On Friday 26 September 2008 03:00:15 pm Rob Dixon wrote:
> Jack Gates wrote:
> > On Friday 26 September 2008 01:23:23 pm John W. Krahn wrote:
> >> Jack Gates wrote:
> >>> s!(<|</)([^\!][A-Z0-9 ]+>)!$1\L$2\E!g;
> >>> or
> >>> s/(<|<\/)([^!][A-Z0-9 ]+>)/$1\L$2\E/g;
> >>>
> >>> The RE above captures and replaces all HTML tags with
> >>> lowercase as desired except for any tag that has only one
> >>> letter such as <P>, <B> or <I>
> >>>
> >>> It will get the </B>, </P> and </I>
> >>>
> >>> It properly ignores the <!DOCTYPE> tag
> >>>
> >>> What is the correct way to write the above RE?
> >>
> >> Perhaps this is what you want?
> >>
> >> s{ ( < (?!!) /? [[:upper:]]{2,} > | < [[:upper:]]{2,} \s* /> )
> >> } {\L$1}xg;
> >
> > Yours worked with three exceptions it was missing all the
> > single letter tags open and close and the H 1-6 tags and the
> > tags that had and element or attribute
> >
> > A little tweaking and it works. It gets only and all of what I
> > want it to get.
> >
> > s{ ( < (?!!) /? [[:upper:][:digit:] ]{1,} > | < [[:upper:]
> > [:digit:] ]{1,} \s* /> ) } {\L$1}xg;
>
> That won't modify tags that have attributes either. Are you sure
> you know what you want?
>
> RYes, why don't you try reading what I wrote. -- Silverfox Please don't CC me. I am subscribed to the list. -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/
