I don't think there is a Jira issue for this yet, but I think we've
made good progress towards this, eg IndexWriter can now delete by
term or query, IndexReader no longer writes a commit lock, etc.
I *think* we don't need to carry over delete by document number
(delete by query should work for most cases)?
Then I think the biggie that remains is setNorm. But if
LUCENE-1231 / flexible indexing make it more generally possible to
update field values, I guess we'd need to figure out how that would
happen in the context of an IndexWriter. I think this needs more
fleshing out. Really, IndexReader, since it holds all these cached
values or can access them all from the Directory, does seem like a
"good" place to manage updates to these values.
But before we get a fully read-only IndexReader, you should still be
able to pare back quite a bit, under oal.index.*, if you're only
doing searching?
Mike
DM Smith wrote:
One of the recent threads talked about separating Index reading
from Index writing. Is there a Jira issue for that? I'd like to
know what all that encompasses.
Specifically, I'd like to know if it would be possible to pair
lucene-core.jar down to just the classes that read the index. I'd
like to explore what it would take to make Lucene work on a JavaME
device to search pre-built, read-only indexes.
-- DM Smith
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
