I would add to its use cases processing documents after they are indexed. For instance, in Solr, the standard update chain is to index, then log. I inserted "notify" (my update handler) between index and log.
On Sat, Oct 12, 2013 at 8:44 PM, Jack Park <[email protected]> wrote: > Thanks. Will watch for progress on that. > > On Sat, Oct 12, 2013 at 6:04 PM, Aaron McCurry <[email protected]> wrote: >> Not yet. But I have added a task to build that capability, I've been >> meaning to add it for some time now. >> >> https://issues.apache.org/jira/browse/BLUR-276 >> >> Aaron >> >> >> On Sat, Oct 12, 2013 at 5:48 PM, Jack Park <[email protected]> wrote: >> >>> My question is predicated on an observation from coding with Solr: I >>> am able to declare update handlers in which I have the opportunity to >>> intercept a new document at some point in the cycle and send it out to >>> other agents. Case in point: I am creating a topic map in Solr which >>> needs to have each new document examined by a "merge engine" which >>> makes decisions and sometimes performs surgery on the index to reflect >>> merged documents (topics). >>> >>> Do I have access in Blur to such capabilities? >>> >>> Many thanks in advance. >>> Jack >>>
