Files
GerbilManager/GerbilManager.Tests/IngestResolvedServiceTests.cs
Gulum cf672c0933 feat(provenance): Datenherkunft als chronologischer, datei-attribuierter Verlauf
Die Herkunft liest sich jetzt wie eine History: pro Feld wird genannt, aus
welcher Quelldatei der Wert kam, plus Merge-/Entscheidungs-/Wurfchronik-Schritte.
Beispiel: „In ‚X.xlsx' gefunden." → „Geburtsdatum (…) aus ‚X.xlsx'." → „Auch in
‚Y.docx' gefunden → zusammengeführt." → „Eltern über Position erkannt (Quelle …)."

Import: build_entity_provenance() um history[] erweitert; ein field_source-
Tracking im Gerbil-Dedup hält fest, welcher Datensatz/welche Datei jedes Feld
(DOB/Genotyp/Geschlecht/Farbschlag/Sterbedatum) lieferte — auch über Empty-Fill
und Mehrheitsentscheid hinweg. Kontakte und Würfe erhalten analoge, datei-
attribuierte Verläufe. Kein DB-Migration nötig (history liegt im Provenance-JSON).

Frontend: ProvenanceDialog rendert den Verlauf als nummerierte Timeline.
Tests: history wird erzeugt, nennt Dateien, überlebt Ingest (dotnet 212,
vitest 129, playwright 17, python merge/extract grün).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-22 16:21:24 +02:00

205 lines
8.6 KiB
C#

using GerbilManagerWebAPI.Import;
using GerbilManagerWebAPI.Models;
using Microsoft.EntityFrameworkCore;
using Microsoft.Extensions.Configuration;
using System.Text.Json;
namespace GerbilManager.Tests
{
public class IngestResolvedServiceTests : IDisposable
{
private readonly string _dir;
private readonly string _resolvedJsonPath;
public IngestResolvedServiceTests()
{
_dir = Path.Combine(Path.GetTempPath(), "ingest-resolved-" + Guid.NewGuid().ToString("N"));
Directory.CreateDirectory(_dir);
_resolvedJsonPath = Path.Combine(_dir, "resolved_import.json");
var contactId = Guid.NewGuid();
var fatherId = Guid.NewGuid();
var motherId = Guid.NewGuid();
var litterId = Guid.NewGuid();
var photoId = Guid.NewGuid();
var data = new
{
Contacts = new[]
{
new
{
Id = contactId,
Name = "Test Breeder",
ZuchtName = "Test Zucht",
City = "Test City",
Email = "test@example.com",
Homepage = "",
Phone = "",
Address = "",
Provenance = (string?)"{\"sourceFiles\":[\"Stammbaum\"],\"mergedRecordCount\":1,\"notes\":[\"als Züchter erkannt\"]}"
}
},
Litters = new[]
{
new
{
Id = litterId,
Name = "Wurf A",
Date = "2026-01-01",
TotalBorn = 5,
DeathsWithin8Weeks = 0,
FatherId = fatherId,
MotherId = motherId,
ExpectedGoHomeDate = (string?)null,
Notes = "Test litter notes",
PairingCode = "PC01",
ExternalRef = "ext-litter-1",
LitterLetter = "A",
Provenance = (string?)"{\"sourceFiles\":[\"Wurfchronik-Detail.docx\"],\"mergedRecordCount\":1,\"fromWurfchronik\":true,\"notes\":[\"aus Wurfchronik\"]}"
}
},
Gerbils = new[]
{
new
{
Id = fatherId,
Name = "Papa",
Gender = "male",
Status = "Breeding",
LitterId = (Guid?)null,
OriginContactId = contactId,
ReceiverContactId = (Guid?)null,
EnclosureId = (Guid?)null,
ColorVarietyId = new Guid("00000000-0000-0000-0000-000000000006"), // Schwarz
DateOfBirth = "2025-01-01",
DateOfDeath = (string?)null,
CauseOfDeath = (string?)null,
GoHomeDate = (string?)null,
Genotype = "aa CC DD EE GG PP spsp rere",
Notes = "",
ImportSource = "docx-export",
ExternalRef = "ext-papa",
RawImportData = "{}",
OriginBreeder = "Test Zucht",
NameSearch = "papa",
CharacterTraits = new string[] {},
CharacterNote = (string?)null,
IsDeaf = false,
IsResident = true,
Provenance = (string?)"{\"sourceFiles\":[\"Stammbaum von Papa.xlsx\",\"Wurfchronik-Detail.docx\"],\"mergedRecordCount\":2,\"fromWurfchronik\":true,\"notes\":[\"aus 2 Datensätzen zusammengeführt\"],\"history\":[\"In \\u201eStammbaum von Papa.xlsx\\u201c gefunden.\",\"Genotyp aus \\u201eWurfchronik-Detail.docx\\u201c.\"]}"
},
new
{
Id = motherId,
Name = "Mama",
Gender = "female",
Status = "Breeding",
LitterId = (Guid?)null,
OriginContactId = contactId,
ReceiverContactId = (Guid?)null,
EnclosureId = (Guid?)null,
ColorVarietyId = new Guid("00000000-0000-0000-0000-000000000006"), // Schwarz
DateOfBirth = "2025-01-01",
DateOfDeath = (string?)null,
CauseOfDeath = (string?)null,
GoHomeDate = (string?)null,
Genotype = "aa CC DD EE GG PP spsp rere",
Notes = "",
ImportSource = "docx-export",
ExternalRef = "ext-mama",
RawImportData = "{}",
OriginBreeder = "Test Zucht",
NameSearch = "mama",
CharacterTraits = new string[] {},
CharacterNote = (string?)null,
IsDeaf = false,
IsResident = true,
Provenance = (string?)null
}
},
GerbilPhotos = new[]
{
new
{
Id = photoId,
GerbilId = fatherId,
FileName = "photo1.jpg",
SortOrder = 0
}
}
};
File.WriteAllText(_resolvedJsonPath, JsonSerializer.Serialize(data));
}
public void Dispose()
{
try { Directory.Delete(_dir, recursive: true); } catch { }
}
private ApplicationContext NewDb()
{
var opts = new DbContextOptionsBuilder<ApplicationContext>()
.UseInMemoryDatabase("ingest-resolved-" + Guid.NewGuid().ToString("N"))
.Options;
var db = new ApplicationContext(opts);
db.Database.EnsureCreated();
return db;
}
[Fact]
public async Task IngestResolved_loads_data_and_resolves_all_relations()
{
using var db = NewDb();
var config = new ConfigurationBuilder()
.AddInMemoryCollection(new Dictionary<string, string?>
{
{ "Import:SourcePath", _dir }
})
.Build();
var service = new IngestResolvedService(db, config, null!);
var result = await service.RunAsync();
Assert.Contains("Ingestion successful!", result);
// Verify counts
Assert.Equal(1, await db.Contacts.CountAsync());
Assert.Equal(1, await db.Litters.CountAsync());
Assert.Equal(2, await db.Gerbils.CountAsync());
Assert.Equal(1, await db.GerbilPhotos.CountAsync());
// Verify relations
var litter = await db.Litters.SingleAsync();
var father = await db.Gerbils.SingleAsync(g => g.Name == "Papa");
var mother = await db.Gerbils.SingleAsync(g => g.Name == "Mama");
var photo = await db.GerbilPhotos.SingleAsync();
Assert.Equal(father.Id, litter.FatherId);
Assert.Equal(mother.Id, litter.MotherId);
Assert.Equal(father.Id, photo.GerbilId);
Assert.Equal(father.OriginContactId, mother.OriginContactId);
// Provenance round-trips verbatim through ingest (and stays null when absent).
Assert.NotNull(father.Provenance);
Assert.Contains("Stammbaum von Papa.xlsx", father.Provenance);
Assert.Contains("mergedRecordCount", father.Provenance);
// The chronological, file-attributed history survives ingest verbatim
// (stored as opaque JSON on the text column — no schema for it).
Assert.Contains("history", father.Provenance);
Assert.Contains("Genotyp aus", father.Provenance);
Assert.Null(mother.Provenance);
// Contact + litter provenance also round-trips through the ingest.
var contact = await db.Contacts.SingleAsync();
Assert.NotNull(contact.Provenance);
Assert.Contains("als Züchter erkannt", contact.Provenance);
Assert.NotNull(litter.Provenance);
Assert.Contains("aus Wurfchronik", litter.Provenance);
Assert.Contains("fromWurfchronik", litter.Provenance);
}
}
}