> > In comp.lang.javascript message <1179589728.556362.297@u30g2000hsc.go
> > oglegroups.com>, Sat, 19 May 2007 08:48:48, hende@gmail.com posted:

> > >     I am trying to get a regular expression test to work and can't
> > >figure out why.  I will give you the code below:

> > >for (var j=0; j<document.getElementById('cmbList').options.length; j+
> > >+) {
> > >   if (document.getElementById('cmbList').options[j].value ==
> > >object.firstChild.data) { strAnswer = "specific"; break; }

> > >   alert('does .' +
> > >document.getElementById('cmbList').options[j].value + '. contain .' +
> > >object.firstChild.data + '.');

> > >   if (/
> > >^object.firstChild.data/.test(document.getElementById('cmbList').option
> > >s[j].value))
> > >{
> > >      alert('in matches');
> > >      strAnswer = "child";
> > >      break;
> > >   }
> > >}

> > >I know the everything is working except the test function due to the
> > >alert statement and the output generated.  The two values being
> > >compared contain directory paths (ie '/some/dir') and I can see (by
> > >the alert statement) while its comparing values that some directories
> > >are child of the parent directory, yet the .test function fails.  Is
> > >there a better way to perform this loop or is there an error with my
> > >test statement?

> > There,   document.getElementById('cmbList').options.length   appears to
> > be evaluated once for each option and once for the non-next one.  Set
> > that into a simple variable before the loop, and use that in the loop
> > control.

> > And   document.getElementById('cmbList')   is repeatedly evaluated in
> > the first "if".  Save that in a variable, and use it there and above.

> > And   document.getElementById('cmbList').options[j].value   is called
> > twice.   Save that ...

> > That which the RegExp literal   /^object.firstChild.data/   contains
> > seems unlikely to be what you want to be there.  It will match the
> > beginning of a string    object.firstChild.data   (with dot wild) and
> > has nothing to do with, for example, the data element of the first child
> > of object.

> > Create simpler code to develop that you are trying to do there.  You may
> > well need something more like

> >         RE = new RegExp(object.firstChild.data)
> >         if RE.test(valueJ) { ... }

> > Use of "object" as an identifier mat well be legitimate, but it can
> > confuse.  Just re-spell it so that it looks more different from
> > "Object".

> > It's a good idea to read the newsgroup c.l.j and its FAQ.  See below.

> > --
> >  (c) John Stockton, Surrey, UK.  ?@merlyn.demon.co.uk   Turnpike v6.05   IE 6
> >  news:comp.lang.javascript FAQ <URL:http://www.jibbering.com/faq/index.html>.
> >  <URL:http://www.merlyn.demon.co.uk/js-index.htm> jscr maths, dates, sources.
> >  <URL:http://www.merlyn.demon.co.uk/> TP/BP/Delphi/jscr/&c, FAQ items, links.

> Thanks for the input as well.  I have another issue.  I can't figure
> regex's out apparently.  The below is a very over simplified example
> of what I am trying to accomplish.  Take for instance this code:

> //re = /\w+\s/g;
> re = /.+\s/;
> str = "fee fi fo fum";
> myArray = str.match(re);
> alert(myArray);

> if I uncomment the first line (re = /\w+\s/g;) and comment out the
> second line (re = /.+\s/;) and run the script I get the following
> results:

> fee ,fi ,fo ,fum

> stored in an array.  So far so good.  I would like to accomplish the
> same goal using the period wildcard (.), so I commentted out the first
> line (re = /\w+\s/g;) and uncommented the second (re = /.+\s/;).  The
> results were an array of 1 containing the entire line.  Why does the
> regular expression evalute to this?  Shouldn't it match *anything* up
> to a space since that is the terminating delimiter (in the example
> from above, it should only return "fee ".  Also note, that I didn't
> give the "g" at the end for a global search.    How could I get my
> logic to work using the . wildcard in this example?

> Thanks,

> Dave