Conrad, Is it possible to add a view (materialized or not) to the RDBMS? That view could take care of the denormalization and then QueryDatabaseTable could point at the view. The DB would take care of the push-down filters, which functionally is like if you had a QueryDatabaseTable for each table then did the joins.
In NiFi 1.0 there is a GenerateTableFetch processor which is like QueryDatabaseTable except it generates SQL instead of executing SQL. That might be used in your a) option above but you'd have to reconcile the SQL statements into a JOIN. A possible improvement to either/both processors would be to add attributes for the maximum value columns whose values are the maximum observed values. Then you wouldn't have to parse or manipulate the SQL if you really just want the max values. I have been thinking about how QueryDatabaseTable and GenerateTableFetch would work if they accepted incoming flow files (to allow dynamic table names for example). It's a bit more tricky because those processors run without input to get max values, so their behavior would change when a flow file is present but would return to the original behavior if no flow file is present. Since the ListDatabaseTables processor is also in 1.0, it would be nice to use that as input to the other two processors. I'm definitely interested in any thoughts or discussion around these things :) Regards, Matt On Wed, Aug 3, 2016 at 8:37 AM, Conrad Crampton <conrad.cramp...@secdata.com> wrote: > Hi, > > My use case is that I want to ship a load of rows from an RDMS periodically > and put in HDFS as Avro. > > QueryTable processor has functionality that would be great i.e. maxcolumn > value (there are couple of columns I could use for this from the data) and > it is this functionality I am looking for, BUT the data is not from one > single table. The nature of the RDBMS is that the business view on the data > requires a bunch of joins from other tables/schemas to get the correct Avro > file so the options I appear to have are > > a) Use QueryTable for each table that make up the business view and do > the joins etc. in HDFS (Spark or something) – or potentially do the > reconciliation within NiFi??? > > b) Use ExecuteSQL to run the complete SQL to get the rows which can > easily be put into HDFS as Avro given that the line will be the business > (denormalised) data that is required. > > The problem with a) is the reconciliation (denormalisation) of the data and > the problem with b) is how to maintain the maxcolumn value so I only get the > data since the last run. > > > > In order to address b) can I use the DistrubutedMapCacheServer & Client to > hold a key/value pair of last run date and extract from this date as a > parameter? > > > > Thanks for any suggestions. > > > > Conrad > > > > SecureData, combating cyber threats > > ________________________________ > > The information contained in this message or any of its attachments may be > privileged and confidential and intended for the exclusive use of the > intended recipient. If you are not the intended recipient any disclosure, > reproduction, distribution or other dissemination or use of this > communications is strictly prohibited. The views expressed in this email are > those of the individual and not necessarily of SecureData Europe Ltd. Any > prices quoted are only valid if followed up by a formal written quote. > > SecureData Europe Limited. Registered in England & Wales 04365896. > Registered Address: SecureData House, Hermitage Court, Hermitage Lane, > Maidstone, Kent, ME16 9NT
