From f2da062c49a92ece91efc5ff7bca823a528879c3 Mon Sep 17 00:00:00 2001
From: Rob <rob.tranquillo@gmx.de>
Date: Tue, 21 Apr 2026 15:28:13 +0200
Subject: [PATCH] Add master data parser with German decimal conversion

---
 .github/copilot-tasks.md          | 10 ++--
 src/taetigkeitsnachweis/parser.py | 93 ++++++++++++++++++++++++++++++-
 2 files changed, 97 insertions(+), 6 deletions(-)

diff --git a/.github/copilot-tasks.md b/.github/copilot-tasks.md
index cf19780..fa502a1 100644
--- a/.github/copilot-tasks.md
+++ b/.github/copilot-tasks.md
@@ -36,10 +36,10 @@ Projekt vollstaendig umsetzen: aus PS_PW.xlsx und Stellenbeschreibungen.xlsx mon
 	- [x] Mitarbeiterkontext ueber Mehrzeilen-Eintraege vererben
 	- [x] Werte normalisieren (`-`, `0`, leer, numerisch)
 	- [x] Normalisierte Records erzeugen
-- [ ] T6: Parser fuer Stellenbeschreibungen.xlsx implementieren
-	- [ ] Felder ID, Antragsnummer, Name, Qualifikation, Stellenanteil, Funktion, Taetigkeit einlesen
-	- [ ] Prozentwerte (deutsches Dezimalformat) robust in Zahlen umwandeln
-	- [ ] Mehrfache Eintraege je Name korrekt behalten
+- [x] T6: Parser fuer Stellenbeschreibungen.xlsx implementieren
+	- [x] Felder ID, Antragsnummer, Name, Qualifikation, Stellenanteil, Funktion, Taetigkeit einlesen
+	- [x] Prozentwerte (deutsches Dezimalformat) robust in Zahlen umwandeln
+	- [x] Mehrfache Eintraege je Name korrekt behalten
 
 ## Phase 4 - Matching und Fachlogik
 - [ ] T7: Foerderquote berechnen
@@ -111,4 +111,4 @@ Projekt vollstaendig umsetzen: aus PS_PW.xlsx und Stellenbeschreibungen.xlsx mon
 8. T15 -> T16 -> T17
 
 ## Aktueller Startpunkt
-- [ ] Naechste aktive Aufgabe: T6 (Parser fuer Stellenbeschreibungen.xlsx implementieren)
+- [ ] Naechste aktive Aufgabe: T7 (Foerderquote berechnen)
diff --git a/src/taetigkeitsnachweis/parser.py b/src/taetigkeitsnachweis/parser.py
index dfd8b32..45cc403 100644
--- a/src/taetigkeitsnachweis/parser.py
+++ b/src/taetigkeitsnachweis/parser.py
@@ -10,7 +10,7 @@ from openpyxl import load_workbook
 from openpyxl.utils.cell import column_index_from_string
 
 from .errors import ValidationError
-from .models import AllocationRecord
+from .models import AllocationRecord, MasterDataRecord
 
 
 def _to_text(value: Any) -> str:
@@ -30,6 +30,27 @@ def _to_float(value: Any, field_name: str) -> float:
 		raise ValidationError(f"Invalid numeric value for {field_name}: {value}") from exc
 
 
+def _parse_german_decimal(
+	value: Any,
+	field_name: str,
+	decimal_separator: str,
+	thousands_separator: str,
+) -> float:
+	if value is None or str(value).strip() == "":
+		raise ValidationError(f"Missing numeric value for {field_name}.")
+
+	text = str(value).strip()
+	if thousands_separator:
+		text = text.replace(thousands_separator, "")
+	if decimal_separator:
+		text = text.replace(decimal_separator, ".")
+
+	try:
+		return float(text)
+	except ValueError as exc:
+		raise ValidationError(f"Invalid numeric value for {field_name}: {value}") from exc
+
+
 def _parse_month_marker(value: Any, prefix: str, date_format: str) -> datetime | None:
 	marker = _to_text(value)
 	if not marker:
@@ -182,3 +203,73 @@ def parse_allocation_records(
 
 	workbook.close()
 	return normalized_records
+
+
+def parse_masterdata_records(
+	config: dict[str, Any],
+	project_root: Path,
+) -> list[MasterDataRecord]:
+	files_cfg = config["files"]
+	sheets_cfg = config["sheets"]
+	paths_cfg = config["paths"]
+	master_cfg = config["masterdata"]
+	columns_cfg = master_cfg["columns"]
+	decimal_cfg = master_cfg["decimal"]
+
+	workbook_path = (
+		project_root
+		/ paths_cfg["input_dir"]
+		/ files_cfg["masterdata_workbook"]
+	)
+
+	if not workbook_path.exists():
+		raise ValidationError(f"Master data workbook not found: {workbook_path}")
+
+	workbook = load_workbook(workbook_path, data_only=True)
+	sheet_name = sheets_cfg["masterdata"]
+	if sheet_name not in workbook.sheetnames:
+		raise ValidationError(f"Sheet '{sheet_name}' not found in {workbook_path.name}")
+	worksheet = workbook[sheet_name]
+
+	id_col = column_index_from_string(columns_cfg["id"])
+	antragsnummer_col = column_index_from_string(columns_cfg["antragsnummer"])
+	name_col = column_index_from_string(columns_cfg["name"])
+	qualifikation_col = column_index_from_string(columns_cfg["qualifikation"])
+	stellenanteil_col = column_index_from_string(columns_cfg["stellenanteil"])
+	funktion_col = column_index_from_string(columns_cfg["funktion"])
+	taetigkeit_col = column_index_from_string(columns_cfg["taetigkeit"])
+
+	records: list[MasterDataRecord] = []
+
+	for row_index in range(2, worksheet.max_row + 1):
+		name = _to_text(worksheet.cell(row=row_index, column=name_col).value)
+		if not name:
+			continue
+
+		stellenanteil = _parse_german_decimal(
+			worksheet.cell(row=row_index, column=stellenanteil_col).value,
+			field_name=f"stellenanteil row {row_index}",
+			decimal_separator=decimal_cfg.get("decimal_separator", ","),
+			thousands_separator=decimal_cfg.get("thousands_separator", "."),
+		)
+
+		records.append(
+			MasterDataRecord(
+				record_id=_to_text(worksheet.cell(row=row_index, column=id_col).value),
+				antragsnummer=_to_text(
+					worksheet.cell(row=row_index, column=antragsnummer_col).value
+				),
+				name=name,
+				qualifikation=_to_text(
+					worksheet.cell(row=row_index, column=qualifikation_col).value
+				),
+				stellenanteil_percent=stellenanteil,
+				funktion=_to_text(worksheet.cell(row=row_index, column=funktion_col).value),
+				taetigkeit=_to_text(
+					worksheet.cell(row=row_index, column=taetigkeit_col).value
+				),
+			)
+		)
+
+	workbook.close()
+	return records