public void ExtractionInformationTriangle()
{
var t = new TableInfo(CatalogueRepository, "t");
var col = new ColumnInfo(CatalogueRepository, "mycol", "varchar(10)", t);
var cat = new Catalogue(CatalogueRepository, "MyCat");
var ci = new CatalogueItem(CatalogueRepository, cat, "myci");
try
{
//col depends on tr
Assert.Contains(t, col.GetObjectsThisDependsOn());
Assert.Contains(col, t.GetObjectsDependingOnThis());
//catalogue depends on catalogue items existing (slightly counter intuitive but think of it as data flow out of technical low level data through transforms into datasets - and then into researchers and research projects)
Assert.Contains(cat, ci.GetObjectsDependingOnThis());
Assert.Contains(ci, cat.GetObjectsThisDependsOn());
//catalogue item should not be relying on anything currently (no link to underlying technical data)
Assert.IsNull(ci.GetObjectsThisDependsOn());
//now they are associated so the ci should be dependent on the col
ci.SetColumnInfo(col);
Assert.Contains(col, ci.GetObjectsDependingOnThis());
}
finally
{
t.DeleteInDatabase();
cat.DeleteInDatabase();
}
}