Hello group,
this message is a word of warning and plea to wiki writers.
Reading the wiki and documentation in general, there seems to be an accepted
consensus that most in SOLR is working in utf-8. To my opinion this is
absolutely good.
But this may be a remain of the times. Several efforts seem to have been done
to revert to more tolerant defaults.
Believing the wiki, I suspected all sorts of broken things in my indexer
pipeline.
At the end, thanks to my dear Luke desktop application, I realized that it was
just the defaults that were wrong: both the the Velocity response-writer was
using iso-8859-1.
With velocity, you just need
<str name="v.contentType">text/html; charset=utf-8</str>
(the default seems to be text/html and not output any form of charset)
Others may be different.
I don't see a wiki page specialized to that.
paul
