Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!us.feeder.erje.net!feeder.erje.net!eu.feeder.erje.net!newsreader4.netcologne.de!news.netcologne.de!newsfeed.arcor.de!newsspool3.arcor-online.net!news.arcor.de.POSTED!not-for-mail
Content-Type: text/plain; charset="UTF-8"
Message-ID: <16779241.CEys5BfS3R@PointedEars.de>
From: Thomas 'PointedEars' Lahn <PointedEars@web.de>
Reply-To: Thomas 'PointedEars' Lahn <usenet@PointedEars.de>
Organization: PointedEars Software (PES)
Date: Sat, 12 Jan 2013 16:16:14 +0100
User-Agent: KNode/4.4.11
Content-Transfer-Encoding: 8Bit
Subject: Re: Need regexp help
Newsgroups: comp.lang.javascript
References: <eed9c2f1-0e66-4502-849d-9d538eddd1a8@googlegroups.com> <d1e1f8d6shakdd5sjro0vcudj1utmsa4q7@4ax.com> <vcr1f8ptnnqptopbi07esct071uuo38htl@4ax.com>
Followup-To: comp.lang.javascript
MIME-Version: 1.0
Lines: 60
NNTP-Posting-Date: 12 Jan 2013 16:16:15 CET
NNTP-Posting-Host: 357b4895.newsspool3.arcor-online.net
X-Trace: DXC=>nDV\_HCIMnJ00P1S40fZgMcF=Q^Z^V3h4Fo<]lROoRa8kF<OcfhCOk1Na@Z_S6VelDZm8W4\YJNlb@mF9jNikde3>YBYU3JEom<LEcK^1SeSg
X-Complaints-To: usenet-abuse@arcor.de
Xref: csiph.com comp.lang.javascript:18090

Richard Yates wrote:

> On Fri, 11 Jan 2013 17:14:16 -0800, Richard Yates
> <richard@yatesguitar.com> wrote:
>> justaguy <lichunshen84@gmail.com> wrote:
>>> How could use regexp (replace) to find all occurences of the following
                                           ^^^
>>> CSS code in a long string and remove them?   Thanks.
>>>
>>> <xmp>
>>> body { font-family : "Courier New", Courier, monospace; font-size : 9pt;
>>> valign : top; text-align : left; line-height: 9pt }
>>>
>>> td {
>>> font-family : "Courier New", Courier, monospace; font-size : 9pt; valign
>>> : top; text-align : left; line-height: 9pt } </xmp>
>>
>> Not sure why it's a regexp problem. Use indexOf() to find the location
>> of the section you want to remove, and then a couple substring() to
>> close up the gap.
> 
> var ind = haystack.indexOf(needle);
> var newstring =
> haystack.substring(0,ind)+haystack.substring(ind+needle.length);

The key word here is “all”.  Your code would replace only *one* occurrence, 
so to replace *all* occurrences you would need to run it in a loop.  We had 
to do comparably inefficient stuff like that before regular expressions were 
introduced in ECMAScript Edition 3.  That was a little more than 13 years 
ago (1999-12).  (But even then we used String.prototype.replace() already.)

That said, one should not use regular expressions on a context-free language 
unless one really knows what one is doing.  In this exceptional case a 
*single* regular expression would be appropriate to parse CSS because the 
probability of a false positive for pattern matching can be reduced to close 
to zero.  Watch for word-wrap (or instead use RegExp.prototype.concat(), 
provided by jsx.regexp.concat() [1]):

  css = css.replace(/(^|\s*)body\s*\{\s*font-family\s*:\s*"Courier New",
\s*Courier,\s*monospace;\s*font-size\s*:\s*9pt;\s*valign\s*:\s*top;\s*text-
align\s*:\s*left;\s*line-height:\s*9pt\s*\}\s*td\s*\{\s*font-family\s*:
\s*"Courier New",\s*Courier,\s*monospace;\s*font-size\s*:\s*9pt;
\s*valign\s*:\s*top;text-align\s*:\s*left;\s*line-height:\s*9pt\s*\}/g, 
"$1");

This begs the question, however, why the OP thinks this would be necessary 
here in the first place.  I certainly would not recommend it.

BTW, the code to remove is not Valid (there is no “valign” CSS property to 
begin with; the property name is “vertical-align” [2]).  But that could be 
one reason why it is to be removed.

_________
[1] <http://PointedEars.de/scripts/test/regexp>
[2] <http://www.w3.org/TR/CSS21/>
-- 
PointedEars

Twitter: @PointedEars2
Please do not Cc: me. / Bitte keine Kopien per E-Mail.