Add targeted SQL data-fix script to remap mislabeled Mikrodilution tests to Micronaut

[Copilot]

Some AntimicrobialSensitivityTest rows were labeled with TestingMethod = 1 (Microdilution / “Mikrodilution”) even though they belong to Micronaut. This PR adds a focused data-fix script that remaps only the affected SentinelEntry ranges provided by the lab.

• What changed

  • Added Scripts/2026-05-10-FixTestMethodMikrodilutionToMicronaut.sql.
  • Script contains:
    • a short change header,
    • an enhanced pre-update verification SELECT that joins SentinelEntries and includes:
      • SenderLaboratoryNumber
      • computed LaboratoryNumber (SN-<Year>-<YearlySequentialEntryNumber>)
    • a scoped UPDATE from TestingMethod = 1 to TestingMethod = 3.
  • Added SQL lint configuration file .sqlfluff to correctly configure SQL dialect (tsql) for CI linting.

• Update scope

  • Applies only to rows where TestingMethod = 1.
  • Restricts affected entries via compact SentinelEntry ID ranges:
    • 1304–1598
    • 1683–1693
    • 1795–1801
    • 1942
    • 2788–2805

• Implementation snippet

  UPDATE AntimicrobialSensitivityTest
  SET TestingMethod = 3 -- Micronaut
  WHERE TestingMethod = 1 -- Microdilution (displayed as "Mikrodilution")
    AND SentinelEntryId IN (
        SELECT Id FROM SentinelEntries
        WHERE (Id BETWEEN 1304 AND 1598)
           OR (Id BETWEEN 1683 AND 1693)
           OR (Id BETWEEN 1795 AND 1801)
           OR Id = 1942
           OR (Id BETWEEN 2788 AND 2805)
    );

Original prompt

Background

Issue #218 reports that a set of AntimicrobialSensitivityTest entries are mislabelled as Microdilution (displayed as "Mikrodilution" in the UI) but should be Micronaut.

The SpeciesTestingMethod enum in NRZMyk.Services/Data/Entities/SpeciesTestingMethod.cs defines:

public enum SpeciesTestingMethod
{
    Vitek,         // 0
    Microdilution, // 1  — Description: "Mikrodilution"
    YeastOne,      // 2  — Description: "Yeast One"
    Micronaut,     // 3
    ETest          // 4  — Description: "E-Test"
}

So the fix is: set TestingMethod = 3 (Micronaut) where TestingMethod = 1 (Mikrodilution) for the affected sentinel entries.

Task

Add a new SQL script at Scripts/2026-05-10-FixTestMethodMikrodilutionToMicronaut.sql following the conventions of existing scripts in that folder (e.g. Scripts/2023-06-04-FixMicResistanceBasedOnReferenceMethod.sql).

The script should:

1. Start with a comment block describing when / what the change is.
2. Include a verification SELECT so the affected rows can be checked before running the UPDATE.
3. Include the UPDATE statement.

The affected SentinelEntryId values come from a CSV provided by the laboratory. They fall into these compact ID ranges (use ranges, not individual IDs):

Range	Notes
1304 – 1598	main batch, SN-2022-xxxx
1683 – 1693	additional SN-2022-xxxx
1795 – 1801	SN-2023-xxxx early batch
1942	single SN-2023-0241 entry (only its Mikrodilution test remains to fix)
2788 – 2805	SN-2024-xxxx

Use BETWEEN for ranges and OR to combine them. The TestingMethod = 1 filter ensures only currently-mislabelled rows are updated (already-correct Micronaut tests, e.g. on SentinelEntry 1942, will not be touched).

Keep the script minimal — no per-row detail, just the ranges above and the two TestingMethod integer constants with inline comments explaining what they mean.

The UPDATE should look roughly like:

-- ~NNN rows affected (verify with SELECT above before running)
UPDATE AntimicrobialSensitivityTest
SET TestingMethod = 3   -- Micronaut
WHERE TestingMethod = 1 -- Microdilution (displayed as "Mikrodilution")
  AND SentinelEntryId IN (
      SELECT Id FROM SentinelEntries
      WHERE (Id BETWEEN 1304 AND 1598)
         OR (Id BETWEEN 1683 AND 1693)
         OR (Id BETWEEN 1795 AND 1801)
         OR Id = 1942
         OR (Id BETWEEN 2788 AND 2805)
  );

Match the style and commenting conventions of the existing scripts in Scripts/.

The following is the prior conversation context from the user's chat exploration (may be truncated):

User: Please prepare a SQL script for this Issue @markusrt/NRZMyk/issues/218

Raise a PR that adds a new SQL script to the Scripts/ folder.

CSV on records to change is attached. Information from Email: "im Anhang die Tabelle mit den SN Nummern bei denen die Testmethode von Mikrodilution auf Micronaut geändert werden muss."

Important: Do include as less as possible detailed information to the script. Some parts of internal IDs, e.g. a range is fine, but the script should, if possible, not contain every detailed ID.

Assistant: Now I have everything I need. The enum shows:

• Microdilution = 1 (displayed as "Mikrodilution")
• Micronaut = 3

From the CSV, the affected SentinelEntry IDs cluster into these groups: 1304–1598, 1683–1693, 1795–1801, 1942, and 2788–2805. The TestingMethod = 1 filter ensures only actually-labelled Mikrodilution rows are touched (e.g. ID 1942's already-correct Micronaut tests won't be changed).

Assistant: We're waiting for your response to the tool confirmation dialog to continue creating the pull request.

This pull request was created from Copilot chat.

[markusrt]

@copilot Please fix linting config, it does not seem to be correctly setup for SQL yet: https://github.com/markusrt/NRZMyk/actions/runs/25628461114

[markusrt]

@copilot please add the laboratorynumber and senderlaboratorynumber to this select via join so a human can verify the data is correct.

[Copilot]

Implemented in fb87f25. The verification SELECT now joins SentinelEntries and includes both SenderLaboratoryNumber and computed LaboratoryNumber so the affected rows can be reviewed by humans before running the UPDATE.

[Copilot]

@copilot Please fix linting config, it does not seem to be correctly setup for SQL yet: https://github.com/markusrt/NRZMyk/actions/runs/25628461114

Implemented in df2c714. I added a repository-level .sqlfluff config so SQLFluff uses the tsql dialect in CI, and documented the excluded SQLFluff rules to keep linting focused on parse/syntax validity for our data-fix SQL scripts.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud