Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!news.glorb.com!news.linkpendium.com!news.linkpendium.com!panix!gordon
From: John Gordon <gordon@panix.com>
Newsgroups: comp.lang.python
Subject: Re: Robust regex
Date: Mon, 19 Nov 2012 20:50:10 +0000 (UTC)
Organization: PANIX Public Access Internet and UNIX, NYC
Lines: 23
Message-ID: <k8e622$p08$1@reader1.panix.com>
References: <mailman.7.1353357285.29569.python-list@python.org>
NNTP-Posting-Host: panix1.panix.com
X-Trace: reader1.panix.com 1353358210 25608 166.84.1.1 (19 Nov 2012 20:50:10 GMT)
X-Complaints-To: abuse@panix.com
NNTP-Posting-Date: Mon, 19 Nov 2012 20:50:10 +0000 (UTC)
User-Agent: nn/6.7.3
Xref: csiph.com comp.lang.python:33555

In <mailman.7.1353357285.29569.python-list@python.org> "Joseph L. Casale" <jcasale@activenetwerx.com> writes:

> Trying to robustly parse a string that will have key/value pairs separated
> by three pipes, where each additional key/value (if more than one exists)
> will be delineated by four more pipes.
>     string = 'key_1|||value_1||||key_2|||value_2'
>     regex = '((?:(?!\|\|\|).)+)(?:\|\|\|)((?:(?!\|\|\|).)+)(?:\|\|\|\|)?'
> I am not convinced this is the most effective or safest, any opinions would
> be greatly appreciated!

Regexes may be overkill here.  A simple string split might be better:

    string = 'key_1|||value_1||||key_2|||value_2'
    pairs = string.split('||||')
    for pair in pairs:
        keyval = pair.split('|||')
        print '%s=%s' % (keyval[0], keyval[1])

-- 
John Gordon                   A is for Amy, who fell down the stairs
gordon@panix.com              B is for Basil, assaulted by bears
                                -- Edward Gorey, "The Gashlycrumb Tinies"