I will get this reg_exp for .name added/verified/clarified for you. Also, the "thrid and fourth character" dash limitation, I believe, is so that some form of RACE (etc.) encoding can be introduced in the future for multilingual .name names (so people are denied if they do bq-- in any segment of the name). Sounds like .name *may* want to do this, but aren't sure how they're going to yet so they're making sure there's no conflicts when the time comes. (I'm guessing here)
Charles Daminato OpenSRS Product Manager Tucows Inc. - [EMAIL PROTECTED] > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On > Behalf Of Robert L Mathews > Sent: March 22, 2002 2:18 PM > To: [EMAIL PROTECTED] > Subject: Re: 2.52: XML_Client favors .com > > > At 3/21/02 10:31 PM, Gregory Neil Shapiro wrote: > > >A customer complained about the error message he received when trying to > >pre-register a .us domain that wasn't invalid. It told him to 'try > >something similar to "yourname.com"'. Instead of pushing .com, > the message > >should probably use the original TLD. Here is a patch: > > > >Index: XML_Client.pm > >=================================================================== > >RCS file: > /usr/local/src/cvsroot/www/retsiger.com/lib/OpenSRS/XML_Client.pm,v > >retrieving revision 1.1.1.11 > >diff -u -u -r1.1.1.11 XML_Client.pm > >--- XML_Client.pm 20 Mar 2002 07:02:26 -0000 1.1.1.11 > >+++ XML_Client.pm 22 Mar 2002 06:30:27 -0000 > >@@ -897,7 +897,12 @@ > > } elsif ($domain !~ /$OPENSRS_TLDS_REGEX$/) { > > return "Top level domain in \"$domain\" is unavailable"; > > } elsif ($domain !~ > >/^[a-zA-Z0-9][.a-zA-Z0-9\-]*[a-zA-Z0-9]\.name$|^[a-zA-Z0-9][a-zA- > Z0-9\-]{1 > >,}$OPENSRS_TLDS_REGEX$/) { > >- return "Invalid domain format (try something similar to > \"yourname.com\""; > >+ my $tld = "com"; > >+ if ($domain =~ /$OPENSRS_TLDS_REGEX$/) > >+ { > >+ $tld = $1; > >+ } > >+ return "Invalid domain format (try something similar to > \"yourname$tld\""; > > } > > return undef; > > } > > > While I'm looking at this, it seems that the line that checks the syntax > of .name domains in the original OpenSRS code is incorrect: > > /^[a-zA-Z0-9][.a-zA-Z0-9\-]*[a-zA-Z0-9]\.name$ > > If this is trying to validate the syntax as either "first.last.name" or > "last.name", it's not quite right. It allows things like > "first...last.name", "first-.last.name", "first.second.third.last.name", > "r.b.name" and "first-first.last.name", all of which are illegal > according to http://www.opensrs.org/dotname_FAQ.shtml. > > I think it needs something more like: > > /^([a-zA-Z0-9]+\.)?[a-zA-Z0-9][a-zA-Z0-9\-]*[a-zA-Z0-9]\.name$ > > It allows "smith.name", "john.smith.name", "smith-jones.name", > "j.smith.name", "chuck.smith-jones-daminato.name", etc., but not any of > the naughty ones I mentioned above. (It also allows > "john.smith--jones.name" with consecutive hyphens, but the implication on > the page is that this is legal -- it doesn't say it's not, anyway.) > > Speaking of that page, I found this statement confusing: "Hyphens are > allowed, but not in the third and fourth character positions." This seems > to literally state that this is legal: "a-b.smith.name" (hyphen is second > character) and this is illegal: "ab-c.smith.name" (hyphen is third > character), but that doesn't seem to make sense. So I took this to mean > "but not in the third and fourth LEVEL" of the domain name (that is, > hyphens are only allowed in the second level -- the last name part). If I > misunderstood, my suggested regexp above is not correct. > > -- > Robert L Mathews, Tiger Technologies > > "The trouble with doing something right the first time is that nobody > appreciates how difficult it was." >
