Merge remote-tracking branch 'upstream/main' into dagger-parquet-file-processing

# Conflicts:
#	dagger-common/build.gradle
#	dagger-core/src/main/java/io/odpf/dagger/core/StreamManager.java
#	version.txt

Author: Meghajit

.github/workflows/python_release.yml

@@ -0,0 +1,25 @@
+name: Python Package
+on:
+  release:
+    types: [created]
+
+jobs:
+  publishPythonZip:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Zip Python Udf
+        run: |
+          cd dagger-py-functions
+          zip -r python_udfs.zip udfs -x "*/__init__.py"
+          zip -jr data.zip data
+          zip -r dagger-py-functions.zip requirements.txt data.zip python_udfs.zip
+      - name: Upload Release
+        uses: ncipollo/release-action@v1
+        with:
+          artifacts: dagger-py-functions/dagger-py-functions.zip
+          allowUpdates: true
+          omitNameDuringUpdate: true
+          omitBodyDuringUpdate: true
+          omitPrereleaseDuringUpdate: true
+          token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/python_validation.yml

@@ -0,0 +1,27 @@
+name: Python Validation
+
+on: push
+
+jobs:
+  pythonValidation:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - name: Set up Python 3.8
+        uses: actions/setup-python@v3
+        with:
+          python-version: '3.8'
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install apache-flink==1.14.3
+          cd dagger-py-functions
+          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
+      - name: Lint with flake8
+        run: |
+          # stop the build if there are Python syntax errors or undefined names
+          flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+      - name: Test with pytest
+        run: |
+          cd dagger-py-functions
+          pytest --disable-warnings

dagger-common/build.gradle

@@ -61,12 +61,13 @@ dependencies {
 
     dependenciesCommonJar ('org.apache.hadoop:hadoop-client:2.8.3') {
         exclude module:"commons-cli"
+        exclude module:"commons-compress"
     }
     dependenciesCommonJar 'com.google.cloud.bigdataoss:gcs-connector:1.9.0-hadoop2'
     dependenciesCommonJar 'org.apache.flink:flink-metrics-dropwizard:' + flinkVersion
     dependenciesCommonJar 'org.apache.flink:flink-json:' + flinkVersion
     dependenciesCommonJar 'com.jayway.jsonpath:json-path:2.4.0'
-    dependenciesCommonJar 'io.odpf:stencil:0.1.6'
+    dependenciesCommonJar 'io.odpf:stencil:0.2.1'
     dependenciesCommonJar 'com.google.code.gson:gson:2.8.2'
     dependenciesCommonJar 'org.apache.parquet:parquet-column:1.12.2'
 

dagger-core/env/local.properties

@@ -26,3 +26,7 @@ METRIC_TELEMETRY_ENABLE=true
 # == Others ==
 FUNCTION_FACTORY_CLASSES=io.odpf.dagger.functions.udfs.factories.FunctionFactory
 FLINK_ROWTIME_ATTRIBUTE_NAME=rowtime
+
+# == Python Udf ==
+PYTHON_UDF_ENABLE=false
+PYTHON_UDF_CONFIG={"PYTHON_FILES":"/path/to/files.zip", "PYTHON_REQUIREMENTS": "requirements.txt", "PYTHON_FN_EXECUTION_BUNDLE_SIZE": "1000"}

dagger-core/src/main/java/io/odpf/dagger/core/StreamManager.java

@@ -28,13 +28,18 @@
 import io.odpf.dagger.core.sink.SinkOrchestrator;
 import io.odpf.dagger.core.source.StreamsFactory;
 import io.odpf.dagger.core.utils.Constants;
+import io.odpf.dagger.functions.udfs.python.PythonUdfConfig;
+import io.odpf.dagger.functions.udfs.python.PythonUdfManager;
 
+import java.io.IOException;
 import java.lang.reflect.Constructor;
 import java.lang.reflect.InvocationTargetException;
 import java.time.Duration;
 import java.util.List;
 
 import static io.odpf.dagger.core.utils.Constants.*;
+import static io.odpf.dagger.functions.common.Constants.PYTHON_UDF_ENABLE_DEFAULT;
+import static io.odpf.dagger.functions.common.Constants.PYTHON_UDF_ENABLE_KEY;
 import static org.apache.flink.table.api.Expressions.$;
 
 /**
@@ -138,7 +143,13 @@ private ApiExpression[] getApiExpressions(StreamInfo streamInfo) {
      *
      * @return the stream manager
      */
-    public StreamManager registerFunctions() {
+    public StreamManager registerFunctions() throws IOException {
+        if (configuration.getBoolean(PYTHON_UDF_ENABLE_KEY, PYTHON_UDF_ENABLE_DEFAULT)) {
+            PythonUdfConfig pythonUdfConfig = PythonUdfConfig.parse(configuration);
+            PythonUdfManager pythonUdfManager = new PythonUdfManager(tableEnvironment, pythonUdfConfig);
+            pythonUdfManager.registerPythonFunctions();
+        }
+
         String[] functionFactoryClasses = configuration
                 .getString(Constants.FUNCTION_FACTORY_CLASSES_KEY, Constants.FUNCTION_FACTORY_CLASSES_DEFAULT)
                 .split(",");

dagger-core/src/main/java/io/odpf/dagger/core/processors/common/EndpointHandler.java

@@ -1,7 +1,7 @@
 package io.odpf.dagger.core.processors.common;
 
 import io.odpf.dagger.core.processors.ColumnNameManager;
-import io.odpf.dagger.core.processors.types.SourceConfig;
+import io.odpf.dagger.core.utils.Constants.ExternalPostProcessorVariableType;
 import org.apache.flink.streaming.api.functions.async.ResultFuture;
 import org.apache.flink.types.Row;
 
@@ -29,7 +29,6 @@
  */
 public class EndpointHandler {
     private static final Logger LOGGER = LoggerFactory.getLogger(EndpointHandler.class.getName());
-    private SourceConfig sourceConfig;
     private MeterStatsManager meterStatsManager;
     private ErrorReporter errorReporter;
     private String[] inputProtoClasses;
@@ -48,13 +47,11 @@ public class EndpointHandler {
      * @param columnNameManager the column name manager
      * @param descriptorManager the descriptor manager
      */
-    public EndpointHandler(SourceConfig sourceConfig,
-                           MeterStatsManager meterStatsManager,
+    public EndpointHandler(MeterStatsManager meterStatsManager,
                            ErrorReporter errorReporter,
                            String[] inputProtoClasses,
                            ColumnNameManager columnNameManager,
                            DescriptorManager descriptorManager) {
-        this.sourceConfig = sourceConfig;
         this.meterStatsManager = meterStatsManager;
         this.errorReporter = errorReporter;
         this.inputProtoClasses = inputProtoClasses;
@@ -63,19 +60,20 @@ public EndpointHandler(SourceConfig sourceConfig,
     }
 
     /**
-     * Get endpoint or query variables values.
+     * Get external post processor variables values.
      *
      * @param rowManager   the row manager
+     * @param variableType the variable type
+     * @parm variables     the variable list
      * @param resultFuture the result future
      * @return the array object
      */
-    public Object[] getEndpointOrQueryVariablesValues(RowManager rowManager, ResultFuture<Row> resultFuture) {
-        String queryVariables = sourceConfig.getVariables();
-        if (StringUtils.isEmpty(queryVariables)) {
+    public Object[] getVariablesValue(RowManager rowManager, ExternalPostProcessorVariableType variableType, String variables, ResultFuture<Row> resultFuture) {
+        if (StringUtils.isEmpty(variables)) {
             return new Object[0];
         }
 
-        String[] requiredInputColumns = queryVariables.split(",");
+        String[] requiredInputColumns = variables.split(",");
         ArrayList<Object> inputColumnValues = new ArrayList<>();
         if (descriptorMap == null) {
             descriptorMap = createDescriptorMap(requiredInputColumns, inputProtoClasses, resultFuture);
@@ -84,7 +82,7 @@ public Object[] getEndpointOrQueryVariablesValues(RowManager rowManager, ResultF
         for (String inputColumnName : requiredInputColumns) {
             int inputColumnIndex = columnNameManager.getInputIndex(inputColumnName);
             if (inputColumnIndex == -1) {
-                throw new InvalidConfigurationException(String.format("Column '%s' not found as configured in the endpoint/query variable", inputColumnName));
+                throw new InvalidConfigurationException(String.format("Column '%s' not found as configured in the '%s' variable", inputColumnName, variableType));
             }
 
             Descriptors.FieldDescriptor fieldDescriptor = descriptorMap.get(inputColumnName);
@@ -105,11 +103,12 @@ public Object[] getEndpointOrQueryVariablesValues(RowManager rowManager, ResultF
      *
      * @param resultFuture            the result future
      * @param rowManager              the row manager
-     * @param endpointVariablesValues the endpoint variables values
+     * @param variables               the request/header variables
+     * @param variablesValue          the variables value
      * @return the boolean
      */
-    public boolean isQueryInvalid(ResultFuture<Row> resultFuture, RowManager rowManager, Object[] endpointVariablesValues) {
-        if (!StringUtils.isEmpty(sourceConfig.getVariables()) && (Arrays.asList(endpointVariablesValues).isEmpty() || Arrays.stream(endpointVariablesValues).allMatch(""::equals))) {
+    public boolean isQueryInvalid(ResultFuture<Row> resultFuture, RowManager rowManager, String variables, Object[] variablesValue) {
+        if (!StringUtils.isEmpty(variables) && (Arrays.asList(variablesValue).isEmpty() || Arrays.stream(variablesValue).allMatch(""::equals))) {
             LOGGER.warn("Could not populate any request variable. Skipping external calls");
             meterStatsManager.markEvent(ExternalSourceAspects.EMPTY_INPUT);
             resultFuture.complete(singleton(rowManager.getAll()));

dagger-core/src/main/java/io/odpf/dagger/core/processors/external/AsyncConnector.java

@@ -161,7 +161,7 @@ public void open(Configuration configuration) throws Exception {
             meterStatsManager = new MeterStatsManager(getRuntimeContext().getMetricGroup(), true);
         }
         if (endpointHandler == null) {
-            endpointHandler = new EndpointHandler(sourceConfig, meterStatsManager, errorReporter,
+            endpointHandler = new EndpointHandler(meterStatsManager, errorReporter,
                     schemaConfig.getInputProtoClasses(), schemaConfig.getColumnNameManager(), descriptorManager);
         }
 

dagger-core/src/main/java/io/odpf/dagger/core/processors/external/es/EsAsyncConnector.java

@@ -9,8 +9,8 @@
 import io.odpf.dagger.core.processors.external.AsyncConnector;
 import org.apache.flink.streaming.api.functions.async.ResultFuture;
 import org.apache.flink.types.Row;
-
 import io.odpf.dagger.core.utils.Constants;
+import io.odpf.dagger.core.utils.Constants.ExternalPostProcessorVariableType;
 import org.apache.http.HttpHost;
 import org.apache.http.auth.AuthScope;
 import org.apache.http.auth.UsernamePasswordCredentials;
@@ -79,8 +79,8 @@ protected void createClient() {
     protected void process(Row input, ResultFuture<Row> resultFuture) {
         RowManager rowManager = new RowManager(input);
         Object[] endpointVariablesValues = getEndpointHandler()
-                .getEndpointOrQueryVariablesValues(rowManager, resultFuture);
-        if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, endpointVariablesValues)) {
+                .getVariablesValue(rowManager, ExternalPostProcessorVariableType.ENDPOINT_VARIABLE, esSourceConfig.getVariables(), resultFuture);
+        if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, esSourceConfig.getVariables(), endpointVariablesValues)) {
             return;
         }
         String esEndpoint = String.format(esSourceConfig.getPattern(), endpointVariablesValues);

dagger-core/src/main/java/io/odpf/dagger/core/processors/external/grpc/GrpcAsyncConnector.java

@@ -20,6 +20,7 @@
 import com.google.protobuf.Descriptors;
 import com.google.protobuf.DynamicMessage;
 import io.odpf.dagger.core.utils.Constants;
+import io.odpf.dagger.core.utils.Constants.ExternalPostProcessorVariableType;
 import org.apache.commons.lang3.StringUtils;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
@@ -90,8 +91,8 @@ protected void process(Row input, ResultFuture<Row> resultFuture) throws Excepti
             RowManager rowManager = new RowManager(input);
 
             Object[] requestVariablesValues = getEndpointHandler()
-                    .getEndpointOrQueryVariablesValues(rowManager, resultFuture);
-            if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, requestVariablesValues)) {
+                    .getVariablesValue(rowManager, ExternalPostProcessorVariableType.REQUEST_VARIABLES, grpcSourceConfig.getVariables(), resultFuture);
+            if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, grpcSourceConfig.getVariables(), requestVariablesValues)) {
                 return;
             }
 

dagger-core/src/main/java/io/odpf/dagger/core/processors/external/http/HttpAsyncConnector.java

@@ -15,6 +15,7 @@
 import org.apache.flink.types.Row;
 
 import io.odpf.dagger.core.utils.Constants;
+import io.odpf.dagger.core.utils.Constants.ExternalPostProcessorVariableType;
 import org.asynchttpclient.AsyncHttpClient;
 import org.asynchttpclient.BoundRequestBuilder;
 import org.slf4j.Logger;
@@ -94,12 +95,14 @@ protected void process(Row input, ResultFuture<Row> resultFuture) {
             RowManager rowManager = new RowManager(input);
 
             Object[] requestVariablesValues = getEndpointHandler()
-                    .getEndpointOrQueryVariablesValues(rowManager, resultFuture);
-            if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, requestVariablesValues)) {
+                    .getVariablesValue(rowManager, ExternalPostProcessorVariableType.REQUEST_VARIABLES, httpSourceConfig.getRequestVariables(), resultFuture);
+            Object[] dynamicHeaderVariablesValues = getEndpointHandler()
+                    .getVariablesValue(rowManager, ExternalPostProcessorVariableType.HEADER_VARIABLES, httpSourceConfig.getHeaderVariables(), resultFuture);
+            if (getEndpointHandler().isQueryInvalid(resultFuture, rowManager, httpSourceConfig.getRequestVariables(), requestVariablesValues) || getEndpointHandler().isQueryInvalid(resultFuture, rowManager, httpSourceConfig.getHeaderVariables(), dynamicHeaderVariablesValues)) {
                 return;
             }
 
-            BoundRequestBuilder request = HttpRequestFactory.createRequest(httpSourceConfig, httpClient, requestVariablesValues);
+            BoundRequestBuilder request = HttpRequestFactory.createRequest(httpSourceConfig, httpClient, requestVariablesValues, dynamicHeaderVariablesValues);
             HttpResponseHandler httpResponseHandler = new HttpResponseHandler(httpSourceConfig, getMeterStatsManager(),
                     rowManager, getColumnNameManager(), getOutputDescriptor(resultFuture), resultFuture, getErrorReporter(), new PostResponseTelemetry());
             httpResponseHandler.startTimer();