source/stats.py

"""Module to crawl the website 'https://www.stats.gov.cn' to fetch and process articles."""
import time
import urllib.request
import http.client
from datetime import datetime, timedelta
import random

from lxml import etree
from prefect import task, get_run_logger

from controllers.utils import crawl_by_url, encode

@task(name = "Data Collection - stats", log_prints = True)
def crawl(delta):
    """
    Crawls the website "https://www.stats.gov.cn/sj/sjjd/" and
    retrieves articles within a specified time range.

    Args:
        delta (int): The number of days to go back from the current date.

    Returns:
        None

    Raises:
        None
    """
    logger = get_run_logger()
    logger.info("stats.gov.hk")
    i = 0
    while i > -1:
        if i == 0:
            category_url = "https://www.stats.gov.cn/sj/sjjd/"
        else:
            category_url = f"https://www.stats.gov.cn/sj/sjjd/index_{i}.html"
        i = i + 1
        max_retries = 5
        backoff_factor = 2
        retries = max_retries
        while retries > 0:
            try:
                req = urllib.request.urlopen(category_url, timeout=60)
                text = req.read()
                html_text = text.decode("utf-8")
                page = etree.HTML(html_text)
                articlelist = page.xpath("//div[contains(@class, 'list-content')]/ul/li")
                break  # Success, exit retry loop
            except (urllib.error.URLError, http.client.IncompleteRead, TimeoutError) as error:
                logger.info(f"Network error: {error}. Retries left: {retries-1}")
                retries -= 1
                if retries > 0:
                    sleep_time = backoff_factor ** (max_retries - retries) + random.uniform(0, 1)
                    time.sleep(sleep_time)
                else:
                    logger.error(f"Failed to fetch {category_url} after {max_retries} attempts.")
                    articlelist = []  # Prevents UnboundLocalError
        for article in articlelist:
            if isinstance(article, etree._Element):
                subelement = etree.tostring(article).decode()
                subpage = etree.HTML(subelement)
                date = encode(subpage.xpath("//span"))
                parsed_datetime = datetime.strptime(
                        time.strftime("%Y-%m-%d", time.strptime(date, "%Y-%m-%d")),
                        "%Y-%m-%d")
                if parsed_datetime < (datetime.today() - timedelta(days=delta)):
                    i = -1
                else:
                    urls = subpage.xpath("//a[@class='fl pc_1600']/@href")
                    for url in urls:
                        try:
                            article = {}
                            url = url.replace('./', "https://www.stats.gov.cn/sj/sjjd/")
                            article['category'] = "Data Interpretation"
                            crawl_by_url(url, article)
                        except (urllib.error.URLError, etree.XMLSyntaxError) as error:
                            logger.info(error)