Path: csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!aioe.org!.POSTED!not-for-mail
From: Rafael Villar <morgano5@hotmail.com>
Newsgroups: comp.lang.java.programmer
Subject: Re: Regex doesn't recognize single quote
Date: Sun, 08 Jan 2012 09:05:11 +1000
Organization: Aioe.org NNTP Server
Lines: 63
Message-ID: <jeaj4f$2dm$1@speranza.aioe.org>
References: <74f4b448-24bf-448f-9f4a-06fd1b79c86d@o12g2000vbd.googlegroups.com> <prmfg79jlt8o86otpnabqfs924cc399var@4ax.com> <bloat-20120107123558@ram.dialup.fu-berlin.de> <tphhg71jbmq4q2vj5dtno5r01igvgdavh2@4ax.com>
NNTP-Posting-Host: 1QofoA7r19pAiEX5BZLHVA.user.speranza.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@aioe.org
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:9.0) Gecko/20111222 Thunderbird/9.0.1
X-Notice: Filtered by postfilter v. 0.8.2
Xref: x330-a1.tempe.blueboxinc.net comp.lang.java.programmer:11095

On 08/01/12 08:41, Roedy Green wrote:
> On 7 Jan 2012 11:42:26 GMT, ram@zedat.fu-berlin.de (Stefan Ram) wrote,
> quoted or indirectly quoted someone who said :
> 
>>> That is not what a regex is for.
>>
>>  How do you know what it is for?
> 
> Regexes are for searching for patterns.  Transforming or deleting
> characters is much simpler done with a for loop. 
> 
> How do I know what a regex is for? I am familiar with the API. I have
> attempted to use them for various purposes and discovered they were
> suitable for some and not for others. 
>>
>>> Just use a StringBuilder the length of your String. Then
>>> loop through the chars with charAt.  If the character is a
>>> ' or \w, ignore it, else append.  If it gets complex, use a
>>> switch or if it gets really complicated use a BitSet.
>>
>>  This might be needless (as far as we know right now)
>>  optimization bloating the code reducing its readability and
>>  low-level thinking, which might be required sometimes, but
>>  does not serve as a general rule. Still it is nice to know
>>  how it could be done if required.
> 
> What is your simpler implementation?  
> 
> /** remove ' and \w from string
>   * @param s string to process
>   * @return string without ' or \w
>   */
> private static String scrunch( final String s ) 
> {
> final Stringbuilder sb = new StringBuilder( s.length() );
> for (int i=0; i<s.length(); i++ )
>   { 
>   char c = s.charAt(i);
>   if ( !( c = '\'' || c = '\w' ) )
>      {
>      sb.append ( c );
>      }
>   }
> return sb.toString();
> }

In most cases is better to use a StringBuilder to perform replacements,
but in this particular case String.replaceAll() is better. By the way,
the escape sequence \w is not a java regular escape sequence but belongs
to the pattern syntax (although you should already know about it, as you
say you are familiar with the API).

Anyway a simpler implementation (and one which works, because yours
doesn't):

/** remove ' and \w from string
 * @param s string to process
 * @return string without ' or \w
 */
private static String scrunch( final String s ) {
   return s.replaceAll("[^'\\w]+", "");
}