Github user rdblue commented on a diff in the pull request:
https://github.com/apache/incubator-nifi/pull/70#discussion_r34176166
--- Diff:
nifi/nifi-nar-bundles/nifi-kite-bundle/nifi-kite-processors/src/main/java/org/apache/nifi/processors/kite/AvroRecordConverter.java
---
@@ -0,0 +1,279 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.nifi.processors.kite;
+
+import java.io.IOException;
+import java.util.List;
+import java.util.Map;
+
+import org.apache.avro.Schema;
+import org.apache.avro.Schema.Field;
+import org.apache.avro.generic.GenericData.Record;
+import org.apache.avro.generic.IndexedRecord;
+import org.codehaus.jackson.JsonNode;
+
+import com.google.common.base.Preconditions;
+import com.google.common.collect.Lists;
+import com.google.common.collect.Maps;
+
+/**
+ * Responsible for converting records of one Avro type to another. Supports
+ * syntax like "record.field" to unpack fields and will try to do simple
type
+ * conversion.
+ */
+public class AvroRecordConverter {
+ private final Schema inputSchema;
+ private final Schema outputSchema;
+ private final Map<String, String> fieldMapping;
+
+ public AvroRecordConverter(Schema inputSchema, Schema outputSchema,
+ JsonNode fieldMapping) {
+ this.inputSchema = inputSchema;
+ this.outputSchema = outputSchema;
+ this.fieldMapping = getFieldMapping(fieldMapping);
+ }
+
+ /**
+ * Converts one record to another given a input and output schema plus
+ * explicit mappings for certain target fields.
+ *
+ * @param record
+ * @param inputSchema
+ * @param outputSchema
+ * @param fieldMapping
+ * @return
+ * @throws AvroConversionException
+ */
+ public Record convert(Record input) throws AvroConversionException {
+ Record result = new Record(outputSchema);
+ for (Field outputField : outputSchema.getFields()) {
+ String inputFieldName = outputField.name();
+ if (fieldMapping.containsKey(outputField.name())) {
+ inputFieldName =
fieldMapping.get(outputField.name());
+ }
+
+ List<String> fieldParts =
Lists.newArrayList(inputFieldName
+ .split("\\."));
+ Record current = input;
+ while (fieldParts.size() > 1) {
+ // Step into the nested records as far as
needed.
+ current = (Record)
current.get(fieldParts.remove(0));
+ }
+
+ // Current should now be in the right place to read the
record.
+ Field f = getFieldForName(inputFieldName, inputSchema);
+ Object content = getContentForName(input,
inputFieldName,
+ input.getSchema());
+ result.put(outputField.name(),
+ convertData(content, f.schema(),
outputField.schema()));
+ }
+ return result;
+ }
+
+ /**
+ * @return the inputSchema
+ */
+ public Schema getInputSchema() {
+ return inputSchema;
+ }
+
+ /**
+ * @return the outputSchema
+ */
+ public Schema getOutputSchema() {
+ return outputSchema;
+ }
+
+ /**
+ * Converts the data from one schema to another. If the types are the
same,
+ * no change will be made, but simple conversions will be attempted for
+ * other types.
+ *
+ * @param content
+ * @param inputSchema
+ * @param outputSchema
+ * @return
+ * @throws AvroConversionException
+ */
+ private Object convertData(Object content, Schema inputSchema,
+ Schema outputSchema) throws AvroConversionException {
+ if (content == null) {
+ // No conversion can happen here.
+ return null;
+ }
+
+ Schema nonNillInput = undoNillableSchema(inputSchema);
+ Schema nonNillOutput = undoNillableSchema(outputSchema);
+ if (nonNillInput.getType().equals(nonNillOutput.getType())) {
+ return content;
+ } else {
+ switch (nonNillOutput.getType()) {
+ case STRING:
+ // This is the easiest conversion case.
Converting to string we
+ // assume
+ // that String.valueOf knows what to do.
+ return String.valueOf(content);
+
+ // For the rest of these, we will try to
convert through string
+ // which
+ // isn't super-efficient but should work.
+ case LONG:
+ try {
+ return
Long.parseLong(String.valueOf(content));
+ } catch (NumberFormatException e) {
+ throw new
AvroConversionException("Cannot convert "
+ + content + " to long");
+ }
+ case INT:
+ try {
+ return
Integer.parseInt(String.valueOf(content));
+ } catch (NumberFormatException e) {
+ throw new
AvroConversionException("Cannot convert "
+ + content + " to int");
+ }
+ case DOUBLE:
+ try {
+ return
Double.parseDouble(String.valueOf(content));
+ } catch (NumberFormatException e) {
+ throw new
AvroConversionException("Cannot convert "
+ + content + " to
double");
+ }
+ case FLOAT:
+ try {
+ return
Float.parseFloat(String.valueOf(content));
+ } catch (NumberFormatException e) {
+ throw new
AvroConversionException("Cannot convert "
+ + content + " to
float");
+ }
+ default:
+ throw new AvroConversionException("Cannot
convert to type "
+ + nonNillOutput.getType());
+ }
+ }
+ }
+
+ /**
+ * Get a mapping from output column to input column definition.
+ *
+ * @param mappingStr
+ * @return
+ * @throws AvroConversionException
+ */
+ private Map<String, String> getFieldMapping(JsonNode listNode) {
+ Map<String, String> result = Maps.newHashMap();
+ for (JsonNode mappingNode : listNode) {
--- End diff --
This seems like a really simple JSON structure. Why not encode this
directly in the processor's properties using dynamic properties? That's how the
route processor works and I think it would work well for this. It would
certainly simplify configuration for users (no writing JSON by hand) and you
wouldn't have to parse JSON either. Each could also have a validator to ensure
the name is a valid Avro path.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
