mengw15 commented on code in PR #5898: URL: https://github.com/apache/texera/pull/5898#discussion_r3457956314
########## common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/huggingFace/HuggingFaceIrisLogisticRegressionOpDescSpec.scala: ########## @@ -0,0 +1,114 @@ +/* + * 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.texera.amber.operator.huggingFace + +import org.apache.texera.amber.core.tuple.{AttributeType, Schema} +import org.apache.texera.amber.operator.LogicalOp +import org.apache.texera.amber.operator.metadata.OperatorGroupConstants +import org.apache.texera.amber.util.JSONUtils.objectMapper +import org.scalatest.flatspec.AnyFlatSpec +import org.scalatest.matchers.should.Matchers + +import java.nio.charset.StandardCharsets +import java.util.Base64 + +class HuggingFaceIrisLogisticRegressionOpDescSpec extends AnyFlatSpec with Matchers { + + private def b64(s: String): String = + Base64.getEncoder.encodeToString(s.getBytes(StandardCharsets.UTF_8)) + + // EncodableString fields are always base64-wrapped in .encode mode + // (self.decode_python_template('<base64>')), so assert on the base64 form only rather than + // the raw column name, which could appear in the generated Python for unrelated reasons. + private def carries(output: String, name: String): Boolean = + output.contains(b64(name)) + + private def configured(): HuggingFaceIrisLogisticRegressionOpDesc = { + val d = new HuggingFaceIrisLogisticRegressionOpDesc + d.petalLengthCmAttribute = "petalLength" + d.petalWidthCmAttribute = "petalWidth" + d.predictionClassName = "species" + d.predictionProbabilityName = "probability" + d + } + + "HuggingFaceIrisLogisticRegressionOpDesc.operatorInfo" should + "advertise the name, Hugging Face group, and a 1-in/1-out shape" in { + val info = (new HuggingFaceIrisLogisticRegressionOpDesc).operatorInfo + info.userFriendlyName shouldBe "Hugging Face Iris Logistic Regression" + info.operatorDescription shouldBe + "Predict whether an iris is an Iris-setosa using a pre-trained logistic regression model" + info.operatorGroupName shouldBe OperatorGroupConstants.HUGGINGFACE_GROUP + info.inputPorts should have length 1 + info.outputPorts should have length 1 + } + + "HuggingFaceIrisLogisticRegressionOpDesc" should "default all column fields to null" in { + val d = new HuggingFaceIrisLogisticRegressionOpDesc + d.petalLengthCmAttribute shouldBe null + d.petalWidthCmAttribute shouldBe null + d.predictionClassName shouldBe null + d.predictionProbabilityName shouldBe null + } + + "HuggingFaceIrisLogisticRegressionOpDesc.getOutputSchemas" should + "reject a blank prediction result name" in { + val d = new HuggingFaceIrisLogisticRegressionOpDesc + val in = Schema().add("sepal", AttributeType.STRING) + val ex = intercept[RuntimeException] { + d.getOutputSchemas(Map(d.operatorInfo.inputPorts.head.id -> in)) + } + ex.getMessage shouldBe "Result attribute name should not be empty" + } + + it should "append a STRING class column and a DOUBLE probability column, keyed by the output port" in { + val d = configured() + val in = Schema().add("sepal", AttributeType.STRING) + val out = d.getOutputSchemas(Map(d.operatorInfo.inputPorts.head.id -> in)) + val schema = out(d.operatorInfo.outputPorts.head.id) + schema.getAttribute("sepal").getType shouldBe AttributeType.STRING + schema.getAttribute("species").getType shouldBe AttributeType.STRING + schema.getAttribute("probability").getType shouldBe AttributeType.DOUBLE + } + + "HuggingFaceIrisLogisticRegressionOpDesc.generatePythonCode" should + "emit the logistic-regression operator carrying the configured columns (encoded)" in { + val d = configured() + val code = d.generatePythonCode() + code should include("class ProcessTupleOperator(UDFOperatorV2)") + code should include("LinearModel.from_pretrained") + code should include("sadhaklal/logistic-regression-iris") + code should include("self.decode_python_template(") + carries(code, "petalLength") shouldBe true + carries(code, "species") shouldBe true + } + + "HuggingFaceIrisLogisticRegressionOpDesc" should + "round-trip its config fields through the polymorphic base" in { + val d = configured() + val restored = objectMapper.readValue(objectMapper.writeValueAsString(d), classOf[LogicalOp]) + restored shouldBe a[HuggingFaceIrisLogisticRegressionOpDesc] + val h = restored.asInstanceOf[HuggingFaceIrisLogisticRegressionOpDesc] + h.petalLengthCmAttribute shouldBe "petalLength" + h.petalWidthCmAttribute shouldBe "petalWidth" + h.predictionClassName shouldBe "species" + h.predictionProbabilityName shouldBe "probability" + } +} Review Comment: Within-PR consistency: the other three Hugging Face specs in this PR each include a `getPhysicalOp` test that asserts `OpExecWithCode(..., "python")` plus port-identity carry-forward (`physical.inputPorts.keySet shouldBe d.operatorInfo.inputPorts.map(_.id).toSet`). The Iris spec doesn't — it never imports `OpExecWithCode` / `WorkflowIdentity` / `ExecutionIdentity` and skips the test entirely. Iris extends the same `PythonOperatorDescriptor` base as its siblings, so `getPhysicalOp` is on the same contract surface. Mirroring the test (≈10 lines) would bring this spec up to parity with the others. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
