Server vs Client - JDK compatibility
Currently our application is running on BroadVision server which is actually based on very older version of JDK that is v1.3. Now we've a requirement for searching documents and we want to integrate Apache Solr within our application. We have also decided to run Apache Solr service separately in a server for various other reasons. Now my question is, is it possible/compatible to have the Apache Solr server running the latest JDK version (running as a separate server as said before) and the client library (SolrJ) running behind a lower version (JDK 1.3), since the client API will be invoked/used from within BroadVision server? If this is possible/compatible, what are all the features that cannot be used/leveraged because of running a lower version of client library and more particularly what type of file extensions will not be supported, indexed and searchable? -- View this message in context: http://lucene.472066.n3.nabble.com/Server-vs-Client-JDK-compatibility-tp4109982.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: Server vs Client - JDK compatibility
Shawn Heisey-4 wrote > Although SolrJ is a standalone component, it is also an integral part of > Solr itself, so it has the same Java requirement as Solr. It is highly > unlikely that it will work at all with a 1.3.x Java version. Thanks Shawn. So which version of Solr supports JDK 1.3? Or any pointers/way to figure out this from the archive/previous release folder? -- View this message in context: http://lucene.472066.n3.nabble.com/Server-vs-Client-JDK-compatibility-tp4109982p4110132.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: Server vs Client - JDK compatibility
Chris Hostetter-3 wrote > Your best bet is probably to ignore SolrJ -- just pick an HTTP library and > an XML serialiation library that you are familiar with and know will run > in your client app and use them to talk to Solr with some custom code to > format your docs in XML for indexing and parsing the response XML when > searching. Can you please give the list of available HTTP client library? Which one among the list is highly recommended and easy to integrate & use from within Java platform? Working example for the same are highly appreciated. -- View this message in context: http://lucene.472066.n3.nabble.com/Server-vs-Client-JDK-compatibility-tp4109982p4110136.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: Server vs Client - JDK compatibility
Thanks Alex for the response. Alexandre Rafalovitch wrote > What features are you planning to use from SolrJ client? Because you can > always just talk to Solr server directly with HTTP query URLs and JSON > results. Because SolrJ is a Java client, we've chosen that as the default client that without much thought. We just need to query/run search in the Solr server and get the result response in the form of Java Bean, as all the parsing is done by SolrJ. Nothing else. Alexandre Rafalovitch wrote > Do you need to support multiple Solr servers? No, we don't need multi Solr server support. Alexandre Rafalovitch wrote > Do you want bean mapping? Something else? Of course, yes we need bean mapping which will be more convenient and easy to use. We don't want to roll out our own response parsing. -- View this message in context: http://lucene.472066.n3.nabble.com/Server-vs-Client-JDK-compatibility-tp4109982p4110146.html Sent from the Solr - User mailing list archive at Nabble.com.
