DateTime backend exception following User Defined Function

[BenComer98]

What happened?

Release: 1.1.0
Platform: Windows, using Powershell
UI: Google Chrome

Background: I'm trying to use a User Defined Function to define a Priority Score for a dataset I'm using. To do so, I'm using a standard CSV File scan to pull in data from a csv. I've attached the CSV as a dataset, I'm pulling it through a UDF that adds a field "priority_score," and I'm sorting in descending order by that score. The execution unit crashes and produces the error shown in the log section.

I have attached the files I used to produce this error. I tried using a smaller, made-up-for-this-issue csv, but it didn't error out.

appointments.csv <- CSV

Issue Repro.json <- Exported Issue repro

How to reproduce?

1. Add a dataset containing appointments.csv to Texera.
2. Add a Workflow. In that Workflow, add a CSV File Scan, a Python UDF (normal, no variation), and a Sort, and connect them in that order.
3. In the CSV File Scan, select the appointments.csv file with headers on.
4. In the Python UDF, add a new field called "priority_score" with type double and paste the following:
   `from pytexera import *
   from datetime import date
   import math

class ProcessTupleOperator(UDFOperatorV2):

@overrides
def process_tuple(self, tuple_: Tuple, port: int) -> Iterator[Optional[TupleLike]]:
    appointment_date = tuple_["appointment_datetime"].date()

    today = date.today()
    days_until = (appointment_date - today).days
    if days_until >= 0:
        tuple_["priority_score"] = 1 / math.log(days_until + 2) + 1
    else:
        tuple_["priority_score"] = 0.0

    # Output the tuple
    yield tuple_

`
5. In the Sort function, sort by priority_score DESC.
6. Add a Computing Unit and Run. Your backend will throw errors.

Version

1.1.0-incubating (Pre-release/Master)

Commit Hash (Optional)

No response

What browsers are you seeing the problem on?

No response

Relevant log output

texera-computing-unit-master            | [2025-12-05 00:08:11,439] [WARN] [edu.uci.ics.amber.util.ArrowUtils$] [flight-server-default-executor-0] - Caught error during parsing Arrow value back to Texera value
texera-computing-unit-master            | edu.uci.ics.amber.core.tuple.AttributeTypeUtils$AttributeTypeException: Failed to parse type java.time.LocalDateTime to Timestamp: 2026-09-05T01:01:12
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$$anonfun$parseTimestamp$2.applyOrElse(AttributeTypeUtils.scala:219)
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$$anonfun$parseTimestamp$2.applyOrElse(AttributeTypeUtils.scala:215)
texera-computing-unit-master            |       at scala.util.Failure.recover(Try.scala:233)
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$.parseTimestamp(AttributeTypeUtils.scala:215)
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$.parseField(AttributeTypeUtils.scala:128)
texera-computing-unit-master            |       at edu.uci.ics.amber.util.ArrowUtils$.$anonfun$getTexeraTuple$1(ArrowUtils.scala:82)
texera-computing-unit-master            |       at scala.collection.StrictOptimizedIterableOps.map(StrictOptimizedIterableOps.scala:100)
texera-computing-unit-master            |       at scala.collection.StrictOptimizedIterableOps.map$(StrictOptimizedIterableOps.scala:87)
texera-computing-unit-master            |       at scala.collection.convert.JavaCollectionWrappers$JCollectionWrapper.map(JavaCollectionWrappers.scala:98)
texera-computing-unit-master            |       at edu.uci.ics.amber.util.ArrowUtils$.getTexeraTuple(ArrowUtils.scala:77)
texera-computing-unit-master            |       at edu.uci.ics.amber.engine.architecture.pythonworker.AmberProducer.$anonfun$acceptPut$2(PythonProxyServer.scala:143)
texera-computing-unit-master            |       at edu.uci.ics.amber.engine.architecture.pythonworker.AmberProducer.$anonfun$acceptPut$2$adapted(PythonProxyServer.scala:142)
texera-computing-unit-master            |       at scala.collection.immutable.Range.foreach(Range.scala:190)
texera-computing-unit-master            |       at edu.uci.ics.amber.engine.architecture.pythonworker.AmberProducer.$anonfun$acceptPut$1(PythonProxyServer.scala:142)
texera-computing-unit-master            |       at org.apache.arrow.flight.FlightService.lambda$doPutCustom$0(FlightService.java:233)
texera-computing-unit-master            |       at io.grpc.Context$1.run(Context.java:566)
texera-computing-unit-master            |       at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
texera-computing-unit-master            |       at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
texera-computing-unit-master            |       at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
texera-computing-unit-master            |       at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
texera-computing-unit-master            |       at java.base/java.lang.Thread.run(Thread.java:829)
texera-computing-unit-master            | Caused by: edu.uci.ics.amber.core.tuple.AttributeTypeUtils$AttributeTypeException: Unsupported type for parsing to Timestamp: java.time.LocalDateTime
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$.$anonfun$parseTimestamp$1(AttributeTypeUtils.scala:209)
texera-computing-unit-master            |       at scala.util.Try$.apply(Try.scala:210)
texera-computing-unit-master            |       at edu.uci.ics.amber.core.tuple.AttributeTypeUtils$.parseTimestamp(AttributeTypeUtils.scala:202)

[chenlica]

@BenComer98 Thanks for the reporting the issue. @aglinxinyuan Can you take a look?

[aglinxinyuan]

ok

[aglinxinyuan]

I followed the steps and cannot reproduce the error.

The workflow runs ok.

workflow.json

Maybe @yunyad can give it a try?

[chenlica]

@yunyad Please try it as well.

[BenComer98]

I followed the steps and cannot reproduce the error.

The workflow runs ok. workflow.json

Maybe @yunyad can give it a try?

I copied over this workflow to my local setup. It still errors for me in the same way. I'm suspecting this may be a local issue. I'm going to try a fresh install to see if that resolves it.

[yunyad]

I remember encountering this error locally before; it was caused by a timezone issue with Python datetimes.
This time I’m unable to reproduce it.

[Yicong-Huang]

I notice the platform is windows. maybe we need windows for repro?

[aglinxinyuan]

I notice the platform is windows. maybe we need windows for repro?

I used Windows and cannot reproduce it.

[carloea2]

I used windows and I can reproduce it

[carloea2]

Reason: fieldValue is a java.time.LocalDateTime, not a java.sql.Timestamp, so none of the String / Long / Timestamp / Date cases match and it falls into case _ => and throws.

texera/common/workflow-core/src/main/scala/org/apache/texera/amber/core/tuple/AttributeTypeUtils.scala

Lines 200 to 212 in 4f822e3

	@throws[AttributeTypeException]
	def parseTimestamp(fieldValue: Any): Timestamp = {
	val attempt: Try[Timestamp] = Try {
	fieldValue match {
	case str: String => new Timestamp(DateParserUtils.parseDate(str.trim).getTime)
	case long: java.lang.Long => new Timestamp(long)
	case timestamp: Timestamp => timestamp
	case date: java.util.Date => new Timestamp(date.getTime)
	// Integer, Double, Boolean, Binary are considered to be illegal here.
	case _ =>
	throw new AttributeTypeException(
	s"Unsupported type for parsing to Timestamp: ${fieldValue.getClass.getName}"
	)