saving

tests/unit/test_manual.py

@@ -0,0 +1,110 @@
+"""Tests for `openrun.ingest.manual` — CSV import + idempotency."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import pytest
+
+from openrun.ingest.manual import import_csv
+
+
+def _write_csv(tmp_path: Path, text: str) -> Path:
+    p = tmp_path / "manual.csv"
+    p.write_text(text)
+    return p
+
+
+def test_import_csv_inserts_required_columns(tmp_conn, tmp_path: Path) -> None:
+    csv = _write_csv(tmp_path, """\
+activity_date,activity_type,distance_km,duration_min,training_load,notes
+2026-05-15,strength,,45,30,upper body
+2026-05-16,hike,8.0,120,40,morning hike
+""")
+    result = import_csv(tmp_conn, csv)
+    assert result.inserted == 2
+    assert result.updated == 0
+    assert result.skipped == 0
+    assert result.errors == []
+
+    rows = tmp_conn.execute(
+        "SELECT activity_date, activity_type, distance_km, duration_min, training_load, notes "
+        "FROM manual_activities ORDER BY activity_date"
+    ).fetchall()
+    assert rows[0]["activity_date"] == "2026-05-15"
+    assert rows[0]["activity_type"] == "strength"
+    assert rows[0]["distance_km"] is None
+    assert rows[0]["duration_min"] == 45
+    assert rows[0]["training_load"] == 30
+    assert rows[0]["notes"] == "upper body"
+    assert rows[1]["distance_km"] == 8.0
+
+
+def test_import_csv_requires_header_columns(tmp_conn, tmp_path: Path) -> None:
+    csv = _write_csv(tmp_path, "date,type\n2026-05-15,strength\n")  # wrong column names
+    with pytest.raises(ValueError, match="missing required columns"):
+        import_csv(tmp_conn, csv)
+
+
+def test_import_csv_external_id_makes_reimport_idempotent(tmp_conn, tmp_path: Path) -> None:
+    csv = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load,external_id
+2026-05-15,strength,30,strava-12345
+""")
+    import_csv(tmp_conn, csv)
+    second = import_csv(tmp_conn, csv)
+    assert second.inserted == 0
+    assert second.updated == 1
+    count = tmp_conn.execute("SELECT COUNT(*) FROM manual_activities").fetchone()[0]
+    assert count == 1
+
+
+def test_import_csv_reimport_with_new_tl_updates_value(tmp_conn, tmp_path: Path) -> None:
+    """External_id upsert should reflect changed training_load on re-import."""
+    csv1 = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load,external_id
+2026-05-15,strength,30,k1
+""")
+    import_csv(tmp_conn, csv1)
+    csv2 = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load,external_id
+2026-05-15,strength,55,k1
+""")
+    import_csv(tmp_conn, csv2)
+    tl = tmp_conn.execute("SELECT training_load FROM manual_activities WHERE external_id='k1'").fetchone()[0]
+    assert tl == 55
+
+
+def test_import_csv_blank_external_id_allows_duplicates(tmp_conn, tmp_path: Path) -> None:
+    """No external_id = caller is intentionally logging without dedupe key. Two imports = two rows."""
+    csv = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load
+2026-05-15,strength,30
+""")
+    import_csv(tmp_conn, csv)
+    import_csv(tmp_conn, csv)
+    assert tmp_conn.execute("SELECT COUNT(*) FROM manual_activities").fetchone()[0] == 2
+
+
+def test_import_csv_skips_bad_rows_but_keeps_going(tmp_conn, tmp_path: Path) -> None:
+    csv = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load
+2026-05-15,strength,30
+bad-date,hike,40
+2026-05-17,run,
+""")
+    result = import_csv(tmp_conn, csv)
+    assert result.inserted == 2
+    assert result.skipped == 1
+    assert len(result.errors) == 1
+    assert "line 3" in result.errors[0]
+
+
+def test_import_csv_accepts_iso_datetime_in_date_column(tmp_conn, tmp_path: Path) -> None:
+    csv = _write_csv(tmp_path, """\
+activity_date,activity_type,training_load
+2026-05-15T08:30:00,strength,30
+""")
+    import_csv(tmp_conn, csv)
+    d = tmp_conn.execute("SELECT activity_date FROM manual_activities").fetchone()[0]
+    assert d == "2026-05-15"