This is an automated email from the ASF dual-hosted git repository.
aglinxinyuan pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/texera.git
The following commit(s) were added to refs/heads/main by this push:
new 9b2005d5b6 test(amber): add unit tests for visualization OpDescs
(Image, FunnelPlot, VolcanoPlot, BulletChart) (#4809)
9b2005d5b6 is described below
commit 9b2005d5b6718e3a6b3181331bbd0a4d245e130e
Author: Yicong Huang <[email protected]>
AuthorDate: Sun May 3 11:03:01 2026 -0700
test(amber): add unit tests for visualization OpDescs (Image, FunnelPlot,
VolcanoPlot, BulletChart) (#4809)
### What changes were proposed in this PR?
Adds scalatest coverage for four visualization operator descriptors
under `common/workflow-operator/.../visualization/`. The existing thin
spec on `ImageVisualizerOpDesc` is extended; specs for
`FunnelPlotOpDesc`, `VolcanoPlotOpDesc`, and `BulletChartOpDesc` are
new.
The specs verify (per OpDesc) the user-friendly name + group +
outputPorts wiring, the single-port `html-content` STRING schema
returned by `getOutputSchemas`, the rendered Python source from
`generatePythonCode` (operator class + plotly imports +
`decode_python_template` runtime decode sites for each `EncodableString`
field), and the missing-required-field assertion behavior — which
differs by OpDesc:
- `ImageVisualizerOpDesc.binaryContent` defaults to `null` (`var ... =
_`), so the assert path raises `NullPointerException` when called on
null.
- `FunnelPlotOpDesc.x` / `.y` default to `""`, so the same assert raises
`AssertionError`.
- `VolcanoPlotOpDesc` and `BulletChartOpDesc` have no asserts in
`generatePythonCode` — empty defaults render valid Python source.
### Any related issues, documentation, discussions?
Closes #4807.
Bug filed separately: most visualization OpDescs (29 files) have a
duplicate `Map(operatorInfo.outputPorts.head.id -> outputSchema)`
statement in `getOutputSchemas` — the first call is computed and
discarded, only the second is the actual return value. Likely a
copy-paste artifact.
### How was this PR tested?
```
sbt scalafmtCheckAll
sbt "WorkflowOperator/testOnly
org.apache.texera.amber.operator.visualization.ImageViz.ImageVisualizerOpDescSpec
org.apache.texera.amber.operator.visualization.funnelPlot.FunnelPlotOpDescSpec
org.apache.texera.amber.operator.visualization.volcanoPlot.VolcanoPlotOpDescSpec
org.apache.texera.amber.operator.visualization.bulletChart.BulletChartOpDescSpec"
```
### Was this PR authored or co-authored using generative AI tooling?
Generated-by: Claude Code (claude-opus-4-7)
---------
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
---
.../ImageViz/ImageVisualizerOpDescSpec.scala | 59 +++++++++-
.../bulletChart/BulletChartOpDescSpec.scala | 119 +++++++++++++++++++++
.../funnelPlot/FunnelPlotOpDescSpec.scala | 95 ++++++++++++++++
.../volcanoPlot/VolcanoPlotOpDescSpec.scala | 95 ++++++++++++++++
4 files changed, 366 insertions(+), 2 deletions(-)
diff --git
a/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/ImageViz/ImageVisualizerOpDescSpec.scala
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/ImageViz/ImageVisualizerOpDescSpec.scala
index 266f14e4d8..f03271aa36 100644
---
a/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/ImageViz/ImageVisualizerOpDescSpec.scala
+++
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/ImageViz/ImageVisualizerOpDescSpec.scala
@@ -19,18 +19,73 @@
package org.apache.texera.amber.operator.visualization.ImageViz
+import org.apache.texera.amber.core.tuple.AttributeType
+import org.apache.texera.amber.operator.metadata.OperatorGroupConstants
import org.scalatest.BeforeAndAfter
import org.scalatest.flatspec.AnyFlatSpec
+import org.scalatest.matchers.should.Matchers
-class ImageVisualizerOpDescSpec extends AnyFlatSpec with BeforeAndAfter {
+class ImageVisualizerOpDescSpec extends AnyFlatSpec with BeforeAndAfter with
Matchers {
var opDesc: ImageVisualizerOpDesc = _
before {
opDesc = new ImageVisualizerOpDesc()
}
- it should "throw assertion error if BinaryContent is empty" in {
+ it should "currently throw NullPointerException when binaryContent is
uninitialized" in {
+ // Documents the present behavior without claiming it is the contract:
+ // `binaryContent` is declared `var binaryContent: EncodableString = _`,
+ // so an uninitialized reference field defaults to null and the
+ // `assert(binaryContent.nonEmpty)` inside `createBinaryData` reaches
+ // `null.nonEmpty` and throws NPE before the assert message can fire.
assertThrows[NullPointerException] {
opDesc.createBinaryData()
}
}
+
+ it should "eventually reject missing binaryContent with a controlled error
(pendingUntilFixed)" in pendingUntilFixed {
+ // Intended contract: because `binaryContent` is declared
+ // `@JsonProperty(required = true)`, an unconfigured operator should
+ // surface a domain error (AssertionError or IllegalArgumentException),
+ // not an NPE from dereferencing null. Using pendingUntilFixed so a
+ // future validation fix flips this test from Pending to a deliberate
+ // failure that forces removal of the marker.
+ val ex = intercept[RuntimeException] {
+ opDesc.createBinaryData()
+ }
+ ex shouldBe a[AssertionError]
+ }
+
+ "ImageVisualizerOpDesc.operatorInfo" should "advertise the user-friendly
name and Media group" in {
+ val info = opDesc.operatorInfo
+ info.userFriendlyName shouldBe "Image Visualizer"
+ info.operatorGroupName shouldBe
OperatorGroupConstants.VISUALIZATION_MEDIA_GROUP
+ info.operatorDescription should include("image")
+ }
+
+ it should "expose exactly one output port wired through forVisualization" in
{
+ opDesc.operatorInfo.outputPorts should have length 1
+ }
+
+ "ImageVisualizerOpDesc.getOutputSchemas" should "return a single-port schema
with an html-content STRING column" in {
+ opDesc.binaryContent = "image_bytes"
+ val schemas = opDesc.getOutputSchemas(Map.empty)
+ schemas should have size 1
+ val (portId, schema) = schemas.head
+ portId shouldBe opDesc.operatorInfo.outputPorts.head.id
+ schema.getAttributes should have length 1
+ schema.getAttributes.head.getName shouldBe "html-content"
+ schema.getAttributes.head.getType shouldBe AttributeType.STRING
+ }
+
+ "ImageVisualizerOpDesc.generatePythonCode" should "render a UDFOperatorV2
source with a runtime column-decode site" in {
+ // EncodableString fields are NOT emitted as literal strings — the pyb
+ // macro wraps them in `self.decode_python_template.decode("<base64>")`
+ // calls so the column name resolves at runtime. Verify the structure
+ // (operator class, body helper, decode site) instead of a literal name.
+ opDesc.binaryContent = "image_bytes"
+ val code = opDesc.generatePythonCode()
+ code should include("class ProcessTupleOperator(UDFOperatorV2)")
+ code should include("encode_image_to_html")
+ code should include("decode_python_template")
+ }
}
diff --git
a/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/bulletChart/BulletChartOpDescSpec.scala
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/bulletChart/BulletChartOpDescSpec.scala
new file mode 100644
index 0000000000..fff547b939
--- /dev/null
+++
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/bulletChart/BulletChartOpDescSpec.scala
@@ -0,0 +1,119 @@
+/*
+ * 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.visualization.bulletChart
+
+import org.apache.texera.amber.core.tuple.AttributeType
+import org.apache.texera.amber.operator.metadata.OperatorGroupConstants
+import org.scalatest.flatspec.AnyFlatSpec
+import org.scalatest.matchers.should.Matchers
+
+import java.util
+import java.util.{List => JList}
+
+class BulletChartOpDescSpec extends AnyFlatSpec with Matchers {
+
+ private def configured: BulletChartOpDesc = {
+ val op = new BulletChartOpDesc
+ op.value = "actualValue"
+ op.deltaReference = "100"
+ op
+ }
+
+ "BulletChartOpDesc.operatorInfo" should "advertise the user-friendly name
and Financial group" in {
+ val info = (new BulletChartOpDesc).operatorInfo
+ info.userFriendlyName shouldBe "Bullet Chart"
+ info.operatorGroupName shouldBe
OperatorGroupConstants.VISUALIZATION_FINANCIAL_GROUP
+ info.operatorDescription should include("Bullet Chart")
+ }
+
+ it should "expose exactly one output port wired through forVisualization" in
{
+ (new BulletChartOpDesc).operatorInfo.outputPorts should have length 1
+ }
+
+ "BulletChartOpDesc.getOutputSchemas" should "return a single-port schema
with an html-content STRING column" in {
+ val op = configured
+ val schemas = op.getOutputSchemas(Map.empty)
+ schemas should have size 1
+ val (portId, schema) = schemas.head
+ portId shouldBe op.operatorInfo.outputPorts.head.id
+ schema.getAttributes should have length 1
+ schema.getAttributes.head.getName shouldBe "html-content"
+ schema.getAttributes.head.getType shouldBe AttributeType.STRING
+ }
+
+ "BulletChartOpDesc.generatePythonCode" should "render Python source with a
runtime decode site for the value column" in {
+ // EncodableString fields are NOT emitted as literal strings — the pyb
+ // macro wraps them in `self.decode_python_template.decode("<base64>")`
+ // calls. The rendered source must reference the decoder symbol at least
+ // for `value` and `deltaReference`.
+ val code = configured.generatePythonCode()
+ code should include("plotly.graph_objects")
+ val decodeOccurrences = "decode_python_template".r.findAllIn(code).length
+ decodeOccurrences should be >= 2
+ }
+
+ it should "default to an empty steps list when none are configured" in {
+ // The bullet-chart template ships with several unrelated `[]` literals
+ // (`colors`, `valid_steps`, `step_errors`, `steps_list`, `html_chunks`),
+ // so a bare `code should include("[]")` is too weak. Anchor on the
+ // generated `steps_data = ...` literal directly so a regression that
+ // makes it non-empty would actually fail the assertion.
+ val code = configured.generatePythonCode()
+ code should include regex """steps_data\s*=\s*\[\]"""
+ }
+
+ it should "include each configured step's start/end JSON keys with extra
decode sites" in {
+ val op = configured
+ val steps: JList[BulletChartStepDefinition] = new
util.ArrayList[BulletChartStepDefinition]()
+ steps.add(new BulletChartStepDefinition("0", "50"))
+ steps.add(new BulletChartStepDefinition("50", "100"))
+ op.steps = steps
+ val code = op.generatePythonCode()
+ code should include("\"start\":")
+ code should include("\"end\":")
+ // Two steps × 2 EncodableString fields each = 4 extra decode sites on
+ // top of the value/deltaReference decodes from the base configuration.
+ val baseDecodes =
"decode_python_template".r.findAllIn(configured.generatePythonCode()).length
+ val withSteps = "decode_python_template".r.findAllIn(code).length
+ withSteps shouldBe baseDecodes + 4
+ }
+
+ it should "currently render a code block even with the default empty
configuration (no assert guard)" in {
+ // Documents the present behavior: BulletChartOpDesc has no assert
+ // guards inside generatePythonCode, so empty defaults still produce
+ // syntactically valid Python source. The intended contract lives in
+ // the pendingUntilFixed test below.
+ val op = new BulletChartOpDesc
+ val code = op.generatePythonCode()
+ code should include("plotly.graph_objects")
+ }
+
+ it should "eventually reject empty required value/deltaReference like
FunnelPlot/ImageVisualizer (pendingUntilFixed)" in pendingUntilFixed {
+ // Intended contract: `value` and `deltaReference` are marked required
+ // on `BulletChartOpDesc`, so generatePythonCode on a default-constructed
+ // instance should raise instead of rendering empty-string column refs.
+ // Using pendingUntilFixed so a future validation fix flips this test
+ // from Pending to a deliberate failure and forces removal of the marker.
+ val op = new BulletChartOpDesc
+ intercept[RuntimeException] {
+ op.generatePythonCode()
+ }
+ }
+}
diff --git
a/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/funnelPlot/FunnelPlotOpDescSpec.scala
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/funnelPlot/FunnelPlotOpDescSpec.scala
new file mode 100644
index 0000000000..5f46a0d6f7
--- /dev/null
+++
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/funnelPlot/FunnelPlotOpDescSpec.scala
@@ -0,0 +1,95 @@
+/*
+ * 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.visualization.funnelPlot
+
+import org.apache.texera.amber.core.tuple.AttributeType
+import org.apache.texera.amber.operator.metadata.OperatorGroupConstants
+import org.scalatest.flatspec.AnyFlatSpec
+import org.scalatest.matchers.should.Matchers
+
+class FunnelPlotOpDescSpec extends AnyFlatSpec with Matchers {
+
+ private def configured: FunnelPlotOpDesc = {
+ val op = new FunnelPlotOpDesc
+ op.x = "stage"
+ op.y = "count"
+ op
+ }
+
+ "FunnelPlotOpDesc.operatorInfo" should "advertise the user-friendly name and
Financial group" in {
+ val info = (new FunnelPlotOpDesc).operatorInfo
+ info.userFriendlyName shouldBe "Funnel Plot"
+ info.operatorGroupName shouldBe
OperatorGroupConstants.VISUALIZATION_FINANCIAL_GROUP
+ info.operatorDescription should include("Funnel")
+ }
+
+ it should "expose exactly one output port wired through forVisualization" in
{
+ (new FunnelPlotOpDesc).operatorInfo.outputPorts should have length 1
+ }
+
+ "FunnelPlotOpDesc.getOutputSchemas" should "return a single-port schema with
an html-content STRING column" in {
+ val op = configured
+ val schemas = op.getOutputSchemas(Map.empty)
+ schemas should have size 1
+ val (portId, schema) = schemas.head
+ portId shouldBe op.operatorInfo.outputPorts.head.id
+ schema.getAttributes should have length 1
+ schema.getAttributes.head.getName shouldBe "html-content"
+ schema.getAttributes.head.getType shouldBe AttributeType.STRING
+ }
+
+ "FunnelPlotOpDesc.generatePythonCode" should "render a UDFTableOperator
source with runtime decode sites for x and y" in {
+ // EncodableString fields are NOT emitted as literal strings — the pyb
+ // macro wraps them in `self.decode_python_template.decode("<base64>")`
+ // calls. Each configured column becomes one decode site, so x + y must
+ // produce at least two distinct decodes in the rendered source.
+ val code = configured.generatePythonCode()
+ code should include("class ProcessTableOperator(UDFTableOperator)")
+ code should include("plotly.express")
+ val decodeOccurrences = "decode_python_template".r.findAllIn(code).length
+ decodeOccurrences should be >= 2
+ }
+
+ it should "render the optional color argument only when color is configured"
in {
+ val without = configured.generatePythonCode()
+ val withColor = {
+ val op = configured
+ op.color = "category"
+ op.generatePythonCode()
+ }
+ without should not include "color="
+ withColor should include("color=")
+ // With color set, the rendered source has one extra decode site beyond
+ // the two for x and y.
+ val withDecodes = "decode_python_template".r.findAllIn(withColor).length
+ val withoutDecodes = "decode_python_template".r.findAllIn(without).length
+ withDecodes shouldBe withoutDecodes + 1
+ }
+
+ it should "fail-fast when required x/y are unset (the assert guards inside
createPlotlyFigure)" in {
+ // Pin: createPlotlyFigure asserts nonEmpty on both x and y. The fields
+ // are initialized to "" so the assert path is reached (not the NPE path
+ // that ImageVisualizerOpDesc hits).
+ val op = new FunnelPlotOpDesc
+ op.x = ""
+ op.y = ""
+ assertThrows[AssertionError](op.generatePythonCode())
+ }
+}
diff --git
a/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/volcanoPlot/VolcanoPlotOpDescSpec.scala
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/volcanoPlot/VolcanoPlotOpDescSpec.scala
new file mode 100644
index 0000000000..e85ed589ef
--- /dev/null
+++
b/common/workflow-operator/src/test/scala/org/apache/texera/amber/operator/visualization/volcanoPlot/VolcanoPlotOpDescSpec.scala
@@ -0,0 +1,95 @@
+/*
+ * 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.visualization.volcanoPlot
+
+import org.apache.texera.amber.core.tuple.AttributeType
+import org.apache.texera.amber.operator.metadata.OperatorGroupConstants
+import org.scalatest.flatspec.AnyFlatSpec
+import org.scalatest.matchers.should.Matchers
+
+class VolcanoPlotOpDescSpec extends AnyFlatSpec with Matchers {
+
+ private def configured: VolcanoPlotOpDesc = {
+ val op = new VolcanoPlotOpDesc
+ op.effectColumn = "log2fc"
+ op.pvalueColumn = "pvalue"
+ op
+ }
+
+ "VolcanoPlotOpDesc.operatorInfo" should "advertise the user-friendly name
and Scientific group" in {
+ val info = (new VolcanoPlotOpDesc).operatorInfo
+ info.userFriendlyName shouldBe "Volcano Plot"
+ info.operatorGroupName shouldBe
OperatorGroupConstants.VISUALIZATION_SCIENTIFIC_GROUP
+ info.operatorDescription should include("statistical")
+ }
+
+ it should "expose exactly one output port wired through forVisualization" in
{
+ (new VolcanoPlotOpDesc).operatorInfo.outputPorts should have length 1
+ }
+
+ "VolcanoPlotOpDesc.getOutputSchemas" should "return a single-port schema
with an html-content STRING column" in {
+ val op = configured
+ val schemas = op.getOutputSchemas(Map.empty)
+ schemas should have size 1
+ val (portId, schema) = schemas.head
+ portId shouldBe op.operatorInfo.outputPorts.head.id
+ schema.getAttributes should have length 1
+ schema.getAttributes.head.getName shouldBe "html-content"
+ schema.getAttributes.head.getType shouldBe AttributeType.STRING
+ }
+
+ "VolcanoPlotOpDesc.generatePythonCode" should "render a UDFTableOperator
source that decodes both column references" in {
+ // EncodableString fields are NOT emitted as literal column names — the
+ // pyb macro wraps them in `self.decode_python_template.decode("<base64>")`
+ // calls so the column name is resolved at runtime. Verify the structure
+ // (class + import + decode site count) instead of substring matches.
+ val code = configured.generatePythonCode()
+ code should include("class ProcessTableOperator(UDFTableOperator)")
+ code should include("plotly.express")
+ code should include("-log10(pvalue)")
+ val decodeOccurrences = "decode_python_template".r.findAllIn(code).length
+ decodeOccurrences should be >= 2
+ }
+
+ it should "currently render code even when required fields are empty (no
assert guard)" in {
+ // Documents the present behavior: VolcanoPlotOpDesc does not assert on
+ // its required fields inside `generatePythonCode`. An empty
+ // configuration therefore renders syntactically valid Python that
+ // references an empty string. The intended contract is split out into
+ // the pendingUntilFixed test below so this assertion no longer reads
+ // as the contract.
+ val op = new VolcanoPlotOpDesc
+ val code = op.generatePythonCode()
+ code should include("class ProcessTableOperator(UDFTableOperator)")
+ }
+
+ it should "eventually reject empty required fields like
FunnelPlot/ImageVisualizer (pendingUntilFixed)" in pendingUntilFixed {
+ // Intended contract: `effectColumn` and `pvalueColumn` are marked
+ // required on `VolcanoPlotOpDesc`, so generatePythonCode on a
+ // default-constructed instance should raise instead of producing a
+ // string-literal-empty payload. Using pendingUntilFixed so a future
+ // validation fix flips this test from Pending to a deliberate failure
+ // and forces removal of the marker.
+ val op = new VolcanoPlotOpDesc
+ intercept[RuntimeException] {
+ op.generatePythonCode()
+ }
+ }
+}
