#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
苏州市环保产业协会 - 第一次公示爬虫
CMS: 自定义ASP站
列表: div.ny_a_r > a[href^=shownews]（标题...日期）
分页: Swxx.asp?ClassID=2&page=N（6页，18条/页），第1页 Swxx.asp
详情: div.about_nr_jsaa + title标签
"""

import re, sys, json, time, requests, sqlite3
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://www.sz-epia.cn"
LIST_BASE = BASE_URL + "/Swxx.asp"
DB_PATH = "/root/search.db"
SITE_NAME = "苏州市环保产业协会-第一次公示"
CATEGORY = GROUP = "江苏苏州"

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
    "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
}
MAX_PAGES = 5

def init_db():
    conn = sqlite3.connect(DB_PATH, timeout=60)
    conn.execute("PRAGMA journal_mode=WAL")
    conn.execute("PRAGMA busy_timeout=5000")
    return conn

def get_soup(url):
    try:
        r = requests.get(url, headers=HEADERS, timeout=30)
        r.encoding = "utf-8"
        if r.status_code != 200:
            return None
        return BeautifulSoup(r.text, "html.parser")
    except:
        return None

def extract_list_items(soup):
    items = []
    d = soup.find("div", class_="ny_a_r")
    if not d:
        return items
    for a in d.find_all("a", href=True):
        href = a["href"].strip()
        if not href.startswith("shownews"):
            continue
        full = urljoin(BASE_URL, href)
        text = a.get_text(strip=True)
        # Format: ·　Title…YYYY/MM/DD
        # Remove leading · and spaces
        text = re.sub(r"^[·\s　]+", "", text)
        # Split title and date: last 10 chars should be YYYY/MM/DD
        date_str = ""
        m = re.search(r"(\d{4}/\d{1,2}/\d{1,2})$", text)
        if m:
            date_str = m.group(1).replace("/", "-")
            title = text[:m.start()].strip().rstrip("…")
        else:
            title = text
        if not title:
            continue
        items.append({"title": title, "url": full, "date": date_str})
    return items

def extract_detail(url):
    try:
        r = requests.get(url, headers=HEADERS, timeout=30)
        r.encoding = "utf-8"
        if r.status_code != 200:
            return None, None, None, None, None, None
    except Exception as e:
        print("  [ERROR] fetch: %s" % e, file=sys.stderr)
        return None, None, None, None, None, None

    soup = BeautifulSoup(r.text, "html.parser")
    html_str = r.text

    # Title from <title>
    title = ""
    title_tag = soup.find("title")
    if title_tag:
        t = title_tag.get_text(strip=True)
        t = re.sub(r"\s*[-–—]_?苏州市环保产业协会.*$", "", t).strip()
        if t:
            title = t

    # Date from page text: 发布时间：YYYY/MM/DD
    pub_date = ""
    m = re.search(r"发布时间[：:]\s*(\d{4}/\d{1,2}/\d{1,2})", html_str)
    if m:
        pub_date = m.group(1).replace("/", "-")

    src_url = ""

    # Content: div.about_nr_jsaa
    div = soup.find("div", class_="about_nr_jsaa")
    parts = []
    attach = []

    if div:
        for noise in div.find_all(["script", "style"]):
            noise.decompose()

        for child in div.find_all(["p", "table", "img"], recursive=True):
            if child.name == "table":
                rows = []
                for tr in child.find_all("tr"):
                    cells = [td.get_text(" ", strip=True) for td in tr.find_all(["td", "th"])]
                    if cells:
                        rows.append("| " + " | ".join(cells) + " |")
                if rows:
                    parts.append("\n".join(rows))
            elif child.name == "p":
                txt = child.get_text(" ", strip=True)
                if txt and len(txt) > 2:
                    parts.append(txt)
            elif child.name == "img":
                src = child.get("src", "")
                alt = child.get("alt", "")
                if src:
                    if not src.startswith("http"):
                        src = urljoin(url, src)
                    parts.append("![%s](%s)" % (alt or "", src))

        for a in div.find_all("a", href=True):
            h = a["href"].strip()
            if re.search(r"\.(pdf|doc|docx|xls|xlsx|zip|rar)$", h, re.I):
                fh = h if h.startswith("http") else urljoin(url, h)
                attach.append({"name": a.get_text(strip=True) or h.split("/")[-1], "url": fh})

    content = "\n\n".join(parts)
    if len(content.strip()) < 20:
        content = '<p><a href="%s">%s</a></p>' % (url, title or url.split('/')[-1])
    summary = content[:200] if len(content) > 200 else content
    return title, pub_date, src_url, content, summary, json.dumps(attach, ensure_ascii=False) if attach else ""

def main():
    import argparse
    parser = argparse.ArgumentParser()
    parser.add_argument("--max-pages", type=int, default=MAX_PAGES)
    parser.add_argument("--full", action="store_true")
    # Handle bare numeric arg for _MAX_PG
    import sys as _SYS2
    _MAX_PG = int(_SYS2.argv[-1]) if len(_SYS2.argv) > 1 and _SYS2.argv[-1].isdigit() else None
    if _MAX_PG is not None:
        print('[AutoPg] max_pages=' + str(_MAX_PG))
        _SYS2.argv.pop()
    args = parser.parse_args()
    max_pages = 6 if args.full else args.max_pages

    all_items, seen = [], set()

    for pn in range(1, max_pages + 1):
        if pn == 1:
            url = LIST_BASE
        else:
            url = LIST_BASE + "?ClassID=2&page=%d" % pn
        print("列表页 %d/%d: %s" % (pn, max_pages, url))
        soup = get_soup(url)
        if not soup:
            print("  FAILED")
            continue
        items = extract_list_items(soup)
        if not items:
            print("  空列表，停止")
            break
        n = 0
        for it in items:
            if it["url"] not in seen:
                seen.add(it["url"])
                all_items.append(it)
                n += 1
        print("  本页%d条，新增%d条，累计%d条" % (len(items), n, len(all_items)))
        if n == 0:
            break
        time.sleep(0.5)

    print("\n共%d篇文章，开始抓取详情..." % len(all_items))
    conn = init_db()
    ins, err = 0, 0
    SQL = "INSERT OR IGNORE INTO gov_raw (site_name, source_url, page_url, title, publish_date, summary, content, category, attachments, group_name) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"

    for idx, it in enumerate(all_items):
        try:
            print("  [%d/%d] %s..." % (idx + 1, len(all_items), it["title"][:40]))
            t, d, s, c, sm, a = extract_detail(it["url"])
            if t is None:
                err += 1
                continue
            conn.execute(SQL, (SITE_NAME, s or it["url"], it["url"], (t or it["title"]).strip(), d or it["date"],
                               sm or c[:200] if c else it["title"], c, CATEGORY, a, GROUP))
            conn.commit()
            ins += 1
            time.sleep(0.3)
        except Exception as e:
            print("  ERROR: %s" % e)
            err += 1
    conn.close()
    print("\n完成！新增%d条，错误%d条" % (ins, err))

if __name__ == "__main__":
    main()