# -*- coding: utf-8 -*- import os import sys import json import re from playwright.sync_api import sync_playwright CURRENT_DIR = os.path.dirname(os.path.abspath(__file__)) ROOT_DIR = os.path.dirname(CURRENT_DIR) if ROOT_DIR not in sys.path: sys.path.append(ROOT_DIR) USER_AGENT = ( "Mozilla/5.0 (Windows NT 10.0; Win64; x64) " "AppleWebKit/537.36 (KHTML, like Gecko) " "Chrome/136.0.0.0 Safari/537.36" ) class NaverSchoolDomLogger: def __init__(self, article_no, headless=False): self.article_no = str(article_no or "").strip() self.headless = headless self.playwright = None self.browser = None self.context = None self.page = None self.article_data = None def start(self): print("=" * 80) print("[BROWSER START]") self.playwright = sync_playwright().start() self.browser = self.playwright.chromium.launch( headless=self.headless, args=[ "--disable-blink-features=AutomationControlled", "--no-sandbox", ], ) self.context = self.browser.new_context( locale="ko-KR", viewport={ "width": 1365, "height": 900, }, user_agent=USER_AGENT, ) self.page = self.context.new_page() self.page.on("response", self.handle_response) print("[BROWSER READY]") def close(self): print("[BROWSER CLOSED]") try: if self.browser: self.browser.close() except Exception: pass try: if self.playwright: self.playwright.stop() except Exception: pass def build_url(self): return ( "https://new.land.naver.com/complexes" f"?articleNo={self.article_no}" ) def is_json_response(self, response): try: content_type = response.headers.get("content-type", "") return "json" in content_type.lower() except Exception: return False def handle_response(self, response): try: url = str(response.url).lower() if not self.is_json_response(response): return if ( f"/api/articles/{self.article_no}" in url or f"articles/{self.article_no}" in url ): self.article_data = response.json() print("[ARTICLE API CAPTURED]", response.url) except Exception: pass def get_article_meta(self): if not self.article_data: print("[ARTICLE META] not captured") return {} detail = self.article_data.get("articleDetail") or {} meta = { "article_no": detail.get("articleNo") or self.article_no, "article_name": detail.get("articleName"), "hscp_no": detail.get("hscpNo"), "latitude": detail.get("latitude"), "longitude": detail.get("longitude"), "cortar_no": detail.get("cortarNo"), "realtor_id": detail.get("realtorId"), "detail_description": detail.get("detailDescription") or "", "feature_description": detail.get("articleFeatureDescription") or "", } print("=" * 80) print("[ARTICLE META]") for key, value in meta.items(): if key in ["detail_description", "feature_description"]: value = str(value or "")[:250] print(f"{key}: {value}") print("=" * 80) return meta def click_school_related(self): candidates = [ "text=학군", "text=학교", "text=교육", "text=주변", "text=주변시설", "button:has-text('학군')", "button:has-text('학교')", "button:has-text('교육')", "button:has-text('주변')", "button:has-text('주변시설')", "a:has-text('학군')", "a:has-text('학교')", "a:has-text('교육')", "a:has-text('주변')", "a:has-text('주변시설')", ] print("=" * 80) print("[SCHOOL CLICK START]") for selector in candidates: try: loc = self.page.locator(selector) count = loc.count() if count <= 0: continue print("[CLICK TRY]", selector, "count=", count) loc.first.click( force=True, timeout=2500, ) self.page.wait_for_timeout(2500) self.dump_body_text( label=self.safe_label(selector) ) except Exception as e: print("[CLICK FAIL]", selector, str(e)[:160]) print("[SCHOOL CLICK END]") print("=" * 80) def safe_label(self, text): text = str(text or "") text = re.sub(r"[^0-9A-Za-z가-힣_]+", "_", text) return text[:50] or "dump" def dump_body_text(self, label="body"): try: text = self.page.locator("body").inner_text(timeout=5000) except Exception as e: print("[BODY TEXT FAIL]", str(e)[:160]) return "" out_dir = os.path.join(ROOT_DIR, "storage", "debug") os.makedirs(out_dir, exist_ok=True) file_path = os.path.join( out_dir, f"school_dom_{self.article_no}_{label}.txt" ) with open(file_path, "w", encoding="utf-8") as f: f.write(text) print("[BODY TEXT SAVED]", file_path) self.print_school_lines(text) return text def print_school_lines(self, text): lines = [] keywords = [ "초등학교", "중학교", "고등학교", "학교", "학군", "교육", "초,중,고", "초·중·고", "초중고", ] for line in str(text or "").splitlines(): line = line.strip() if not line: continue if any(keyword in line for keyword in keywords): lines.append(line) print("=" * 80) print("[SCHOOL RELATED LINES]", len(lines)) for line in lines[:120]: print(line) print("=" * 80) def extract_school_candidates_from_text(self, text): result = [] # 명확한 학교명만 추출 patterns = [ (r"([가-힣A-Za-z0-9]{2,30}초등학교)", "초등학교", "elementary"), (r"([가-힣A-Za-z0-9]{2,30}중학교)", "중학교", "middle"), (r"([가-힣A-Za-z0-9]{2,30}고등학교)", "고등학교", "high"), ] seen = set() for pattern, school_type, level in patterns: for name in re.findall(pattern, text or ""): name = str(name or "").strip() if not name: continue if name in seen: continue seen.add(name) result.append({ "school_name": name, "school_type": school_type, "school_level": level, "source": "dom", }) return result def print_dom_school_candidates(self): try: text = self.page.locator("body").inner_text(timeout=5000) except Exception as e: print("[DOM CANDIDATE FAIL]", str(e)[:160]) return [] candidates = self.extract_school_candidates_from_text(text) print("=" * 80) print("[DOM SCHOOL CANDIDATES]", len(candidates)) for item in candidates: print( item.get("school_name"), item.get("school_type"), item.get("school_level"), ) print("=" * 80) return candidates def run(self): if not self.article_no: raise ValueError("article_no is required") self.start() try: url = self.build_url() print("[OPEN]", url) self.page.goto( url, wait_until="commit", timeout=30000, ) self.page.wait_for_timeout(7000) self.get_article_meta() # 초기 화면 DOM 덤프 self.dump_body_text(label="initial") # 학군/학교/교육/주변시설 관련 UI 클릭 후 DOM 덤프 self.click_school_related() # 최종 DOM에서 학교명 후보 추출 self.print_dom_school_candidates() finally: self.close() def main(): import argparse parser = argparse.ArgumentParser() parser.add_argument( "--article-no", type=str, default="2627920510", ) parser.add_argument( "--headless", action="store_true", help="headless mode", ) args = parser.parse_args() logger = NaverSchoolDomLogger( article_no=args.article_no, headless=args.headless, ) logger.run() if __name__ == "__main__": main()