On Wed, Sep 17, 2014 at 5:16 PM, Robert Haas <robertmh...@gmail.com> wrote: > Of course, there's also the question of whether ICU would have similar > issues. You're assuming that they *don't* whack the collation order > around in minor releases, or at least that they do so to some lesser > degree than glibc, but is that actually true?
No, but they're disciplined about it. They clearly do versioning properly, which seems to not be the case with glibc, based on Peter's remarks: http://userguide.icu-project.org/collation/architecture#TOC-Versioning (they talk about a 32-bit identifier here). PostgreSQL's problems in this area are exactly the same as every other database system's (the Unicode consortium anticipated these problems too, and as I pointed out have commented on these problems.). A bunch of prominent database systems are listed as using ICU. -- Peter Geoghegan -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
