On Wed, Sep 20, 2017 at 9:35 AM, Tom Lane <t...@sss.pgh.pa.us> wrote: > TBH, the more I learn about ICU, the less faith I have in the proposition > that it's going to fix anything at all for us in this area. It seems to > be just about as squishy as glibc in terms of locale identification, > if not worse.
That may be our fault, to a significant degree. Did you read my mail from yesterday, over on the "CREATE COLLATION does not sanitize ICU's BCP 47 language tags" thread? I think that we've been incorrectly specifying the locale name that is passed to ucol_open() this whole time. At least, for ICU versions prior to ICU 54. This will need to be addressed very soon. -- 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