Hi Önder, do you consult any other sources of data, or is this all in the same database? If it's all in the same database I would just use SQL triggers or views (possibly materialised). Otherwise I think aggregation is probably something you need to add to your list,
zoran On Fri, Jul 7, 2017 at 11:01 AM, onders <[email protected]> wrote: > Hello, > > I want to ask you question about a design choice. (maybe there is an > existing EIP / composite EIPS together out of box, which i am not already > familiar ) > > Think of you have X number of rows in a table. (which will continuosly get > new rows) > And you want to consume every row in that table and decide which message > really needs to processed based on a certain column. > (This requires filtering) and save in a another table. > Additionally you want to make sure in the second table (where you keep rows > to be further processed) where you always have the (let's say the latest > row which is based on its timestamp column) latest row. > > I would guess i need to implement something like idempotent consumer EIP in > which i need to insert & update (which is like merge in PLSQL - assuming we > have Oracle backend) for the records in the second table(which is my > repository and final table holding all the latest records). > > I can see > - Idempotent Consumer EIP > - Selective Consumer EIP > > which are not i can use directly out of box. > > Is there an EIP or composite EIPs to achieve something similar (like > Comparing Consumer Against Repository which i named it bogusly)? > > And could you think of a way not doing too many queries or updates to second > table considering performance? > > Thanks in advance. > > > > -- > View this message in context: > http://camel.465427.n5.nabble.com/SQL-component-comparing-consumer-against-database-tp5806564.html > Sent from the Camel - Users mailing list archive at Nabble.com. -- Zoran Regvart
