«

this is starting to bother me:

if (
$req =~ /\.html$/ ||
$req =~ /\.pl$/ ||
$req =~ /\.shtml$/ ||
$req =~ /\.txt$/ ||
$req =~ /\.cgi$/ ||
$req =~ /\.cgi\?/ ||
$req =~ /\/\?/ ||
$req =~ /\/$/
) {
$countPage++;
}


there's got to be a better way.

"Ken Sington" <ken_sington@nospam_abcdefg.com> wrote in message
news:vJmdnbh-yuSIGIzcRVn-...
> this is starting to bother me:
>
> if (
> $req =~ /\.html$/ ||
> $req =~ /\.pl$/ ||
> $req =~ /\.shtml$/ ||
> $req =~ /\.txt$/ ||
> $req =~ /\.cgi$/ ||
> $req =~ /\.cgi\?/ ||
> $req =~ /\/\?/ ||
> $req =~ /\/$/
> ) {
> $countPage++;
> }
>
>
> there's got to be a better way.

I see three different clusters that just can't be merged. The following will
decrease the readability immensely, but will compact the code if that's all
you want:

if ( $req =~ m#(\.(s?html?|pl|cgi|txt)$)|(/[$?])|(\.(pl|cgi)\?)# ) {
$countPage++; }

Matt

"Matt Garrish" <> wrote in message
news:tvgQc.28984$ ...
>
> "Ken Sington" <ken_sington@nospam_abcdefg.com> wrote in message
> news:vJmdnbh-yuSIGIzcRVn-...
> > this is starting to bother me:
> >
> > if (
> > $req =~ /\.html$/ ||
> > $req =~ /\.pl$/ ||
> > $req =~ /\.shtml$/ ||
> > $req =~ /\.txt$/ ||
> > $req =~ /\.cgi$/ ||
> > $req =~ /\.cgi\?/ ||
> > $req =~ /\/\?/ ||
> > $req =~ /\/$/
> > ) {
> > $countPage++;
> > }
> >
> >
> > there's got to be a better way.
>
> I see three different clusters that just can't be merged. The following
will
> decrease the readability immensely, but will compact the code if that's
all
> you want:
>
> if ( $req =~ m#(\.(s?html?|pl|cgi|txt)$)|(/[$?])|(\.(pl|cgi)\?)# ) {
> $countPage++; }
>

Oops! For some reason I thought the $ in the last expression was escaped (I
also should have escaped it in the character class (/[\$?]), so double my
bad!):

if ( $req =~ m#(\.(s?html?|pl|cgi|txt)$)|(/\?)|(\.(pl|cgi)\?)# ) {
$countPage++; }

Matt

Matt Garrish <> wrote in comp.lang.perl.misc:
>
> "Ken Sington" <ken_sington@nospam_abcdefg.com> wrote in message
> news:vJmdnbh-yuSIGIzcRVn-...
> > this is starting to bother me:
> >
> > if (
> > $req =~ /\.html$/ ||
> > $req =~ /\.pl$/ ||
> > $req =~ /\.shtml$/ ||
> > $req =~ /\.txt$/ ||
> > $req =~ /\.cgi$/ ||
> > $req =~ /\.cgi\?/ ||
> > $req =~ /\/\?/ ||
> > $req =~ /\/$/
> > ) {
> > $countPage++;
> > }
> >
> >
> > there's got to be a better way.
>
> I see three different clusters that just can't be merged. The following will
> decrease the readability immensely, but will compact the code if that's all
> you want:
>
> if ( $req =~ m#(\.(s?html?|pl|cgi|txt)$)|(/[$?])|(\.(pl|cgi)\?)# ) {
> $countPage++; }

How about a compromise (untested):

$countPage ++ if
$req =~ /\.(?:html|pl|shtml|txt|cgi)$/ ||
$req =~ /\.cgi\?/ ||
$req =~ /\/\?/ ||
$req =~ /\/$/;

or even

/\.(?:html|pl|shtml|txt|cgi)$/ || /\.cgi\?/ || /\/\?/ || /\/$/ and
$countPage ++ for $req;

Anno

> if (
> $req =~ /\.html$/ ||
> $req =~ /\.pl$/ ||
> $req =~ /\.shtml$/ ||
> $req =~ /\.txt$/ ||
> $req =~ /\.cgi$/ ||
> $req =~ /\.cgi\?/ ||
> $req =~ /\/\?/ ||
> $req =~ /\/$/
> ) {
> $countPage++;
> }

In addition to other ways you've been shown you can combine all the
simple suffix ones

$req =~ /\.(\w+)$/ && $suffix{$1}

But this is probably not worth it for only five suffixes.