Merge pull request #148 from Datenschule/update-mv-2

This change updates the data to the latest freely available data and prevents numbers in ids and zips from being parsed to floats before being stringified.

Fixes #147

Author: k-nut

jedeschule/spiders/mecklenburg_vorpommern.py

@@ -6,6 +6,13 @@
 from jedeschule.spiders.school_spider import SchoolSpider
 
 
+def as_string(value: str):
+    try:
+        return str(int(value))
+    except ValueError:
+        return value
+
+
 class MecklenburgVorpommernSpider(SchoolSpider):
     name = "mecklenburg-vorpommern"
     # The state provides the data as an Excel file. The current year's
@@ -15,7 +22,7 @@ class MecklenburgVorpommernSpider(SchoolSpider):
     #   https://www.statistischebibliothek.de/mir/receive/MVSerie_mods_00000396
     # Official documentation on all available data here:
     #   https://www.laiv-mv.de/Statistik/Veröffentlichungen/Verzeichnisse/
-    base_url = "https://www.statistischebibliothek.de/mir/servlets/MCRFileNodeServlet/MVHeft_derivate_00006849/V034%202022%2000.xlsx"
+    base_url = "https://www.statistischebibliothek.de/mir/servlets/MCRFileNodeServlet/MVHeft_derivate_00007470/V044%202023%2000.xlsx"
     start_urls = [base_url]
 
     def parse(self, response):
@@ -32,10 +39,10 @@ def parse(self, response):
     def normalize(item: Item) -> School:
         return School(
             name=item.get("NAME1"),
-            id="MV-{}".format(item.get("DIENSTSTELLEN-NUMMER")),
+            id="MV-{}".format(as_string(item.get("DIENSTSTELLEN-NUMMER"))),
             address=item.get("STRASSE"),
             address2="",
-            zip=item.get("PLZ"),
+            zip=as_string(item.get("PLZ")).zfill(5),
             city=item.get("ORT"),
             website=item.get("INTERNET"),
             email=item.get("E-MAIL-ADRESSE"),