You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@accumulo.apache.org by GitBox <gi...@apache.org> on 2018/11/09 01:58:13 UTC
[GitHub] ctubbsii edited a comment on issue #753: Improve new MapReduce API
ctubbsii edited a comment on issue #753: Improve new MapReduce API
URL: https://github.com/apache/accumulo/issues/753#issuecomment-437223256
How about `.configure(). ... .store(job)` instead of serialize?
Also, how would we incorporate multiple tables into this? I'm kinda thinking maybe have a `.table` hanging off of each method after the first `.table`, and implement it so you can configure one table, then the next, then the next, etc. Methods chained after one `.table` call, and before the next `.table` affect only the one that precedes it. So,
```java
AccumuloInputFormat.configure().clientInfo(clientInfo)
.table("table1").scanAuths(Authorizations.EMPTY).addIterator(is1)
.table("table2").ranges(ranges).scanAuths(auths).addIterator(is2).addIterator(is3)
.table("table3")
.store(job);
```
In the above, table1 is fully scanned with no auths and one iterator, while table2 is only scanned over specific ranges and with custom auths and two iterators, and the third table is fully scanned with all the default behavior.
Internally, we just edit the most recent internal data structure for a table, and add it to a list of all the tables when we encounter the next `.table` or a terminating `.store`. For keeping the table configs separate, we internally track a 1-up counter for each table.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services