That's not quite right: this will return one string containing all text between the beginning of the first address in the string and the last occurence of . followed by 2 or three characters in the string, probably the end of the last email address. For example:

test <a href="mailto:me@somewhere.co.uk"> more <a href="mailto:you@elsewhere.org.uk">

Becomes:

me@somewhere.co.uk"> more <a href="mailto:you@elsewhere.org.uk

Yes, you are right. I didn't take into account this case ... Thanks, it's my mistake.

      
--------------------------------
SV* sv_bless(SV* sv, HV* stash);

That has the same greediness issues that Bilbo referred to with the '@.*' potentially grabbing half the document. Also you are probably being to restrictive about what form a valid e-mail address should take - but I would rather not get into that discussion here as it has been the subject of a multitude of flamewars^Wdiscussions Elsewhere

/J\

You right. I should use metacharacter of end of world '>' for exclusion of grabbing information which not belong to email address.

      
--------------------------------
SV* sv_bless(SV* sv, HV* stash);