Github user mattyb149 commented on a diff in the pull request:
https://github.com/apache/nifi/pull/1471#discussion_r101091302
--- Diff:
nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/ExecuteSQL.java
---
@@ -192,31 +192,57 @@ public void process(InputStream in) throws
IOException {
try (final Connection con = dbcpService.getConnection();
final Statement st = con.createStatement()) {
st.setQueryTimeout(queryTimeout); // timeout in seconds
- final AtomicLong nrOfRows = new AtomicLong(0L);
- if (fileToProcess == null) {
- fileToProcess = session.create();
- }
- fileToProcess = session.write(fileToProcess, new
OutputStreamCallback() {
- @Override
- public void process(final OutputStream out) throws
IOException {
- try {
- logger.debug("Executing query {}", new
Object[]{selectQuery});
- final ResultSet resultSet =
st.executeQuery(selectQuery);
-
nrOfRows.set(JdbcCommon.convertToAvroStream(resultSet, out,
convertNamesForAvro));
- } catch (final SQLException e) {
- throw new ProcessException(e);
- }
+
+
+ logger.debug("Executing query {}", new Object[]{selectQuery});
+ boolean results = st.execute(selectQuery);
+ int resultCount = 0;
+ while(results){
+ FlowFile resultSetFF;
+ if(fileToProcess==null)
+ resultSetFF = session.create();
+ else {
+ resultSetFF = session.create(fileToProcess);
+ resultSetFF = session.putAllAttributes(resultSetFF,
fileToProcess.getAttributes());
}
- });
- // set attribute how many rows were selected
- fileToProcess = session.putAttribute(fileToProcess,
RESULT_ROW_COUNT, String.valueOf(nrOfRows.get()));
+ final AtomicLong nrOfRows = new AtomicLong(0L);
+
+ resultSetFF = session.write(resultSetFF, new
OutputStreamCallback() {
+ @Override
+ public void process(final OutputStream out) throws
IOException {
+ try {
+ ResultSet resultSet = st.getResultSet();
+
nrOfRows.set(JdbcCommon.convertToAvroStream(resultSet, out,
convertNamesForAvro));
+ } catch (final SQLException e) {
+ throw new ProcessException(e);
+ }
+ }
+ });
+
+ // set attribute how many rows were selected
+ resultSetFF = session.putAttribute(resultSetFF,
RESULT_ROW_COUNT, String.valueOf(nrOfRows.get()));
+
+ logger.info("{} contains {} Avro records; transferring to
'success'",
+ new Object[]{resultSetFF, nrOfRows.get()});
+ session.getProvenanceReporter().modifyContent(resultSetFF,
"Retrieved " + nrOfRows.get() + " rows",
+ stopWatch.getElapsed(TimeUnit.MILLISECONDS));
+ session.transfer(resultSetFF, REL_SUCCESS);
+ resultCount++;
+
+ // are there anymore result sets?
+ results = st.getMoreResults();
+ }
- logger.info("{} contains {} Avro records; transferring to
'success'",
- new Object[]{fileToProcess, nrOfRows.get()});
- session.getProvenanceReporter().modifyContent(fileToProcess,
"Retrieved " + nrOfRows.get() + " rows",
- stopWatch.getElapsed(TimeUnit.MILLISECONDS));
- session.transfer(fileToProcess, REL_SUCCESS);
+ //If we had at least one result then it's OK to drop the
original file, but if we had no results then
+ // pass the original flow file down the line to trigger
downstream processors
+ if(fileToProcess != null) {
+ if (resultCount > 0) {
+ session.remove(fileToProcess);
+ } else {
+ session.transfer(fileToProcess, REL_SUCCESS);
--- End diff --
What was the original behavior? I would've guessed that no flow files were
output if no ResultSets were returned. I'm not sure I've seen that case myself,
I think usually I would get a ResultSet with no rows or something.
My concern with transferring the original to success is that it can cause
flow files of different formats (SQL or Avro) to go to the same relationship.
If your next processor is ConvertAvroToJSON for example, then if you execute a
statement with no ResultSets, the following processor will barf on the SQL
input. If it is desired to represent the event that no ResultSets were
returned, I think an "original" relationship would be a good way to go. The
designer could always route both original and success to the same downstream
processor if they so choose, or perhaps they want to handle "no results" some
different way, so a different relationship would allow them to do that.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
