为什么我只是从最后一个 url 中获取图片 url？

Question

我写了一个从网络漫画中提取链接图片的程序，但是，当我运行它时，它只是从最后一个链接章节中提取图像链接，而不是从所有章节中提取所有图像链接。我的程序有什么问题？ 我尝试了几种方法，但都没有用。

from PyQt5 import QtNetwork, QtCore
from requests_html import HTML
from functools import cached_property
from PyQt5.QtCore import QCoreApplication, QUrl

url1 = "https://saytruyen.net/truyen-su-tro-lai-cua-phap-su-hac-am-sau-66666-nam.html"

class Manager:
    def __init__(self):
        self.manager.finished.connect(self.handle_response)

    @cached_property
    def manager(self):
        return QtNetwork.QNetworkAccessManager()

    def start(self):
        self.start_request(QtCore.QUrl(url1))

    def start_request(self, url):
        request = QtNetwork.QNetworkRequest(url)
        self.manager.get(request)

    def handle_response(self, reply):
        err = reply.error()
        if err == QtNetwork.QNetworkReply.NoError:
            self.process(str(reply.readAll(), 'utf-8'))
        else:
            print("Error occured: ", err)
            print(reply.errorString())

    def process(self, data):
        html = HTML(html=data)
        rs = html.find("#list-chapter a", first=False)
        for i in reversed(rs):
            url2 = "https://saytruyen.net/" + i.attrs["href"]
            #print(url2)
            #self.start_request(QtCore.QUrl(url2))
            req = QtNetwork.QNetworkRequest(QUrl(url2))

            self.nam = QtNetwork.QNetworkAccessManager()
            self.nam.finished.connect(self.handleResponse)
            self.nam.get(req)

    def handleResponse(self, reply):

        er = reply.error()

        if er == QtNetwork.QNetworkReply.NoError:

            bytes_string = reply.readAll()
            html2 = HTML(html = str(bytes_string, 'utf-8'))
            rs_c = html2.find("#lst_content img")
            for x in rs_c:
                img ="https://saytruyen.net/" + x.attrs['src']
                print(img)

        else:
            print("Error occured: ", er)
            print(reply.errorString())
        
        QCoreApplication.quit()

Answer 1

有两个问题：

1. 用于下载的 QNetworkAccessManager 正在不断重新创建；由于网络请求是异步的，它不会被立即处理，并且会在for循环的下一个循环中与网络管理器一起被销毁，因为它被覆盖了；结果就是前一个请求会被销毁，只有最后一个会“存活”；
2. 收到第一个回复后立即退出应用程序，阻止处理所有其他请求；

解决方案是在__init__ 中为下载过程创建一个单个管理器，并在收到所有请求后立即退出。

class Manager:
    def __init__(self):
        self.manager.finished.connect(self.handle_response)
        self.nam = QtNetwork.QNetworkAccessManager()
        self.nam.finished.connect(self.handleResponse)
        self.urls = set()

    # ...

    def process(self, data):
        html = HTML(html=data)
        rs = html.find("#list-chapter a", first=False)
        for i in reversed(rs):
            url2 = QUrl("https://saytruyen.net/" + i.attrs["href"])
            if url2 in self.urls:
                continue
            self.urls.add(url2)
            req = QtNetwork.QNetworkRequest(url2)
            self.nam.get(req)

    def handleResponse(self, reply):
        self.urls.discard(reply.url())
        er = reply.error()

        if er == QtNetwork.QNetworkReply.NoError:
            bytes_string = reply.readAll()
            html2 = HTML(html = str(bytes_string, 'utf-8'))
            rs_c = html2.find("#lst_content img")
            for x in rs_c:
                img ="https://saytruyen.net/" + x.attrs['src']
                print(img)

        else:
            print("Error occured: ", er)
            print(reply.errorString())

        if not self.urls:
            QCoreApplication.quit()

请注意，拥有一个单个网络管理器并根据排队的请求正确处理响应通常就足够了（而且更好），但对于像这样的简单情况，拥有两个管理器并不代表巨大问题。