"""Extract useful collection metadata from In-Stock Kitchen Catalog 2026 spreads PDF.""" from __future__ import annotations import json import re from pathlib import Path from pypdf import PdfReader PDF = Path(r"C:/Users/Mohsin.Abbas/Downloads/In-Stock-Kitchen-Catalog-2026-spreads.pdf") TAX = Path(r"E:/2026/plytixmage_dbs_data/kitchen_cabinets_taxonomy.json") BROCHURE = Path(r"E:/2026/plytixmage_dbs_data/Brochure_Data") GENERAL_GAP = Path(r"E:/2026/plytixmage_dbs_data/audit_reports/general_catalog_collections_gap.json") OUT_TXT = Path(r"E:/2026/plytixmage_dbs_data/audit_reports/_instock_kitchen_2026_full.txt") OUT_JSON = Path(r"E:/2026/plytixmage_dbs_data/audit_reports/instock_kitchen_2026_extract.json") # Known / suspected kitchen collection names (expanded) KNOWN = [ "Harriet Pearl White", "Anna Snow White", "Anna Stone Gray", "Anna Natural Rift White Oak", "Anna Caramel Harvest", "Eleanor Snow White", "Eleanor Ocean Blue", "Maya Snow White", "Maya Natural Rift White Oak", "Maya Black Rift White Oak", "Cooper Slate Gray", "Merrill Polar White", "Merrill Slate", "Hurston Midnight Espresso", "Johnson Sandstone", "Johnson White", "Belmont Drift White", "Belmont Toffee Pecan", "Chesapeake Golden Oak", "Metro Mist", "Metro Frost", "Galaxy Horizon", "Hallmark Frost", "Riley", "Sanibel", "Plymouth Espresso Maple", "Grace Snow White", "Coronet Honey", "Whitney", "Williamson", "Quest Metro Frost", "Fabuwood Metro Mist", "Fabuwood Galaxy Horizon", "Fabuwood Hallmark Frost", ] def norm(s: str) -> str: s = re.sub(r"\s+", " ", str(s or "").strip().lower()) s = s.replace("&", "and") return re.sub(r"[^a-z0-9 ]+", "", s) def main() -> None: reader = PdfReader(str(PDF)) parts = [] for i, page in enumerate(reader.pages): parts.append(f"--- PAGE {i+1} ---\n{page.extract_text() or ''}") text = "\n".join(parts) OUT_TXT.write_text(text, encoding="utf-8") # Prefix codes like ACH-_, HPW-_, SRL_, LPE_ prefixes = sorted(set(re.findall(r"\b([A-Z]{2,5})[_-]?\s*_?\b", text))) prefix_hits = sorted(set(re.findall(r"\b([A-Z]{2,5})-_\b", text))) prefix_hits2 = sorted(set(re.findall(r"\b([A-Z]{2,5})_\b", text))) # ALL CAPS multi-word headings caps = [] for ln in text.splitlines(): ln = ln.strip() if re.match(r"^[A-Z][A-Z0-9'&\- ]{5,80}$", ln) and len(ln.split()) >= 2: if ln.lower() not in { "other features", "base cabinets", "wall cabinets", "tall cabinets", "special bases", "special wall", "basic bases", "basic wall", "bridge wall", "corner wall", "corner bases", "drawer bases", "sink bases", "panels fillers", "panels & fillers", "moldings", "mouldings", "accessories", "miscellaneous", "specifications", "construction", "matching styles", "kitchen cabinets", "made in the usa", "limited warranty", }: caps.append(ln) # Find known collection mentions (case-insensitive) found = [] for name in KNOWN: # allow flexible whitespace pat = re.compile(re.escape(name).replace(r"\ ", r"\s+"), re.I) ms = list(pat.finditer(text)) if not ms: continue # nearby context snippets contexts = [] for m in ms[:3]: start = max(0, m.start() - 80) end = min(len(text), m.end() + 120) contexts.append(re.sub(r"\s+", " ", text[start:end]).strip()) found.append({"name": name, "mentions": len(ms), "contexts": contexts}) # Matching Styles blocks matching_blocks = [] lines = [ln.strip() for ln in text.splitlines()] for i, ln in enumerate(lines): if ln.lower() in {"matching styles", "matching style"}: block = [] for nxt in lines[i + 1 : i + 15]: if not nxt: continue low = nxt.lower() if low in {"style", "construction", "other features", "specifications", "cabinetry"}: break if "additional items" in low: break block.append(nxt) matching_blocks.append(block) # Stock / availability keywords stock_terms = {} for term in [ "Grab & Go", "GRAB & GO", "Special Order", "AVAILABLE", "In-Stock", "In Stock", "Take Home", "Keyport", "North Bergen", "Bellmawr", ]: stock_terms[term] = len(re.findall(re.escape(term), text, flags=re.I)) # SKU-like tokens sku_samples = sorted(set(re.findall(r"\b(?:[A-Z]{2,5}-)?[A-Z]{1,6}\d{2,4}[A-Z0-9-]*\b", text)))[:80] tax = json.loads(TAX.read_text(encoding="utf-8")) tax_names = { norm(v.get("display_name") or v.get("name") or k): {"code": k, "name": v.get("display_name") or v.get("name")} for k, v in (tax.get("collections") or {}).items() } brochure_files = [p.name for p in sorted(BROCHURE.glob("*.json"))] general = {} if GENERAL_GAP.exists(): general = json.loads(GENERAL_GAP.read_text(encoding="utf-8")) general_missing_tax = [r["name"] for r in general.get("missing_in_taxonomy") or []] rows = [] for item in found: key = norm(item["name"]) # alias normalize aliases = { "quest metro frost": "metro frost", "fabuwood metro mist": "metro mist", "fabuwood galaxy horizon": "galaxy horizon", "fabuwood hallmark frost": "hallmark frost", } lookup_key = aliases.get(key, key) in_tax = lookup_key in tax_names or key in tax_names rows.append({**item, "in_taxonomy": in_tax, "taxonomy": tax_names.get(lookup_key) or tax_names.get(key)}) new_vs_general = [] for item in rows: if item["name"] not in [r.get("name") for r in (general.get("pdf_collections") or [])]: # also check norm gkeys = {norm(r["name"]) for r in (general.get("pdf_collections") or [])} if norm(item["name"]) not in gkeys and norm(item["name"]).replace("quest ", "").replace("fabuwood ", "") not in gkeys: new_vs_general.append(item) # Unique primary caps cleaned caps_clean = [] seen = set() for c in caps: name = re.sub(r"\s+", " ", c).title() if c.isupper() else c name = name.replace("Whit E", "White").replace("W Ood", "Wood") key = norm(name) if key in seen: continue if any(x in key for x in ("warranty", "limited", "grab", "available", "special order")): continue seen.add(key) caps_clean.append({"raw": c, "name": name}) report = { "pdf": str(PDF), "pages": len(reader.pages), "collection_mentions": rows, "collections_not_in_general_catalog_parse": new_vs_general, "still_missing_taxonomy_from_either_catalog": sorted( set(general_missing_tax) | { r["name"] for r in rows if not r["in_taxonomy"] and norm(r["name"]) not in { "quest metro frost", "fabuwood metro mist", "fabuwood galaxy horizon", "fabuwood hallmark frost", "johnson white", "merrill slate", "coronet honey", "whitney", "williamson", "grace snow white", } or ( not r["in_taxonomy"] and r["name"] in { "Metro Mist", "Metro Frost", "Riley", "Sanibel", "Quest Metro Frost", "Fabuwood Metro Mist", } ) } ), "prefix_codes_underscore": prefix_hits2, "prefix_codes_dash_underscore": prefix_hits, "matching_styles_blocks": matching_blocks, "stock_term_counts": stock_terms, "caps_headings": caps_clean, "sku_samples": sku_samples, "brochure_files": brochure_files, "full_text_path": str(OUT_TXT), } # Cleaner missing list missing = [] for name in [ "Metro Mist", "Metro Frost", "Riley", "Sanibel", "Plymouth Espresso Maple", "Grace Snow White", "Johnson White", "Merrill Slate", "Coronet Honey", "Whitney", "Williamson", ]: hit = next((r for r in rows if norm(r["name"]) == norm(name)), None) in_tax = norm(name) in tax_names or any(norm(name) in k or k in norm(name) for k in tax_names) # brochure fuzzy bro_hit = any(norm(name).replace(" ", "") in norm(f).replace(" ", "") for f in brochure_files) missing.append( { "name": name, "mentioned_in_instock_pdf": bool(hit), "mentions": hit["mentions"] if hit else 0, "in_taxonomy": bool(in_tax), "in_brochure_files": bro_hit, } ) report["gap_checklist"] = missing OUT_JSON.write_text(json.dumps(report, indent=2), encoding="utf-8") print(f"pages={len(reader.pages)} wrote {OUT_TXT}") print("\nCollections mentioned:") for r in rows: print(f" {r['name']:35} mentions={r['mentions']:<3} tax={'Y' if r['in_taxonomy'] else 'N'}") print("\nPrefix codes (*-_ / *_):", prefix_hits, prefix_hits2) print("\nMatching Styles blocks:") for i, b in enumerate(matching_blocks, 1): print(f" #{i}: {b}") print("\nGap checklist:") for g in missing: print( f" {g['name']:30} instock={g['mentioned_in_instock_pdf']} tax={g['in_taxonomy']} brochure={g['in_brochure_files']}" ) print("\nStock terms:", stock_terms) print("wrote", OUT_JSON) if __name__ == "__main__": main()