1138733736--