On 10/24/13 12:00 PM, Patrik Fältström wrote: > > On 24 okt 2013, at 19:27, Peter Saint-Andre <[email protected]> > wrote: > >> On 10/16/13 1:16 AM, Patrik Fältström wrote: >>> Yeah, I had to do it anyways -- as I am expert reviewer of the >>> IANA tables. New versions of the IANA tables where created by >>> IANA the other day, and I just approved them as they match my >>> own calculations. I.e. we have two completely independent >>> implementations of IDNA2008 (one at IANA, one that I have -- see >>> link below) and I compare the output of the two. If the output >>> matches, then we are pretty sure we are correct. >> >> Patrik, if I may ask, which implementation does IANA use? > > Not mine :-) > > I know Marc Blanchet was involved in developing it, because he asked > me some questions, but honestly I do not know. You must ask IANA. > > I had my code ready (I actually have two versions of the code, ruby > and python -- where the ruby code is completely written from base up > while python use some external libraries for a few things) when IANA > started to implement theirs. I asked them to ensure they would > develop their code completely independent from mine. > > What we do though is that when IANA has calculated a new version of > the tables (for a new version of Unicode) I do the calculations as > well with my code. I do a diff, and if there are no differences, then > I say "go" to IANA and they make the new version available. > > Patrik >
Thanks, that's helpful. We'll need to establish a similar workflow for PRECIS. Peter -- Peter Saint-Andre https://stpeter.im/ _______________________________________________ precis mailing list [email protected] https://www.ietf.org/mailman/listinfo/precis
