Add support for avoiding shifts on specific dates with reasons

leave.py

@@ -1,5 +1,5 @@
 import csv
-from datetime import datetime
+from datetime import datetime, timedelta
 import re
 import sys
 from requests import Session
@@ -264,4 +264,92 @@ def load_academy(Rota):
     """
     Some trainees spend month blocks in the academy. This function loads that information from a google sheet that published it as a csv file.
     """
-    url = "https://docs.google.com/spreadsheets/d/e/2PACX-1vQtNrp9F1fjtAh8vFX-W_R3RlD9H-2P_vCLMoHrVKR0e24yabbsdrD65VjwxoHlV1Qnatmw1NeghQHG/pub?gid=1113837782&single=true&output=csv"
+    url = "https://docs.google.com/spreadsheets/d/e/2PACX-1vQtNrp9F1fjtAh8vFX-W_R3RlD9H-2P_vCLMoHrVKR0e24yabbsdrD65VjwxoHlV1Qnatmw1NeghQHG/pub?gid=1113837782&single=true&output=csv"
+
+    with Session() as s:
+        headers = {"Cache-Control": "no-cache", "Pragma": "no-cache"}
+        download = s.get(url, headers=headers)
+        decoded = download.content.decode("utf-8")
+
+        reader = csv.reader(decoded.splitlines(), delimiter=',')
+
+        n = 0
+        names = []
+        # collect date rows (these are the month start boundaries) and their cells
+        date_rows: list[datetime.date] = []
+        date_row_cells: list[list[str]] = []
+
+        for row in reader:
+            n += 1
+            if n == 1:
+                # header row: first cell is column title, remaining are worker names
+                names = [c.strip() for c in row[1:]]
+                continue
+
+            if not row:
+                continue
+
+            row_title = row[0].strip()
+            cells = row[1:]
+
+            # If this row is a date row like '02/03/2026' then treat it as a month-start boundary
+            if re.match(date_re, row_title):
+                try:
+                    try:
+                        d = datetime.strptime(row_title, "%d/%m/%Y").date()
+                    except ValueError:
+                        d = datetime.strptime(row_title, "%d/%m/%y").date()
+                except Exception:
+                    continue
+
+                date_rows.append(d)
+                # normalize cells length to match names
+                row_cells = [ (cells[i] if i < len(cells) else "").strip() for i in range(len(names)) ]
+                date_row_cells.append(row_cells)
+
+        # If no date rows found, nothing to do
+        if not date_rows:
+            return
+
+        # Identify twilight shift names available in the rota
+        twilight_shift_names = [s.name for s in Rota.get_shifts() if "twilight" in s.name]
+
+        # For each date boundary, compute its end (next boundary - 1 day) and register avoids where cell=='academy'
+        for idx, start_date in enumerate(date_rows):
+            if idx + 1 < len(date_rows):
+                end_date = date_rows[idx + 1] - timedelta(days=1)
+            else:
+                end_date = Rota.rota_end_date
+
+            cells = date_row_cells[idx]
+            for i, cell in enumerate(cells):
+                if i >= len(names):
+                    break
+                v = (cell or "").lower()
+                if "academy" in v:
+                    worker_name = names[i]
+                    matching_workers = [w for w in Rota.workers if w.name.strip() == worker_name.strip()]
+                    if not matching_workers:
+                        logger.warning(f"Academy data: worker '{worker_name}' not found in rota, skipping")
+                        continue
+                    worker = matching_workers[0]
+
+                    # choose twilight shifts relevant to the worker's site
+                    candidate_shifts = []
+                    for sh_name in twilight_shift_names:
+                        try:
+                            sh = Rota.get_shift_by_name(sh_name)
+                        except Exception:
+                            continue
+                        if hasattr(worker, "site") and worker.site in sh.sites:
+                            candidate_shifts.append(sh_name)
+                        elif sh_name.startswith(f"{worker.site}_"):
+                            candidate_shifts.append(sh_name)
+
+                    if not candidate_shifts:
+                        candidate_shifts = twilight_shift_names.copy()
+
+                    # register as a date range with a reason 'academy'
+                    Rota.add_avoid_shifts_for_workers_on_dates(
+                        names=[worker.name], shifts=candidate_shifts, start_date=start_date, end_date=end_date, reason="academy"
+                    )