"""Listeneintrag-Versionen, Papierkorb-Metadaten, Seiten mit canvasLayout, Inhaltstypen der Listen, Site-Pages-Liste und Strukturprüfung (Stichprobe von 5 Sites, live gegen Backup). Nur GET. Ausgabe: graph/sharepoint//{versions/.json, recycle_bin.json, pages_canvas.json, list_contenttypes.json, termstore.json}, graph/structure_check.json und graph/sharepoint_extended_status.json.""" import glob, json, os, random, re, sys, threading from concurrent.futures import ThreadPoolExecutor sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) import m365_auth as auth def sanitize(s): return re.sub(r'[^A-Za-z0-9._-]', "_", s)[:80] def main(argv=None): p = auth.make_parser("List item versions, recycle bin, canvas pages and structure check (read-only)") p.add_argument("--max-items", type=int, default=1000, help="items per list (default 1000)") p.add_argument("--max-sites", type=int, default=0, help="limit the number of sites (0 = all)") args = p.parse_args(argv) out, max_items = args.out, args.max_items site_root = os.path.join(out, "graph", "sharepoint") lock = threading.Lock() st = {"sites": 0, "lists": 0, "items_with_versions": 0, "versions": 0, "items_skipped_limit": 0, "recycle_bin_entries": 0, "canvas_pages": 0, "attachment_items": 0, "errors": []} def site_work(f): j = json.load(open(f)); site = j["site"]; sid = site["id"] stem = os.path.basename(f)[:-5]; d = os.path.join(site_root, stem) os.makedirs(d + "/versions", exist_ok=True) # Papierkorb (nur Metadaten) bin_items, e = auth.get_all(f"/sites/{sid}/recycleBin/items", True) auth.write_json(d + "/recycle_bin.json", bin_items if e is None else {"error": e}) with lock: if e: st["errors"].append((stem, "recycle_bin", e)) else: st["recycle_bin_entries"] += len(bin_items) # Listen: Backup-Listen plus Site Pages lists = [(x["id"], x.get("displayName")) for x in (j["lists"] if isinstance(j["lists"], list) else [])] r = auth.g("GET", f"/sites/{sid}/lists/Site%20Pages?$select=id,displayName") if r.status_code == 200 and r.json()["id"] not in [x[0] for x in lists]: lists.append((r.json()["id"], r.json().get("displayName", "Site Pages"))) content_types = {} for lid, lname in lists: c, e = auth.get_all(f"/sites/{sid}/lists/{lid}/contentTypes") content_types[lname] = c if e is None else {"error": e} items, e = auth.get_all(f"/sites/{sid}/lists/{lid}/items?$top=999&$expand=fields") if e: with lock: st["errors"].append((stem, lname, "items " + e)) continue with lock: st["lists"] += 1 st["attachment_items"] += sum(1 for i in items if (i.get("fields") or {}).get("Attachments") is True) res = [] for i in items[:max_items]: v, e = auth.get_all(f"/sites/{sid}/lists/{lid}/items/{i['id']}/versions") if e: with lock: st["errors"].append((stem, lname, f"item {i['id']} versions {e}")) continue res.append({"itemId": i["id"], "versions": v}) with lock: st["items_with_versions"] += 1; st["versions"] += len(v) if len(items) > max_items: with lock: st["items_skipped_limit"] += len(items) - max_items auth.write_json(f"{d}/versions/{sanitize(lname)}-{lid[:8]}.json", {"list": lname, "listId": lid, "items_total": len(items), "items_exported": len(res), "items": res}) auth.write_json(d + "/list_contenttypes.json", content_types) # Seiten mit canvasLayout pages = j["pages"] if isinstance(j["pages"], list) else [] canvas = [] for pg in pages: r = auth.g("GET", f"https://graph.microsoft.com/beta/sites/{sid}/pages/{pg['id']}/microsoft.graph.sitePage?$expand=canvasLayout") if r.status_code == 200: canvas.append(r.json()) else: with lock: st["errors"].append((stem, "page " + pg.get("name", ""), f"canvas {r.status_code}")) auth.write_json(d + "/pages_canvas.json", canvas) with lock: st["canvas_pages"] += len(canvas); st["sites"] += 1; print(st["sites"], stem, flush=True) # Termspeicher (nur Versuch) r = auth.g("GET", f"/sites/{sid}/termStore/sets") if r.status_code == 200: auth.write_json(d + "/termstore.json", r.json()) return stem, r.status_code files = sorted(glob.glob(site_root + "/*.json")) if args.max_sites: files = files[:args.max_sites] with ThreadPoolExecutor(3) as ex: results = list(ex.map(site_work, files)) st["termstore_status"] = {} for _, c in results: st["termstore_status"][c] = st["termstore_status"].get(c, 0) + 1 # Strukturprüfung: Stichprobe live gegen Backup random.seed(5) sample = random.sample(files, min(5, len(files))); check = [] for f in sample: j = json.load(open(f)); sid = j["site"]["id"]; entry = {"site": j["site"].get("webUrl")} for key, path, beta in [("lists", f"/sites/{sid}/lists", False), ("contentTypes", f"/sites/{sid}/contentTypes", False), ("columns", f"/sites/{sid}/columns", False), ("pages", f"/sites/{sid}/pages", True), ("drives", f"/sites/{sid}/drives", False)]: live, er = auth.get_all(path, beta); bk = j[key] entry[key] = {"live": None if er else len(live), "backup": len(bk) if isinstance(bk, list) else bk, "error": er} per_list = {} for lst in (j["lists"] if isinstance(j["lists"], list) else []): if lst.get("list", {}).get("hidden"): continue live, er = auth.get_all(f"/sites/{sid}/lists/{lst['id']}/items?$top=999&$expand=fields") bk = j["items"].get(lst["displayName"]) with_fields = sum(1 for i in (bk or []) if isinstance(i, dict) and i.get("fields")) if isinstance(bk, list) else None per_list[lst["displayName"]] = {"live": None if er else len(live), "backup": len(bk) if isinstance(bk, list) else bk, "backup_with_fields": with_fields, "columns_in_list": len(lst.get("columns", []))} entry["list_items"] = per_list cvf = os.path.join(site_root, os.path.basename(f)[:-5], "pages_canvas.json") entry["canvas_pages_in_backup"] = len(json.load(open(cvf))); check.append(entry) auth.write_json(os.path.join(out, "graph", "structure_check.json"), check) auth.write_json(os.path.join(out, "graph", "sharepoint_extended_status.json"), st) print("DONE", {k: v for k, v in st.items() if k != "errors"}, len(st["errors"]), "errors") return 0 if __name__ == "__main__": sys.exit(main())