[pmwiki-users] Need help writing a simple markup for IP addresses inside of pages

[Patrick R. Michaud]

On Sun, Oct 15, 2006 at 09:53:45PM -0500, Chris Cox wrote:
> I have:
> 
> Markup('IPAddy', 'inline',
> '/([0-9][0-9]*)\.([0-9][0-9]*)\.([0-9][0-9]*)\.([0-9][0-9]*)/',
> '%newwin%[[Network/Ip$1-$2-$3-$4|$1.$2.$3.$4]]%%');
> ...
> Markup('IPRevAddy', 'inline',
> '/([0-9][0-9]*)\.([0-9][0-9]*)\.([0-9][0-9]*)\.([0-9][0-9]*)\.in-addr\.arpa/',
> '[[Network/Ip$4-$3-$2-$1|$1.$2.$3.$4.in-addr.arpa]]');
> 
> However, the two rules DO step on each other.  I know there is some
> fancy way of doing conditional regexp... but I don't even know that
> kind of regexp is used by PmWiki in this case.

The problem is that the replacement text "[[Network/Ip$1-$2-$3-$4|$1.$2.$3.$4]]"
matches the original pattern in the part after the '|'.  So, if the replacement
rule gets called multiple times (as can happen if there are any PRR-based
rules in the sequence), then it will be processed multiple times.  The
same goes for [[...|$1.$2.$3.$4.in-addr.arpa]].

The trick is to make sure that the replacement can't re-trigger the original
rule.  One way to do it:

    Markup('IPAddy', 'inline',
      '/\\b(?<!\\|)(\\d+)\\.(\\d+)\\.(\\d+)\\.(\\d+)\\b/',
      '%newwin%[[Network/Ip$1-$2-$3-$4|$0]]');

The (?<!\\|) is a "negative lookbehind" that says "don't match this pattern
if the preceding character is a vertical brace".

Similarly, do:

    Markup('IPRevAddy', '<IPAddy',
      '/\\b(?<!\\|)(\\d+)\\.(\\d+)\\.(\\d+)\\.(\\d+)\\.in-addr\\.arpa\\b/',
      '[[Network/Ip$4-$3-$2-$1|$0]]');

Note that this second rule is set to be processed _before_ the 'IPAddy'
rule, so that in-addr.arpa addresses are processed before the standard
IP addresses.

Pm