[Bug] Conversion error - Couldn't find param for shortName Identification

[FEVLuM]

Description

While converting several different PDX files the converter fails with Error while converting file .pdx: Couldn't find param for shortName Identification:
java.lang.IllegalStateException: Couldn't find param for shortName Identification

Steps to Reproduce

java -jar converter/build/libs/converter-all.jar /mnt/hgfs/temp/CARAlyzer_VM_exchange/CDD_PDX_MDD/.pdx

Expected Behavior

Successful conversion from PDX to MDD

Processing <ecuname>.pdx
[2026-06-24 05:56:54.169Z] [INFO   ] Converting '<ecuname>.pdx' to mdd
[2026-06-24 05:57:04.198Z] [INFO   ] Chunk '<ecuname>' (DIAGNOSTIC_DESCRIPTION) was added to the file with 1,553,390 bytes of data, initial size: 8,272,624 bytes
[2026-06-24 05:57:04.213Z] [INFO   ] Writing database took 5.542918220s total (compression: 5.323728652s) - sizes: odx raw: 140,046,377 bytes, uncompressed chunks: 8,272,624 bytes, compressed mdd: 1,553,640 bytes

Actual Behavior

Conversion fails due to "Couldn't find param shortName Identification"

Screenshots or Logs (if applicable)

[2026-06-11 17:48:19.084Z] [SEVERE ] Error while converting file <ecuname>.pdx: Couldn't find param for shortName Identification:
java.lang.IllegalStateException: Couldn't find param for shortName Identification
	at DatabaseWriter.offset(DatabaseWriter.kt:2098)
	at DatabaseWriter.offset(DatabaseWriter.kt:2118)
	at DatabaseWriter.offset(DatabaseWriter.kt:1900)
	at DatabaseWriter.<init>(DatabaseWriter.kt:231)
	at ChunkBuilder.createEcuDataChunk(ChunkBuilder.kt:119)
	at FileConverter.convert(Converter.kt:177)
	at Converter.run$lambda$5$lambda$4(Converter.kt:378)
	at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:572)
	at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:317)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
	at java.base/java.lang.Thread.run(Thread.java:1583)

Environment

• OS: Ubuntu26.04 LTS
• Version: odx-converter - version: 0.1.0-SNAPSHOT

Additional Context

Important

Link the pull request that addresses this bug once created.

[floroks]

Can you describe the structure of the PDX, how that shortName should be resolved? (e.g. structure until that element, and structure for the element that references it). Maybe even an LLM can assist with extracting the relevant data/a minimal reproducer?

[FEVLuM]

Stick with the help of claude deeper into it to solve the issue from PDX file perspective:
Identified parameter naming mismatch causing configuration error

There’s the error — and it is indeed due to an inconsistency in the master data of the PDX file, not in the converter.

The POS-RESPONSE from ActiveDiagnosticInformation_Read has three parameters:

• SID_PR
• RecordDataIdentifier
• ActiveDiagnosticInformation ← this is the actual data parameter

There is no parameter called Identification. However, all 11 ECU-VARIANT-PATTERN blocks refer, via

<OUT-PARAM-IF-SNREF SHORT-NAME="Identification"/>

to a parameter with exactly that name — which simply does not exist. The actual name is ActiveDiagnosticInformation.

Cause

This is a data authorisation inconsistency within the PDX itself (presumably the parameter was renamed from ‘Identification’ to ‘ActiveDiagnosticInformation’ at some point, but the SNREFs in the ECU-VARIANT-PATTERN blocks were not updated accordingly). Other PDX files that work for you probably still refer to this parameter as ‘Identification’ or reference the patterns correctly.

Option A — XML patch (fastest fix, local):

Replace all 11 instances

sed -i 's/<OUT-PARAM-IF-SNREF SHORT-NAME="Identification"\/>/<OUT-PARAM-IF-SNREF SHORT-NAME="ActiveDiagnosticInformation"\/>/g' <ecuname>.odx-d

Option B — Make the converter more robust:

If you have access to the Kotlin source code (ChunkBuilder.kt / DatabaseWriter.kt), you could implement a fallback there: if the referenced PARAM cannot be found via shortName, log a warning instead of terminating the process abruptly, or simply ignore the ECU-VARIANT-PATTERN (as it is only needed for automatic variant detection, not for the core diagnostic data).

Solved wit Option A

java -jar converter/build/libs/converter-all.jar <ecuname>_patch.pdx 
odx-converter - version: 0.1.0-SNAPSHOT

Processing ecuname>_patch.pdx
[2026-06-24 10:02:13.440Z] [INFO   ] Converting '<ecuname>.pdx' to mdd
[2026-06-24 10:02:19.640Z] [INFO   ] Chunk '<ecuname>' (DIAGNOSTIC_DESCRIPTION) was added to the file with 1,127,804 bytes of data, initial size: 4,976,448 bytes
[2026-06-24 10:02:19.661Z] [INFO   ] Writing database took 3.587392687s total (compression: 3.337016261s) - sizes: odx raw: 78,709,230 bytes, uncompressed chunks: 4,976,448 bytes, compressed mdd: 1,128,055 bytes

@floroks : Can you check Option B as patching the different PDX files always should not be the best option.

[floroks]

Option B would be theoretically possible (with a "create broken mdd, idc" option), but a MDD converted with that option would not be able to include that specific parameter, due to it missing. At runtime, the CDA would not be able to resolve that parameter and run into errors.

In your case (ActiveDiagnosticInformation not working), this would likely mean that even ECU Detection fails with "wild" runtime errors.