code/2.py

import requests, sys, json, lzma, os, time
from bs4 import BeautifulSoup # pip install beautifulsoup4
import trafilatura # pip install trafilatura
from multiprocessing import Process
from multiprocessing.pool import Pool

DOMAIN = "https://thichtruyen.vn"

stories = []
with lzma.open("3-stories.jsonl.xz", "rt") as fin:
    for idx, line in enumerate(fin):
        data = json.loads(line)
        data["_id"] = idx
        stories.append(data)
print(len(stories), flush=True)


def get_content(url):
    print("get_content", url, flush=True)
    r = requests.get(url)
    soup = BeautifulSoup(r.content, "html.parser")
    try:
        content = soup.select(".story-detail-content")[0]
        content = trafilatura.extract(str(content), output_format='txt')
        return content
    except:
        with open("ERRORS.txt", "at") as f: f.write(url + "\n")
        return None

def crawl(story):
    story_id = story["_id"]
    chapters_file = f"stories/{story_id}_chapters.jsonl.xz"
    new_chapters_file = f"stories/{story_id}_chapters.jsonl"

    crawled_urls = set()
    if os.path.exists(chapters_file):
        # Chuyển giao giữa xz và jsonl
        try:
            with open(new_chapters_file, "wt") as fout:
                with lzma.open(chapters_file, "rt") as fin:
                    for line in fin:
                        data = json.loads(line)
                        if len(data["content"]) <= 200: continue
                        url = data["url"]
                        if url not in crawled_urls:
                            crawled_urls.add(url)
                            fout.write(line)
        except: # có lỗi đọc file thì xóa đi crawl từ đầu :)
            crawled_urls = set()
            os.remove(chapters_file)
            os.remove(new_chapters_file)
        # đọc xong thì xóa
        os.remove(chapters_file)

    elif os.path.exists(new_chapters_file):
        with open(new_chapters_file, "rt") as fin:
            for line in fin:
                data = json.loads(line)
                if len(data["content"]) <= 200: continue
                url = data["url"]
                if url not in crawled_urls:
                    crawled_urls.add(url)

    chapters = story["chapters"]
    problems = 0
 
    for chapter in chapters:
        url = chapter["url"]
        if url not in crawled_urls:
            crawled_urls.add(url)
            content = get_content(DOMAIN + url)
            if content and len(content) > 200:
                chapter["content"] = content
                with open(new_chapters_file, "at") as fout:
                    fout.write(json.dumps(chapter, ensure_ascii=False) + "\n")
            else:
                problems += 1

    if problems > len(chapters) // 4:
        with open("PROBLEMS.txt", "at") as f:
            f.write(story["url"] + "\n")

count = 0
with Pool() as pool:
    for _ in pool.imap_unordered(crawl, stories):
        count += 1
        print(count, "/", len(stories), flush=True)