Github user MikeThomsen commented on a diff in the pull request:
https://github.com/apache/nifi/pull/2448#discussion_r171302727
--- Diff:
nifi-nar-bundles/nifi-mongodb-bundle/nifi-mongodb-processors/src/main/java/org/apache/nifi/processors/mongodb/GetMongo.java
---
@@ -129,26 +144,44 @@
.addValidator(StandardValidators.POSITIVE_INTEGER_VALIDATOR)
.build();
static final PropertyDescriptor RESULTS_PER_FLOWFILE = new
PropertyDescriptor.Builder()
- .name("results-per-flowfile")
- .displayName("Results Per FlowFile")
- .description("How many results to put into a flowfile at once.
The whole body will be treated as a JSON array of results.")
- .required(false)
- .expressionLanguageSupported(true)
- .addValidator(StandardValidators.POSITIVE_INTEGER_VALIDATOR)
- .build();
+ .name("results-per-flowfile")
+ .displayName("Results Per FlowFile")
+ .description("How many results to put into a flowfile at once. The
whole body will be treated as a JSON array of results.")
+ .required(false)
+ .expressionLanguageSupported(true)
+ .addValidator(StandardValidators.POSITIVE_INTEGER_VALIDATOR)
+ .build();
+ static final PropertyDescriptor ESTIMATE_PROGRESS = new
PropertyDescriptor.Builder()
+ .name("estimate-progress")
+ .displayName("Estimate Progress")
+ .description("If enabled, a count query will be run first, using
the configured query, and attributes will be added to each flowfile showing how
far they are into the result set.")
+ .required(true)
+ .addValidator(StandardValidators.BOOLEAN_VALIDATOR)
+ .allowableValues(GM_TRUE, GM_FALSE)
+ .defaultValue(GM_FALSE.getValue())
+ .build();
+ static final PropertyDescriptor PROGRESSIVE_COMMITS = new
PropertyDescriptor.Builder()
+ .name("progressive-commits")
+ .displayName("Commit After Each Batch")
--- End diff --
It works in coordination with the results per flowfile property. The idea
is to emulate the ExecuteSQL where after a batch of X number has been built up
in the processor and it sends the data to a flowfile, it commits.
I'm tempted to change the batch size property's display name to be
something like Query Fetch Size. I think Results Per Flowfile is probably even
clearer than Batch Size for this. Thoughts?
---
