#!/usr/bin/env python3 from pathlib import Path import sys root = Path(__file__).resolve().parents[2] errors = [] required_files = [ root / 'index.html', root / 'index-en.html', root / 'shared/archive-core/archive-app.js', root / 'shared/archive-core/archive-styles.css', root / 'shared/map/cdv-map.js', root / 'shared/map/map.css', ] for p in required_files: if not p.exists(): errors.append(f'FEHLT: {p.relative_to(root)}') # Workflow-Anweisungen muessen Teil des Projekts sein und online hochladbar bleiben. workflow_required = [ root / 'workflow/index.html', root / 'workflow/MASTER-INSTRUCTIONS.md', root / 'workflow/PROJECT-INSTRUCTIONS.md', root / 'workflow/ai-instructions.json', root / 'workflow/IMPORT-PROMPT-TEMPLATE.md', ] for p in workflow_required: if not p.exists(): errors.append(f'FEHLT WORKFLOW-DATEI: {p.relative_to(root)}') # Root-Seiten sollen den Workflow diskret verlinken, damit er online auffindbar ist. for name in ['index.html', 'index-en.html']: p = root / name if p.exists(): s = p.read_text(errors='ignore') if 'workflow/index.html' not in s and '/workflow/' not in s: errors.append(f'FEHLT WORKFLOW-LINK: {name}') # Workflow-Regeln: MASTER-INSTRUCTIONS.md ist die einzige verbindliche Quelle. master = root / 'workflow/MASTER-INSTRUCTIONS.md' if master.exists(): mt = master.read_text(encoding='utf-8', errors='ignore') for marker in [ 'einzige verbindliche Arbeitsanweisung', 'TEMP-store', 'Vorderseite', 'Rückseite', 'captionEn', 'Open original', 'action.php?download=', 'kein Speichern-Dialog', 'Zählcheck', 'vsN-historicbits-clean.zip' ]: if marker not in mt: errors.append(f'MASTER-ANWEISUNG UNVOLLSTAENDIG: MASTER-INSTRUCTIONS.md ohne {marker!r}') # Andere Workflow-Dateien duerfen nur auf die Master-Anweisung verweisen und keine konkurrierenden Langregeln bilden. for rel in ['workflow/PROJECT-INSTRUCTIONS.md', 'workflow/IMPORT-PROMPT-TEMPLATE.md']: q = root / rel if q.exists(): qt = q.read_text(encoding='utf-8', errors='ignore') if 'MASTER-INSTRUCTIONS.md' not in qt: errors.append(f'WORKFLOW-DATEI VERWEIST NICHT AUF MASTER: {rel}') ai_json = root / 'workflow/ai-instructions.json' if ai_json.exists(): t = ai_json.read_text(encoding='utf-8', errors='ignore') for marker in ['MASTER-INSTRUCTIONS.md', 'captionEn', 'reverse_logic', 'theme_folder_forbidden_files', 'vsN-historicbits-clean.zip', 'open_original_rule']: if marker not in t: errors.append(f'WORKFLOW-REGEL UNVOLLSTAENDIG: ai-instructions.json ohne {marker}') # Keine versehentlichen XML-Index-Dateien als Ersatz für HTML. for p in root.rglob('index.xml'): errors.append(f'NICHT ERLAUBT: {p.relative_to(root)} — Themenordner brauchen index.html') # Root-Startseite darf nicht auf reine Ordner-URLs zeigen, weil file:// lokal dann Directory Listing anzeigen kann. for name in ['index.html', 'index-en.html']: p = root / name if p.exists(): s = p.read_text(errors='ignore') if 'href="archive/"' in s or "href='archive/'" in s: errors.append(f'RISKANTER ORDNER-LINK: {name} benutzt archive/ statt archive/index.html') if 'archive/?year=' in s: errors.append(f'RISKANTER TIMELINE-LINK: {name} benutzt archive/?year= statt archive/index.html?year=') for name in ['app.js', 'app-en.js']: p = root / name if p.exists(): s = p.read_text(errors='ignore') if 'archive/?year=' in s: errors.append(f'RISKANTER TIMELINE-LINK: {name} benutzt archive/?year= statt archive/index.html?year=') if 'LIVE_DATA_PAGE = "archive/"' in s: errors.append(f'RISKANTER LIVE_DATA_PAGE: {name} benutzt archive/ statt archive/index.html') skip_dirs = {'assets', 'shared', 'workflow', '__MACOSX'} for d in root.iterdir(): if not d.is_dir() or d.name.startswith('.') or d.name in skip_dirs: continue # Themen-/App-Ordner brauchen index.html, wenn sie Daten oder Karten enthalten. looks_like_topic = any((d / x).exists() for x in ['cdv-data-en.js', 'theme-config.js', 'cdv-map-en.html']) or d.name.startswith('_template') if looks_like_topic and not (d / 'index.html').exists(): errors.append(f'FEHLT: {d.name}/index.html') continue if (d / 'index.html').exists() and looks_like_topic: if (d / 'cdv-app.js').exists(): errors.append(f'NICHT ERLAUBT: {d.name}/cdv-app.js') if (d / 'styles.css').exists(): errors.append(f'NICHT ERLAUBT: {d.name}/styles.css') s = (d / 'index.html').read_text(errors='ignore') if 'shared/archive-core/archive-app.js' not in s: errors.append(f'FEHLT SHARED-APP-LINK: {d.name}/index.html') if 'shared/archive-core/archive-styles.css' not in s: errors.append(f'FEHLT SHARED-CSS-LINK: {d.name}/index.html') # Open original must open a high-res browser view when possible, but never trigger direct downloads. app_file = root / 'shared/archive-core/archive-app.js' if app_file.exists(): app_text = app_file.read_text(encoding='utf-8', errors='ignore') if 'function originalOpenUrl(item)' not in app_text: errors.append('OPEN-ORIGINAL-REGEL FEHLT: archive-app.js ohne originalOpenUrl(item)') if 'function browserOpenImageUrl(url' not in app_text: errors.append('OPEN-ORIGINAL-REGEL FEHLT: archive-app.js ohne browserOpenImageUrl(url)') if 'download parameter' not in app_text and 'download flag' not in app_text: errors.append('OPEN-ORIGINAL-REGEL UNVOLLSTAENDIG: download-Parameter wird nicht sichtbar behandelt') if 'href="${esc(highResImageUrl(item.image, item.pageUrl))}"' in app_text: errors.append('OPEN-ORIGINAL FEHLER: sichtbarer Link benutzt highResImageUrl(item.image, item.pageUrl)') if 'href="${esc(highResImageUrl(item.reverse.image, item.reverse.pageUrl))}"' in app_text: errors.append('OPEN-ORIGINAL FEHLER: Rueckseiten-Link benutzt highResImageUrl(reverse.image, reverse.pageUrl)') master_instructions = root / 'workflow/MASTER-INSTRUCTIONS.md' if master_instructions.exists(): instr = master_instructions.read_text(encoding='utf-8', errors='ignore') for marker in ['Open original', 'High-Res', 'action.php?download=', 'kein Speichern-Dialog']: if marker not in instr: errors.append(f'MASTER-REGEL UNVOLLSTAENDIG: MASTER-INSTRUCTIONS.md ohne Open-original-Marker {marker!r}') # Datenprüfung für Bückeburger Import: englische Detailansicht darf keine gemischten deutschen captionEn-Texte enthalten. import json, re def _load_timeline_items(path): text = path.read_text(encoding='utf-8', errors='ignore') m = re.search(r'window\.timelineItems\s*=\s*(\[.*\]);?\s*$', text, re.S) if not m: return [] return json.loads(m.group(1)) german_caption_markers = [ ' tätig ', ' aufgenommen ', ' um 1880', ' war ', ' Seine ', ' handkolorierten ', 'typischer Flügelhaube', 'Fürstenhauses', 'Alltags- und Festkultur', 'um die Jahrhundertwende' ] for rel in ['bueckeburger-trachten/cdv-data-en.js', 'archive/cdv-data-en.js']: p = root / rel if p.exists(): for item in _load_timeline_items(p): if item.get('collection') == 'Bückeburger Trachten' or '/category/48' in str(item.get('pageUrl', '')): caption_en = str(item.get('captionEn', '')) bad = [marker for marker in german_caption_markers if marker in caption_en] if bad: errors.append(f'DE/EN-MIX IN captionEn: {rel} ID {item.get("id")} enthält {bad[0]!r}') # vs16 hard check: new Piwigo imports must carry explicit high-res Open Original URLs. import json as _json def _load_items(_path): _txt = open(_path, encoding="utf-8").read() return _json.loads(_txt[_txt.find("["):_txt.rfind("]")+1]) for _rel in ["schnellfotos-kaufhaus/cdv-data-en.js", "sequenzen/cdv-data-en.js", "bueckeburger-trachten/cdv-data-en.js"]: _path = root / _rel if _path.exists(): for _item in _load_items(_path): if str(_item.get("pageUrl", "")).startswith("https://reinhard-krause.de/sammlung/picture.php?"): _url = str(_item.get("openOriginalUrl") or _item.get("originalImage") or "") if "action.php?download=" not in _url or f"id={_item.get('id')}" not in _url: errors.append(f"{_rel}: ID {_item.get('id')} has no explicit Piwigo high-res openOriginalUrl") _rev = _item.get("reverse") if isinstance(_rev, dict) and str(_rev.get("pageUrl", "")).startswith("https://reinhard-krause.de/sammlung/picture.php?"): _url = str(_rev.get("openOriginalUrl") or _rev.get("originalImage") or "") if "action.php?download=" not in _url or f"id={_rev.get('id')}" not in _url: errors.append(f"{_rel}: reverse ID {_rev.get('id')} has no explicit Piwigo high-res openOriginalUrl") if errors: print('STRUKTURCHECK FEHLGESCHLAGEN') print('\n'.join(errors)) sys.exit(1) print('STRUKTURCHECK OK') # Additional static guard for vs20: Core must prioritize openOriginalUrl through browserOpenImageUrl. from pathlib import Path as _Path _core = root / 'shared/archive-core/archive-app.js' if _core.exists(): _txt = _core.read_text(encoding='utf-8') if 'browserOpenImageUrl(item.openOriginalUrl)' not in _txt: raise SystemExit('STRUCTURECHECK FAILED: Open original must prioritize browserOpenImageUrl(item.openOriginalUrl).') _master = root / 'workflow/MASTER-INSTRUCTIONS.md' if _master.exists(): _m = _master.read_text(encoding='utf-8') if 'openOriginalUrl = verbindliches Ziel' not in _m or 'pageUrl allein ist' not in _m: raise SystemExit('STRUCTURECHECK FAILED: MASTER-INSTRUCTIONS missing strict High-Res openOriginalUrl rule.') # vs27 additional standalone project safety check from pathlib import Path as _Path for _folder in []: _idx = _Path(_folder) / "index.html" if _idx.exists(): _txt = _idx.read_text(encoding="utf-8") assert 'class="nav-link"' in _txt and 'cdv-map-en.html' in _txt, f"{_folder} must include map nav-link; archive-app expects .top-actions .nav-link" assert (_Path(_folder) / "cdv-data-en.js").exists(), f"{_folder} missing cdv-data-en.js" # ENGLISH_FIELD_INTEGRITY_CHECK_REQUIRED: titleEn/captionEn must not contain German source fragments. # Blocking examples: eines, einer, einem, mit, und, Datierung, Innenansicht, Maenner, Frauen, Kinder, Gebaeude, umlauts.