On Fri, 3 Apr 98 12:59:14 -0800, Andrew Edelstein
<andrew(_at_)pure-chaos(_dot_)com> wrote:
era eriksson era(_at_)iki(_dot_)fi 4/3/98 2:42 AM
How about additional protocols such as ftp://, https://, etc?
(news, mailto, gopher?)
I suppose one could search for the characters "://", then include
whatever comes before them, up to the word break. Require that the first
character be alphabetic, in case the URL is quoted...
(Except not all URL:s contain the characters :// -- mailto normally
doesn't, news usually doesn't, you can't tell with other protocols you
don't know about yet so why even try to capture those.)
On Fri, 3 Apr 1998 21:54:45 -0330 (NST), Andrew Vardy
<avardy(_at_)morgan(_dot_)ucs(_dot_)mun(_dot_)ca> wrote:
On Fri, 3 Apr 1998, era eriksson wrote:
Given the observation that URL:s very rarely end in punctuation, while
people sometimes write URL:s with punctuation after them, you might
want to require the last captured character to be alphabetic.
Whenever someone wants to type
...And it is described at http://www.foo.com/, and so blah blah...
Better to put
...And it is described at http://www.foo.com/ , and so blah blah...
(Hence, the convention to always write URL:s inside brokets. Then it's
safe to say something about <http://www.foo.com/>, amidst your prose,
while still allowing standard punctuation to be used. The RFC suggests
you put URL: inside the opening broket if it might not be clear from
the context that what you're quoting is a URL.
Many mail readers etc. will automatically interpret URL:s in brokets
so you can click on them or something.)
Or as a last resort, just use HTML itself in e-mail.
(No, don't.)
/* era */
--
Paparazzi of the Net: No matter what you do to protect your privacy,
they'll hunt you down and spam you. <http://www.iki.fi/~era/spam/>