IKEAwin + cookie

Парсер_IKEA/main.py

@@ -433,6 +433,15 @@ def extract_data(url: str) -> dict:
     try:
         resp = requests.get(url, headers=HEADERS, timeout=15)
         resp.raise_for_status()
+                # 🔎 DEBUG: вывести в консоль базовую информацию об ответе
+        print("\n=== FETCH DEBUG ===")
+        print("URL:        ", url)
+        print("Final URL:  ", resp.url)
+        print("Status:     ", resp.status_code)
+        print("ContentType:", resp.headers.get("Content-Type"))
+        print("Length:     ", len(resp.text))
+        print("Snippet ↓↓↓")
+        print(resp.text[:1000])  # покажет первые 1000 символов HTML
         soup = BeautifulSoup(resp.text, "html.parser")
 
         target = soup.select_one(CSS_SELECTOR)
@@ -510,6 +519,7 @@ def extract_data(url: str) -> dict:
         return filtered
 
     except Exception as e:
+        print(e)
         return {"url": url, "error": str(e)}
 
 # ───────────────────── ПОСТРОЕНИЕ ВАРИАНТА / POST ─────────────────

Парсер_IKEA/main_win.py

@@ -1,15 +1,21 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 
-import os, json, re, math, time, html, requests, datetime, http.cookiejar as cookiejar
-from collections import Counter
-from typing import List, Optional
+"""
+IKEA Parser — надежный вариант: только Playwright (Chrome, persistent profile),
+без requests, без параллельности и ускорителей. По одной вкладке, максимум логов.
+"""
+
+import os, sys, re, json, math, time, html as html_mod, datetime, traceback
+from typing import Optional
+import logging
+from logging.handlers import RotatingFileHandler
+from urllib import request as urlrequest
+from urllib.error import URLError, HTTPError
+
 from bs4 import BeautifulSoup
 from openpyxl import Workbook
-from requests.adapters import HTTPAdapter
-from urllib3.util.retry import Retry
-import logging
-import socket
+from playwright.sync_api import sync_playwright, TimeoutError as PWTimeout
 
 # ───────────────────────── ПУТИ / ФАЙЛЫ ───────────────────────────
 BASE_DIR       = os.path.dirname(os.path.abspath(__file__))
@@ -21,47 +27,19 @@ OUTPUT_FILE    = os.path.join(RECORDS_DIR, "records.xlsx")
 DICT_FILE      = os.path.join(BASE_DIR, "dictionary_main.txt")
 EXCL_FILE      = os.path.join(BASE_DIR, "exclusion_materials.txt")
 POST_LOG       = os.path.join(RECORDS_DIR, "post_log.txt")
-FETCH_LOG      = os.path.join(RECORDS_DIR, "fetch_debug.log")
-COOKIES_TXT    = os.path.join(BASE_DIR, "cookies.txt")
 
-# ───────────────────────── ЛОГИРОВАНИЕ ────────────────────────────
-logger = logging.getLogger("ikea_parser")
-logger.setLevel(logging.DEBUG)
-# файл — максимум подробностей
-fh = logging.FileHandler(FETCH_LOG, encoding="utf-8")
-fh.setLevel(logging.DEBUG)
-fh.setFormatter(logging.Formatter("%(asctime)s [%(levelname)s] %(message)s"))
-# консоль — INFO
-ch = logging.StreamHandler()
-ch.setLevel(logging.INFO)
-ch.setFormatter(logging.Formatter("%(message)s"))
-logger.addHandler(fh)
-logger.addHandler(ch)
+LOGS_DIR       = os.path.join(RECORDS_DIR, "logs")
+HTML_DIR       = os.path.join(RECORDS_DIR, "html_debug")
+JSON_DIR       = os.path.join(RECORDS_DIR, "json_debug")
+PROFILE_DIR    = os.path.join(BASE_DIR, "playwright_profile")
+os.makedirs(LOGS_DIR, exist_ok=True)
+os.makedirs(HTML_DIR, exist_ok=True)
+os.makedirs(JSON_DIR, exist_ok=True)
+os.makedirs(PROFILE_DIR, exist_ok=True)
 
-logger.info("=== IKEA parser started ===")
-logger.info(f"BASE_DIR={BASE_DIR}")
-logger.info(f"Python={os.sys.version}")
-try:
-    logger.info(f"Hostname={socket.gethostname()} IP={socket.gethostbyname(socket.gethostname())}")
-except Exception as _e:
-    logger.info("Hostname/IP: unavailable")
+APP_LOG_FILE   = os.path.join(LOGS_DIR, "app.log")
 
-# ───────────────────────── НАСТРОЙКИ POST ─────────────────────────
-POST_URL     = os.getenv("IKEA_POST_URL", "http://localhost:3005/parser/data")
-POST_API_KEY = os.getenv("IKEA_POST_API_KEY", "")
-POST_TIMEOUT = 20
-BATCH_SIZE   = 50
-
-# ───────────────────────── НАСТРОЙКИ САЙТА ────────────────────────
-HEADERS = {
-    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
-                  "AppleWebKit/537.36 (KHTML, like Gecko) "
-                  "Chrome/124.0.0.0 Safari/537.36",
-    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
-    "Accept-Language": "pl-PL,pl;q=0.9,en;q=0.8,ru;q=0.7",
-    "Cache-Control": "no-cache",
-    "Pragma": "no-cache",
-}
+# ───────────────────────── НАСТРОЙКИ ──────────────────────────────
 CSS_SELECTOR = ".pip-product__subgrid.product-pip.js-product-pip"
 
 BLOCKS = [
@@ -106,45 +84,40 @@ KEEP_COLUMNS = [
     "url",
 ]
 
-# ───────────────────────── HTTP СЕССИЯ ────────────────────────────
-def make_session() -> requests.Session:
-    s = requests.Session()
-    s.headers.update(HEADERS)
-    # игнор системных прокси/mitm переменных окружения Windows
-    s.trust_env = False
-    retries = Retry(
-        total=5,
-        backoff_factor=0.5,
-        status_forcelist=(403, 429, 500, 502, 503, 504),
-        allowed_methods=frozenset(["GET", "POST"])
-    )
-    s.mount("https://", HTTPAdapter(max_retries=retries))
-    s.mount("http://", HTTPAdapter(max_retries=retries))
-    return s
+UA = os.getenv(
+    "IKEA_UA",
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+    "(KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36"
+)
+ACCEPT_LANG = os.getenv("IKEA_ACCEPT_LANGUAGE", "pl-PL,pl;q=0.9,en;q=0.8,ru;q=0.7")
 
-SESSION = make_session()
+# Playwright запускаем Chrome, persistent profile; по умолчанию не headless —
+# так Cloudflare реже блокирует.
+HEADLESS = os.getenv("IKEA_HEADLESS", "0") not in {"0", "false", "False", ""}
 
-def load_netscape_cookies(session: requests.Session, path: str):
-    if os.path.isfile(path):
-        cj = cookiejar.MozillaCookieJar()
-        try:
-            cj.load(path, ignore_discard=True, ignore_expires=True)
-            session.cookies.update(cj)
-            logger.info(f"🍪 Cookies loaded: {path} ({len(cj)} pcs)")
-        except Exception as e:
-            logger.warning(f"⚠️ Failed to load cookies.txt: {e}")
-    else:
-        logger.info("cookies.txt not found — proceeding without external cookies")
+# POST/API без requests
+POST_URL     = os.getenv("IKEA_POST_URL", "http://localhost:3005/parser/data")
+POST_API_KEY = os.getenv("IKEA_POST_API_KEY", "")
+POST_TIMEOUT = int(os.getenv("IKEA_POST_TIMEOUT", "20"))
+BATCH_SIZE   = int(os.getenv("IKEA_BATCH_SIZE", "50"))
 
-load_netscape_cookies(SESSION, COOKIES_TXT)
+# ───────────────────────── ЛОГИРОВАНИЕ ────────────────────────────
+logger = logging.getLogger("ikea_pw_simple")
+logger.setLevel(logging.DEBUG)
+_fmt = logging.Formatter("%(asctime)s [%(levelname)s] %(message)s")
 
-# ───────────────────────── УТИЛИТЫ I/O ────────────────────────────
-def ask_bool(prompt: str, default: str = "1") -> bool:
-    try:
-        val = input(f"{prompt} (1=yes, 0=no) [{default}]: ").strip() or default
-    except EOFError:
-        val = default
-    return val == "1"
+fh = RotatingFileHandler(APP_LOG_FILE, maxBytes=2_000_000, backupCount=3, encoding="utf-8")
+fh.setFormatter(_fmt)
+fh.setLevel(logging.DEBUG)
+logger.addHandler(fh)
+
+ch = logging.StreamHandler(sys.stdout)
+ch.setFormatter(_fmt)
+ch.setLevel(logging.INFO)
+logger.addHandler(ch)
+
+def _now_tag():
+    return datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
 
 def _post_log(msg: str):
     try:
@@ -153,50 +126,30 @@ def _post_log(msg: str):
     except Exception:
         pass
 
-def _now_tag():
-    return datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
-
 def _save_json_batch(payload: dict, batch_index: int):
     fname = f"ikea_batch_{_now_tag()}_{batch_index:04d}.json"
-    fpath = os.path.join(RECORDS_DIR, fname)
+    fpath = os.path.join(JSON_DIR, fname)
     with open(fpath, "w", encoding="utf-8") as fh:
         json.dump(payload, fh, ensure_ascii=False, indent=2)
-    logger.info(f"💾 JSON saved: {fname}")
+    logger.debug(f"💾 JSON saved: {fpath}")
     return fpath
 
-def _safe_name_from_url(url: str) -> str:
-    return re.sub(r"[^a-zA-Z0-9]+", "_", url)[:80]
-
-def _dump_meta(prefix: str, url: str, status: int, elapsed: float, text_len: int, final_url: str, headers: dict, note: str = ""):
+def _save_html_snapshot(prefix: str, idx: int, content: str):
+    fname = f"{idx:04d}_{prefix}_{_now_tag()}.html"
+    fpath = os.path.join(HTML_DIR, fname)
     try:
-        base = f"{prefix}_{_now_tag()}_{_safe_name_from_url(url)}"
-        meta = os.path.join(RECORDS_DIR, base + ".meta.txt")
-        with open(meta, "w", encoding="utf-8") as fh:
-            fh.write(f"URL: {url}\n")
-            fh.write(f"FINAL_URL: {final_url}\n")
-            fh.write(f"STATUS: {status}\n")
-            fh.write(f"ELAPSED_SEC: {elapsed:.3f}\n")
-            fh.write(f"RESP_LEN: {text_len}\n")
-            fh.write(f"NOTE: {note}\n")
-            fh.write("HEADERS:\n")
-            for k, v in headers.items():
-                hv = v if isinstance(v, str) else str(v)
-                fh.write(f"  {k}: {hv}\n")
-    except Exception as e:
-        logger.debug(f"Meta dump failed: {e}")
-
-def _save_debug_html(url: str, text: str, prefix: str = "debug", note: str = "", status: Optional[int] = None, elapsed: Optional[float] = None, headers: Optional[dict] = None, final_url: Optional[str] = None):
-    try:
-        base = f"{prefix}_{_now_tag()}_{_safe_name_from_url(url)}"
-        fpath = os.path.join(RECORDS_DIR, base + ".html")
         with open(fpath, "w", encoding="utf-8") as fh:
-            fh.write(text)
-        logger.info(f"🧪 Saved HTML snapshot: {os.path.basename(fpath)}")
-        # мета рядом
-        if status is not None and headers is not None and final_url is not None and elapsed is not None:
-            _dump_meta(prefix, url, status, elapsed, len(text or ""), final_url, headers, note=note)
-    except Exception as e:
-        logger.debug(f"HTML dump failed: {e}")
+            fh.write(content)
+        logger.debug("🧪 HTML snapshot: %s", fpath)
+    except Exception:
+        logger.exception("Failed to save HTML snapshot")
+
+def ask_bool(prompt: str, default: str = "1") -> bool:
+    try:
+        val = input(f"{prompt} (1=yes, 0=no) [{default}]: ").strip() or default
+    except EOFError:
+        val = default
+    return val == "1"
 
 # ───────────────────────── СЛОВАРИ / ФИЛЬТРЫ ──────────────────────
 def load_dictionary(path: str) -> dict:
@@ -207,7 +160,6 @@ def load_dictionary(path: str) -> dict:
     return {k: v for k, v in pairs}
 
 DICT = load_dictionary(DICT_FILE)
-
 def translate_token(token: str) -> str:
     return DICT.get(token, token)
 
@@ -221,8 +173,8 @@ def load_exclusions(path: str) -> set:
 
 EXCLUSIONS = load_exclusions(EXCL_FILE)
 
-def materials_from_details_json(details: dict) -> List[str]:
-    out: List[str] = []
+def materials_from_details_json(details: dict) -> list[str]:
+    out: list[str] = []
     def walk(node):
         if isinstance(node, dict):
             for k, v in node.items():
@@ -430,14 +382,14 @@ def format_product_details(raw_details, add_summary_desc="", with_html=False, sk
 
     if mats:
         out.append("Materiały")
-        for m in mats:
-            ptype = m.get("productType", "")
-            for mat in (m.get("materials") or []):
-                material = mat.get("material", "")
-                if ptype:
-                    out.append(ptype)
-                if material:
-                    out.append(material)
+    for m in mats:
+        ptype = m.get("productType", "")
+        for mat in (m.get("materials") or []):
+            material = mat.get("material", "")
+            if ptype:
+                out.append(ptype)
+            if material:
+                out.append(material)
 
     if care:
         detailsCareText = mac.get("detailsCareText", "Pielęgnacja")
@@ -473,207 +425,160 @@ def build_variant_color_measure(desc: str, type_name: str, measurement: str) ->
     if t:
         pattern = r"^\s*" + re.escape(t) + r"[\s,;:\-–—/]*"
         s = re.sub(pattern, "", s, flags=re.IGNORECASE)
-
     if not re.search(r"[0-9A-Za-zА-Яа-яЁёÀ-ž]", s or ""):
         s = ""
-
     s = s.strip()
     meas = (measurement or "").strip()
-
     if not s:
         return meas if meas else ""
-
     s = s[:1].upper() + s[1:]
     return f"{s}, {meas}" if meas else s
 
-# ───────────────────── СКРАПИНГ КАРТОЧКИ ──────────────────────────
-def extract_data(url: str, force_dump: bool = False) -> dict:
+# ───────────────────── Playwright — одна вкладка ──────────────────
+def open_browser():
+    pw = sync_playwright().start()
+    # persistent Chrome: Cloudflare к нему относится лояльнее
+    ctx = pw.chromium.launch_persistent_context(
+        PROFILE_DIR,
+        headless=HEADLESS,
+        channel="chrome",          # важное отличие
+        user_agent=UA,
+        locale="pl-PL",
+        java_script_enabled=True,
+        accept_downloads=False,
+        viewport={"width": 1366, "height": 864},
+        # можно добавить proxy={"server": "..."} при необходимости
+    )
+    page = ctx.new_page()
+    # базовые заголовки (совпадают с реальным браузером)
+    page.set_extra_http_headers({"Accept-Language": ACCEPT_LANG})
+    return pw, ctx, page
+
+def close_browser(pw, ctx):
     try:
-        logger.debug(f"GET {url}")
-        t0 = time.time()
-        resp = SESSION.get(url, timeout=25, allow_redirects=True)
-        elapsed = time.time() - t0
-        status = resp.status_code
-        final_url = str(getattr(resp, "url", url))
-        text_len = len(resp.text or "")
-        logger.info(f"HTTP {status} {final_url} ({elapsed:.2f}s, {text_len} bytes)")
+        ctx.close()
+        pw.stop()
+    except Exception:
+        pass
 
-        # Всегда сохраняем первые (force_dump=True) или любую «сомнительную» страницу
-        need_dump = force_dump or status != 200 or ("data-hydration-props" not in resp.text)
-        if need_dump:
-            note = "force_dump" if force_dump else ("no_hydration" if "data-hydration-props" not in resp.text else f"status_{status}")
-            _save_debug_html(url, resp.text, prefix=f"resp{status}", note=note, status=status, elapsed=elapsed, headers=resp.headers, final_url=final_url)
+def fetch_page(page, url: str, idx: int) -> tuple[str, Optional[str]]:
+    """
+    Возвращает (full_html, hydration_raw_json_or_None).
+    Сохраняет снапшот, если не нашли data-hydration-props.
+    """
+    t0 = time.time()
+    resp = page.goto(url, wait_until="domcontentloaded", timeout=60_000)
+    status = resp.status if resp else 0
+    # ждём селектор, но не слишком долго
+    try:
+        page.wait_for_selector(CSS_SELECTOR, timeout=25_000, state="attached")
+    except PWTimeout:
+        pass
+    full_html = page.content()
+    # прямой атрибут
+    raw = None
+    try:
+        el = page.locator(CSS_SELECTOR).first
+        raw = el.get_attribute("data-hydration-props")
+    except Exception:
+        raw = None
+    elapsed = time.time() - t0
+    logger.info("PW %s  status=%s  %.2fs  len=%s", url, status, elapsed, len(full_html or ""))
+    # если Cloudflare/403 — сохраним снапшот для диагностики
+    if not raw:
+        _save_html_snapshot("no_hydration", idx, full_html or "")
+    return full_html or "", raw
 
-        resp.raise_for_status()
+# ───────────────────── Парсинг страницы ───────────────────────────
+def parse_page(url: str, full_html: str, raw_json: Optional[str]) -> dict:
+    if not full_html:
+        return {"url": url, "error": "no html"}
 
-        soup = BeautifulSoup(resp.text, "html.parser")
+    soup = BeautifulSoup(full_html, "html.parser")
+    # Fallback: если не пришёл raw, попробуем из DOM
+    if not raw_json:
         target = soup.select_one(CSS_SELECTOR)
-        if not target:
-            logger.warning("CSS selector NOT FOUND")
-            _save_debug_html(url, resp.text, prefix="no_selector", note="css_selector_missing", status=status, elapsed=elapsed, headers=resp.headers, final_url=final_url)
-            return {"url": url, "error": "CSS selector not found", "http_status": status}
-
-        raw = target.get("data-hydration-props")
-        if not raw:
-            logger.warning("data-hydration-props NOT FOUND")
-            _save_debug_html(url, resp.text, prefix="no_hydration", note="attribute_missing", status=status, elapsed=elapsed, headers=resp.headers, final_url=final_url)
-            return {"url": url, "error": "data-hydration-props not found", "http_status": status}
-
-        decoded = html.unescape(raw)
-        try:
-            full_json = json.loads(decoded)
-        except Exception as je:
-            logger.error(f"JSON decode error: {je}")
-            # сохраним кусок для анализа
-            sample_name = f"bad_json_{_now_tag()}_{_safe_name_from_url(url)}.txt"
-            with open(os.path.join(RECORDS_DIR, sample_name), "w", encoding="utf-8") as fh:
-                fh.write(decoded[:20000])
-            return {"url": url, "error": f"json decode error: {je}", "http_status": status}
-
-        result = {"url": url}
-        for block in BLOCKS:
-            result.update(flatten_block(block, full_json.get(block, {})))
-
-        kf_json  = _parse_json_value(result.get("keyFacts.keyFacts"))
-        dim_json = _parse_json_value(result.get("productInformationSection.dimensionProps"))
-        det_json = _parse_json_value(result.get("productInformationSection.productDetailsProps"))
-
-        result["keyFacts.keyFacts_formatted"] = format_keyfacts(kf_json)
-        result["productInformationSection.dimensionProps_formatted"] = format_dimensions(dim_json, with_html=False, translated=False)
-        html_trans = format_dimensions(dim_json, with_html=True, translated=True)
-        if isinstance(html_trans, str) and html_trans.startswith("strong>"):
-            html_trans = "<" + html_trans
-        result["productInformationSection.dimensionProps_formatted_html_translated"] = html_trans
-
-        total_kg = _collect_packaging_total_kg((dim_json or {}).get("packaging") or {})
-        result["total brutto"] = _fmt_float(total_kg)
-
-        summary_desc = result.get("productSummary.description", "") or ""
-        result["productInformationSection.productDetailsProps_formatted"] = format_product_details(det_json, add_summary_desc=summary_desc, with_html=False, skip_assembly=True)
-        result["productInformationSection.productDetailsProps_formatted_html"] = format_product_details(det_json, add_summary_desc=summary_desc, with_html=True,  skip_assembly=True)
-
-        desc  = result.get("pipPricePackage.productDescription", "") or ""
-        tname = result.get("stockcheckSection.typeName", "") or ""
-        meas  = result.get("pipPricePackage.measurementText", "") or ""
-        result["prductVariantColorMeasure"] = build_variant_color_measure(desc, tname, meas)
-
-        # breadcrumb
-        breadcrumb = None
-        for tag in soup.find_all("script", attrs={"type": lambda t: t and "ld+json" in t}):
-            try:
-                data = json.loads(tag.string)
-            except Exception:
-                continue
-            if isinstance(data, list):
-                data = next((d for d in data if isinstance(d, dict) and d.get("@type") == "BreadcrumbList"), None)
-            if isinstance(data, dict) and data.get("@type") == "BreadcrumbList":
-                items = data.get("itemListElement", [])
-                names = [it.get("name", "") for it in items]
-                breadcrumb = "/".join(names)
-                break
-        if breadcrumb:
-            result["categoryBreadcrumb"] = breadcrumb
-
-        # применяем whitelist
-        filtered = {k: result.get(k) for k in KEEP_COLUMNS if k != "originalName"}
-
-        # originalName = productName + " " + typeName
-        pn = (result.get("buyModule.productName") or "").strip()
-        tn = (result.get("stockcheckSection.typeName") or "").strip()
-        if pn and tn:
-            orig_name = f"{pn} {tn}"
-        else:
-            orig_name = pn or tn
-        filtered["originalName"] = orig_name
-
-        return filtered
+        if target:
+            raw_json = target.get("data-hydration-props")
+    if not raw_json:
+        return {"url": url, "error": "data-hydration-props not found"}
 
+    try:
+        decoded   = html_mod.unescape(raw_json)
+        full_json = json.loads(decoded)
     except Exception as e:
-        logger.error(f"Request error for {url}: {e}")
-        return {"url": url, "error": str(e), "http_status": None}
+        return {"url": url, "error": f"json decode error: {e}"}
 
-# ───────────────────── ПОСТРОЕНИЕ ВАРИАНТА / POST ─────────────────
-def _split_color_size(text: str):
-    if not text:
-        return "", ""
-    parts = [p.strip() for p in text.split(",", 1)]
-    if len(parts) == 2:
-        return parts[0], parts[1]
-    return "", parts[0]
+    result = {"url": url}
+    for block in BLOCKS:
+        result.update(flatten_block(block, full_json.get(block, {})))
 
-def _ceil_price(v):
-    try:
-        return int(math.ceil(float(v)))
-    except Exception:
-        return None
+    kf_json  = _parse_json_value(result.get("keyFacts.keyFacts"))
+    dim_json = _parse_json_value(result.get("productInformationSection.dimensionProps"))
+    det_json = _parse_json_value(result.get("productInformationSection.productDetailsProps"))
 
-def _ceil_int(v):
-    try:
-        return int(math.ceil(float(v)))
-    except Exception:
-        return None
+    result["keyFacts.keyFacts_formatted"] = format_keyfacts(kf_json)
+    result["productInformationSection.dimensionProps_formatted"] = format_dimensions(dim_json, with_html=False, translated=False)
+    html_trans = format_dimensions(dim_json, with_html=True, translated=True)
+    if isinstance(html_trans, str) and html_trans.startswith("strong>"):
+        html_trans = "<" + html_trans
+    result["productInformationSection.dimensionProps_formatted_html_translated"] = html_trans
 
-def build_variant(row: dict) -> dict:
-    visible = row.get("productSummary.visibleItemNo") or ""
-    sku = visible.replace(" ", "")
+    total_kg = _collect_packaging_total_kg((dim_json or {}).get("packaging") or {})
+    result["total brutto"] = _fmt_float(total_kg)
 
-    csm = (row.get("prductVariantColorMeasure") or "").strip()
-    color, size = _split_color_size(csm)
-    if not color and not size:
-        size = (row.get("pipPricePackage.measurementText") or "").strip()
+    summary_desc = result.get("productSummary.description", "") or ""
+    result["productInformationSection.productDetailsProps_formatted"] = format_product_details(det_json, add_summary_desc=summary_desc, with_html=False, skip_assembly=True)
+    result["productInformationSection.productDetailsProps_formatted_html"] = format_product_details(det_json, add_summary_desc=summary_desc, with_html=True,  skip_assembly=True)
 
-    cost = _ceil_price(row.get("buyModule.productPrice"))
-    url  = row.get("url") or ""
+    desc  = result.get("pipPricePackage.productDescription", "") or ""
+    tname = result.get("stockcheckSection.typeName", "") or ""
+    meas  = result.get("pipPricePackage.measurementText", "") or ""
+    result["prductVariantColorMeasure"] = build_variant_color_measure(desc, tname, meas)
 
-    name = row.get("originalName") or row.get("buyModule.productName") or ""
-    desc_html = row.get("productInformationSection.productDetailsProps_formatted_html") or ""
-    composition_html = row.get("productInformationSection.dimensionProps_formatted_html_translated") or ""
+    # breadcrumb
+    breadcrumb = None
+    for tag in soup.find_all("script", attrs={"type": lambda t: t and "ld+json" in t}):
+        try:
+            data = json.loads(tag.string)
+        except Exception:
+            continue
+        if isinstance(data, list):
+            data = next((d for d in data if isinstance(d, dict) and d.get("@type") == "BreadcrumbList"), None)
+        if isinstance(data, dict) and data.get("@type") == "BreadcrumbList":
+            items = data.get("itemListElement", [])
+            names = [it.get("name", "") for it in items]
+            breadcrumb = "/".join(names)
+            break
+    if breadcrumb:
+        result["categoryBreadcrumb"] = breadcrumb
 
-    imgs = []
-    raw_imgs = row.get("productGallery.urls") or ""
-    if isinstance(raw_imgs, str):
-        imgs = [x for x in raw_imgs.split("\n") if x.strip()]
+    filtered = {k: result.get(k) for k in KEEP_COLUMNS if k != "originalName"}
+    pn = (result.get("buyModule.productName") or "").strip()
+    tn = (result.get("stockcheckSection.typeName") or "").strip()
+    filtered["originalName"] = (f"{pn} {tn}".strip() or pn or tn)
 
-    in_stock = bool(row.get("availabilityGroup.serverOnlineSellable")) or bool(row.get("buyModule.onlineSellable"))
-
-    weight_kg = _ceil_int(row.get("total brutto"))
-
-    variant = {
-        "status_id": 1,
-        "color": color.capitalize() if color else "none",
-        "sku": sku,
-        "size": size,
-        "cost": cost,
-        "originalUrl": url,
-        "originalName": name,
-        "originalDescription": desc_html,
-        "originalComposition": composition_html,
-        "images": imgs,
-        "inStock": in_stock,
-        "weight": weight_kg if weight_kg is not None else 0,
-    }
-
-    return {
-        "category": {"name": "TEST/IKEA"},  # временно по вашему ТЗ
-        "brand":    {"name": "ikea"},
-        "variant":  variant,
-    }
+    return filtered
 
+# ───────────────────── POST (urllib) ──────────────────────────────
 def post_payload(payload: dict) -> dict:
-    headers = {"Content-Type": "application/json"}
+    headers = {"Content-Type": "application/json; charset=utf-8"}
     if POST_API_KEY:
         headers["Authorization"] = f"Bearer {POST_API_KEY}"
-
-    body = json.dumps(payload, ensure_ascii=False)
-    _post_log(f"→ POST {POST_URL}\nHeaders: {headers}\nBody: {body}")
-
+    body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+    req = urlrequest.Request(POST_URL, data=body, headers=headers, method="POST")
+    _post_log(f"→ POST {POST_URL}\nHeaders: {headers}\nBody: {body.decode('utf-8')}")
     try:
-        r = SESSION.post(POST_URL, headers=headers, data=body.encode("utf-8"), timeout=POST_TIMEOUT)
-        text = r.text
-        _post_log(f"← {r.status_code}\n{text}\n{'-'*60}")
-        ok = 200 <= r.status_code < 300
-        return {"ok": ok, "status": r.status_code, "response": text}
-    except Exception as e:
+        with urlrequest.urlopen(req, timeout=POST_TIMEOUT) as resp:
+            txt = resp.read().decode("utf-8", errors="replace")
+            code = resp.getcode()
+            _post_log(f"← {code}\n{txt}\n{'-'*60}")
+            return {"ok": 200 <= code < 300, "status": code, "response": txt}
+    except HTTPError as e:
+        txt = e.read().decode("utf-8", errors="replace")
+        _post_log(f"← {e.code}\n{txt}\n{'-'*60}")
+        return {"ok": False, "status": e.code, "response": txt}
+    except URLError as e:
         _post_log(f"× ERROR: {e}\n{'-'*60}")
         return {"ok": False, "status": None, "error": str(e)}
 
@@ -683,39 +588,27 @@ def safe_cell(val):
         return json.dumps(val, ensure_ascii=False)
     return "" if val is None else val
 
-def _clean_url(u: str) -> str:
-    if not isinstance(u, str):
-        return ""
-    u = u.strip().replace("\t", " ")
-    u = u.replace("\ufeff", "").replace("\xa0", "")
-    u = u.strip("\r\n ")
-    return u
-
 def main():
-    logger.info(f"POST_URL={POST_URL}  OUTPUT_FILE={OUTPUT_FILE}")
+    logger.info("Playwright-only (simple). BASE_DIR=%s", BASE_DIR)
+    logger.info("Python=%s", sys.version.replace("\n", " "))
+    logger.info("POST_URL=%s  OUTPUT_FILE=%s", POST_URL, OUTPUT_FILE)
+    logger.info("HEADLESS=%s  UA=%s  Accept-Language=%s", HEADLESS, UA, ACCEPT_LANG)
+
     SAVE_JSON = ask_bool("SAVE_JSON (сохранять JSON на диск?)", "1")
     SEND_JSON = ask_bool("SEND_JSON (отправлять на API?)", "1")
 
-    # читаем ссылки (utf-8-sig для BOM)
-    with open(INPUT_FILE, "r", encoding="utf-8-sig", newline="") as f:
-        raw_lines = f.readlines()
-    links = [_clean_url(x) for x in raw_lines if _clean_url(x)]
-    logger.info(f"Всего ссылок: {len(links)}")
-    if not links:
-        logger.warning("Список ссылок пуст — проверьте product_links.txt")
+    with open(INPUT_FILE, "r", encoding="utf-8") as f:
+        links = [line.strip() for line in f if line.strip()]
+    print(f"Всего ссылок: {len(links)}")
 
-    # готовим Excel
     wb = Workbook()
     ws = wb.active
     ws.title = "IKEA Products"
     ws.append(KEEP_COLUMNS)
 
-    # батч для JSON/API
     batch_items = []
     batch_index = 1
 
-    STATUS_COUNTER = Counter()
-
     def flush_batch():
         nonlocal batch_items, batch_index
         if not batch_items:
@@ -726,68 +619,119 @@ def main():
         if SEND_JSON:
             res = post_payload(payload)
             ok = res.get("ok")
-            logger.info(f"POST batch {batch_index}: {'OK' if ok else 'FAIL'} (status={res.get('status')})")
+            print(f"POST batch {batch_index}: {'OK' if ok else 'FAIL'} (status={res.get('status')})")
+            logger.info("POST batch %d: %s (status=%s)", batch_index, "OK" if ok else "FAIL", res.get("status"))
         batch_index += 1
         batch_items = []
 
-    for idx, link in enumerate(links, 1):
-        logger.info(f"[{idx}/{len(links)}] {link}")
-        force_dump = idx <= 3  # ← Принудительно сохраняем HTML для первых 3 ссылок
-        row = extract_data(link, force_dump=force_dump)
-
-        st = row.get("http_status")
-        if st is None and "error" in row:
-            STATUS_COUNTER["err"] += 1
-        else:
-            STATUS_COUNTER[str(st or 200)] += 1
-
-        # пишем в Excel
-        ws.append([safe_cell(row.get(col, "")) for col in KEEP_COLUMNS])
-
-        # ФИЛЬТРЫ для JSON/API
-        try:
-            price = float(row.get("buyModule.productPrice") or 0)
-        except Exception:
-            price = 0.0
-
-        try:
-            total_kg = float(row.get("total brutto") or 0)
-        except Exception:
-            total_kg = 0.0
-
-        details_json = row.get("productInformationSection.productDetailsProps") or {}
-
-        if not (20 <= price <= 1500):
-            logger.debug(f"Skip by price: {price}")
-        elif total_kg > 30:
-            logger.debug(f"Skip by weight: {total_kg} kg")
-        elif materials_match_exclusions(details_json, EXCLUSIONS):
-            logger.debug("Skip by exclusions (materials)")
-        else:
+    pw, ctx, page = open_browser()
+    try:
+        for idx, url in enumerate(links, 1):
+            print(f"[{idx}/{len(links)}] {url}")
             try:
-                item = build_variant(row)
+                full_html, raw = fetch_page(page, url, idx)
+            except Exception:
+                logger.exception("Fetch error for %s", url)
+                continue
+
+            row = parse_page(url, full_html, raw)
+            if row.get("error"):
+                logger.warning("Extract error [%d] %s: %s", idx, url, row["error"])
+
+            # Excel
+            ws.append([safe_cell(row.get(col, "")) for col in KEEP_COLUMNS])
+
+            # Фильтры для JSON/API
+            try:
+                price = float(row.get("buyModule.productPrice") or 0)
+            except Exception:
+                price = 0.0
+            try:
+                total_kg = float(row.get("total brutto") or 0)
+            except Exception:
+                total_kg = 0.0
+            details_json = row.get("productInformationSection.productDetailsProps") or {}
+
+            if not (20 <= price <= 1500):
+                pass
+            elif total_kg > 30:
+                pass
+            elif materials_match_exclusions(details_json, EXCLUSIONS):
+                pass
+            else:
+                # build variant (минимально как раньше)
+                def _ceil_price(v):
+                    try: return int(math.ceil(float(v)))
+                    except: return None
+                def _ceil_int(v):
+                    try: return int(math.ceil(float(v)))
+                    except: return None
+
+                visible = row.get("productSummary.visibleItemNo") or ""
+                sku = visible.replace(" ", "")
+                csm = (row.get("prductVariantColorMeasure") or "").strip()
+                color, size = "", ""
+                if csm:
+                    parts = [p.strip() for p in csm.split(",", 1)]
+                    if len(parts) == 2:
+                        color, size = parts[0], parts[1]
+                    else:
+                        size = parts[0]
+                if not color and not size:
+                    size = (row.get("pipPricePackage.measurementText") or "").strip()
+
+                cost = _ceil_price(row.get("buyModule.productPrice"))
+                name = row.get("originalName") or row.get("buyModule.productName") or ""
+                desc_html = row.get("productInformationSection.productDetailsProps_formatted_html") or ""
+                composition_html = row.get("productInformationSection.dimensionProps_formatted_html_translated") or ""
+                imgs = []
+                raw_imgs = row.get("productGallery.urls") or ""
+                if isinstance(raw_imgs, str):
+                    imgs = [x for x in raw_imgs.split("\n") if x.strip()]
+                in_stock = bool(row.get("availabilityGroup.serverOnlineSellable")) or bool(row.get("buyModule.onlineSellable"))
+                weight_kg = _ceil_int(row.get("total brutto"))
+
+                item = {
+                    "category": {"name": "TEST/IKEA"},
+                    "brand":    {"name": "ikea"},
+                    "variant":  {
+                        "status_id": 1,
+                        "color": color.capitalize() if color else "none",
+                        "sku": sku,
+                        "size": size,
+                        "cost": cost,
+                        "originalUrl": url,
+                        "originalName": name,
+                        "originalDescription": desc_html,
+                        "originalComposition": composition_html,
+                        "images": imgs,
+                        "inStock": in_stock,
+                        "weight": weight_kg if weight_kg is not None else 0,
+                    },
+                }
                 batch_items.append(item)
-            except Exception as e:
-                logger.error(f"build_variant error for {link}: {e}")
-                _post_log(f"× build_variant error for {link}: {e}")
 
-        # авто-сейв Excel каждые 50 строк
-        if idx % 50 == 0:
-            wb.save(OUTPUT_FILE)
-            logger.info(f"💾 autosave: {OUTPUT_FILE}")
+            if idx % 50 == 0:
+                wb.save(OUTPUT_FILE)
+                print(f"💾 autosave: {OUTPUT_FILE}")
 
-        # флаш батча при достижении лимита
-        if len(batch_items) >= BATCH_SIZE:
-            flush_batch()
+            if len(batch_items) >= BATCH_SIZE:
+                flush_batch()
 
-    # финал
-    wb.save(OUTPUT_FILE)
-    logger.info(f"\n✅ Excel готов: {OUTPUT_FILE}")
+        wb.save(OUTPUT_FILE)
+        print(f"\n✅ Excel готов: {OUTPUT_FILE}")
+        flush_batch()
 
-    flush_batch()
-
-    logger.info(f"HTTP stats: {dict(STATUS_COUNTER)}")
-    logger.info("🎯 Готово.")
+    finally:
+        close_browser(pw, ctx)
+        logger.info("Playwright closed. Bye.")
 
 if __name__ == "__main__":
-    main()
+    try:
+        main()
+    except KeyboardInterrupt:
+        print("\nInterrupted by user.")
+        logger.warning("Interrupted by user")
+    except Exception:
+        logger.exception("Fatal error")
+        raise

Парсер_IKEA/playwright_profile/ChromeFeatureState

@@ -0,0 +1 @@
+{"disable-features":"AcceptCHFrame,AutoExpandDetailsElement,AvoidUnnecessaryBeforeUnloadCheckSync,CertificateTransparencyComponentUpdater,DestroyProfileOnBrowserClose,DialMediaRouteProvider,GlobalMediaControls,HttpsUpgrades,ImprovedCookieControls,LazyFrameLoading,MediaRouter,PaintHolding,PlzDedicatedWorker,Translate","enable-features":"UkmSamplingRate\u003CUkmSamplingRate","force-fieldtrial-params":"UkmSamplingRate.Sampled_NoSeed_Stable:_default_sampling/1000000","force-fieldtrials":"*SeedFileTrial/Default/UkmSamplingRate/Sampled_NoSeed_Stable"}

Парсер_IKEA/playwright_profile/Default/BookmarkMergedSurfaceOrdering

@@ -0,0 +1,2 @@
+{
+}