#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
周口市生态环境局 - 环境影响评价公示 (sthb.zhoukou.gov.cn)
CMS: HTML-Import CMS (script.json pagination)
列表: list{N}.html (N=1..33, 10条/页)
详情: article{32位hex}.html
详情容器: div.article-detail#articleDetail
注意: IP 221.14.138.84 + Host header, ArticleTitle可能重复
"""

import sys, os, re, time, json
from datetime import datetime, timezone, timedelta

sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from crawler_lib import push_to_searchdb

import requests
from bs4 import BeautifulSoup

# === 自动页数控制 ===
_MAX_PG = None
for i, a in enumerate(sys.argv):
    if a == '--pages' and i + 1 < len(sys.argv):
        _MAX_PG = int(sys.argv[i + 1])
        break
if _MAX_PG is not None:
    print("[AutoPg] max_pages=" + str(_MAX_PG))

SITE_NAME = "周口市生态环境局-环境影响评价公示"
BASE_IP = "221.14.138.84"
DOMAIN = "sthb.zhoukou.gov.cn"
BASE_URL = f"http://{BASE_IP}"
LIST_PATH = "/sitesources/sthjj/page_pc/hjgs/hjyxpjgs"
TOTAL_PAGES = 33
CUTOFF = (datetime.now(timezone.utc) - timedelta(days=365*3)).strftime("%Y-%m-%d")

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36",
    "Host": DOMAIN,
}


def html_table_to_html(table, base_url=""):
    """保留 HTML 表格结构，仅将相对链接/图片转绝对 URL"""
    import urllib.parse
    tbl = BeautifulSoup(str(table), 'html.parser')
    for a in tbl.find_all('a'):
        href = a.get('href', '')
        if href and not href.startswith(('http', 'javascript', '#')):
            a['href'] = urllib.parse.urljoin(base_url, href) if base_url else href
    for img in tbl.find_all('img'):
        src = img.get('src', '')
        if src and not src.startswith(('http', '//', 'data:')):
            img['src'] = urllib.parse.urljoin(base_url, src) if base_url else src
    return str(tbl)
def fetch(url, timeout=20):
    for attempt in range(3):
        try:
            r = requests.get(url, headers=HEADERS, timeout=timeout)
            r.encoding = "utf-8"
            return r.text
        except Exception as e:
            if attempt < 2:
                time.sleep(2)
            else:
                print(f"  [WARN] {url}: {e}")
                return None


def make_list_url(page):
    return BASE_URL + LIST_PATH + f"/list{page}.html"


def parse_list_page(html):
    items = []
    for m in re.finditer(
        r'<a href="([^"]+)"[^>]*title="([^"]*)"[^>]*>'
        r'\s*<div class="pictext">\s*(.*?)\s*</div>'
        r'\s*<div class="artpub">\s*<font>(\d{4}-\d{2}-\d{2})</font>',
        html, re.DOTALL
    ):
        href = m.group(1).strip()
        title_attr = m.group(2).strip()
        title_text = re.sub(r'<[^>]+>', '', m.group(3)).strip()
        date = m.group(4).strip()
        
        title = title_attr or title_text
        
        if href.startswith("/"):
            href = "http://" + DOMAIN + href
        elif not href.startswith("http"):
            href = "http://" + DOMAIN + "/" + href
        
        if title and href:
            items.append((title, href, date))
    
    return items


def _clean_span_spaces(text):
    text = re.sub(r'(?<=[\u4e00-\u9fff])\s+(?=[\u4e00-\u9fff])', '', text)
    text = re.sub(r'(?<=[\u4e00-\u9fff])\s+(?=\d)', '', text)
    text = re.sub(r'(?<=\d)\s+(?=[\u4e00-\u9fff])', '', text)
    text = re.sub(r'\s+([，。、；：？！)】」》])', r'\1', text)
    text = re.sub(r'([（【「《])\s+', r'\1', text)
    text = re.sub(r'(?<=\d)\s+(?=\d)', '', text)
    return re.sub(r'\s+', ' ', text).strip()


def extract_content(detail_div, page_url):
    parts = []
    
    for child in detail_div.children:
        name = getattr(child, 'name', None)
        if not name:
            continue
        
        if name == 'table':
            tbl_html = html_table_to_html(child, page_url)
            if tbl_html:
                parts.append(tbl_html)
        elif name == 'p':
            if child.find_parent("table"):
                continue
            
            # 附件链接 (PDF等)
            for a_tag in child.find_all("a", href=True):
                href = a_tag["href"]
                if any(href.lower().endswith(e) for e in ('.pdf', '.doc', '.docx', '.xls', '.xlsx', '.zip', '.rar')):
                    a_text = a_tag.get_text(" ", strip=True)
                    if href.startswith("/"):
                        full_href = "http://" + DOMAIN + href
                    elif not href.startswith("http"):
                        full_href = "http://" + DOMAIN + "/" + href
                    else:
                        full_href = href
                    if a_text:
                        parts.append(f"[{a_text}]({full_href})\n\n")
            
            # 正文文本
            txt = child.get_text(" ", strip=True)
            txt = _clean_span_spaces(txt)
            txt = re.sub(r'&#\d+;', '', txt)
            if txt and len(txt) > 3:
                parts.append(txt)
                parts.append("\n\n")
    
    result = ''.join(parts).strip()
    result = re.sub(r'\n{4,}', '\n\n', result)
    return result


def parse_detail(url):
    html = fetch(url)
    if not html:
        return None, None, None
    
    soup = BeautifulSoup(html, "html.parser")
    
    # 标题 — ArticleTitle 可能重复, 去重
    title = ""
    meta = soup.find("meta", attrs={"name": "ArticleTitle"})
    if meta:
        raw = meta.get("content", "").strip()
        # 去重: 如果前半段=后半段, 取一半
        if len(raw) > 10:
            half = len(raw) // 2
            if raw[:half] == raw[half:]:
                raw = raw[:half]
        title = raw
    
    if not title:
        h1 = soup.find("h1")
        title = h1.get_text(strip=True) if h1 else ""
    if not title:
        title_tag = soup.find("title")
        if title_tag:
            t = title_tag.get_text(strip=True)
            title = re.sub(r'_周口市生态环境局.*', '', t).strip()
    
    # 日期
    date = ""
    meta_d = soup.find("meta", attrs={"name": "PubDate"})
    if meta_d:
        d = meta_d.get("content", "")
        m = re.search(r'(\d{4}-\d{2}-\d{2})', d)
        if m:
            date = m.group(1)
    if not date:
        info = soup.find("div", class_="cms-article-xinxi")
        if info:
            m = re.search(r'(\d{4}-\d{2}-\d{2})', info.get_text())
            if m:
                date = m.group(1)
    
    # 正文
    content = ""
    detail_div = soup.find("div", class_="article-detail", id="articleDetail")
    if detail_div:
        content = extract_content(detail_div, url)
    
    return title, content, date


def main():
    pages_to_fetch = min(TOTAL_PAGES, _MAX_PG) if _MAX_PG else TOTAL_PAGES
    
    print(f"=== {SITE_NAME} ===")
    print(f"  总页数: {TOTAL_PAGES}, 本次: {pages_to_fetch}")
    
    all_items = []
    for page in range(1, pages_to_fetch + 1):
        list_url = make_list_url(page)
        print(f"  第 {page}/{pages_to_fetch} 页: {list_url}")
        
        html = fetch(list_url)
        if not html:
            print(f"    无法获取, 停止")
            break
        
        items = parse_list_page(html)
        if not items:
            print(f"    空列表, 停止")
            break
        
        print(f"    找到 {len(items)} 条")
        all_items.extend(items)
        time.sleep(0.3)
    
    print(f"\n  列表总计: {len(all_items)} 条\n")
    
    if not all_items:
        print("  无数据")
        return
    
    results = []
    for idx, (title, url, date) in enumerate(all_items):
        print(f"  [{idx+1}/{len(all_items)}] {title[:40]}...", end=" ", flush=True)
        
        dt, content, dd = parse_detail(url)
        final_title = dt or title
        final_date = date or dd
        
        if not content or len(content.strip()) < 10:
            print("⚠ 空")
            continue
        
        summary = re.sub(r'<[^>]+>', '', content)[:200].strip()
        if not summary:
            summary = final_title[:200]
        
        results.append({
            "site_name": SITE_NAME,
            "title": final_title,
            "url": url,
            "content": content,
            "summary": summary,
            "pub_date": final_date,
        })
        print(f"✅ ({len(content)}字)")
        
        if (idx + 1) % 20 == 0:
            print(f"  进度: {idx+1}/{len(all_items)}, 已入库: {len(results)}")
    
    if results:
        print(f"\n  入库 {len(results)} 条")
        push_to_searchdb(results, "zhoukou_hjyxpj")
    
    print("完成")


if __name__ == "__main__":
    main()