Geoff hendrey wrote:
Unfortunately the Jira server is down right now so I cannot get the
issue number to provide here. But I wanted to encourage users to vote
for the "full text searching/Lucene integration" for the next release.
Lucene provides fantastic full text searching. Imagine how even more
powerful Derby would be if it supported a more freeform way of
searching for rows. It would be amazing if the integration of Lucene
would treat each row as a Lucene Document. There are some powerful
lucene queries such as FuzzyLikeThisQuery that can search over
multiple document fields. So imagine doing a query for "Smith account
expiring" and having the database find a row in the ACCOUNTS table
with "smith" in the NAME column and "expired" in the STATUS column.
Lucene has support for "stemming" through the SnowballAnalyzer class,
which means Lucene can even figure out that "expired" is a match with
"expiring" because it can understand the common stem word, "expire".
Essentially you'd have a google-like search experience combined with a
relational database.
This would be an absolutely amazing feature. So if it's not to late,
vote for Lucene integration!
-geoff
“How many legs does a dog have if you call the tail a leg? Four.
Calling a tail a leg doesn't make it a leg.”
– A. Lincoln
FYI, this issue is DERBY-472.
Regards,
-Rick
