Would it be possible to decouple the dependency of the Java API on the
Lucence jars? The result being a complete standalone Java API jar?
I could see where this might not be possible but it's worth a try to
ask.
The reason why I need it is that we're currently running an older
version of Lucene and while we migrate I would like to use the Java
API to load the data (assuming the stuck TransportClient gets
resolved) while we continue to run our current, old Lucene, search
index. The alternative is to use the REST API, which were using now
for testing, but according to Shay, the Java API should be faster.
In general, the TransportClient should not require the Lucene jars in order
to work (the Node client does). Currently, I can't really validate it "on
compile time" in the project, but if you hit a snag, then its not
intentional and I will fix it.
Would it be possible to decouple the dependency of the Java API on the
Lucence jars? The result being a complete standalone Java API jar?
I could see where this might not be possible but it's worth a try to
ask.
The reason why I need it is that we're currently running an older
version of Lucene and while we migrate I would like to use the Java
API to load the data (assuming the stuck TransportClient gets
resolved) while we continue to run our current, old Lucene, search
index. The alternative is to use the REST API, which were using now
for testing, but according to Shay, the Java API should be faster.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.