pixiv/PixivSearch/dao/Comment.py

import json
import threading
from concurrent import futures

import requests
from lxml import etree


class Comment:
    lock = threading.Lock()  # 多线程全局资源锁

    def __init__(self, keywords_=None) -> None:
        super().__init__()
        self.obj = {'data': {}, 'flag': False}
        self.keywords = keywords_

    # 获取番剧合集弹幕排行榜
    def getCommentSort(self, cids):

        urls = []
        for cid in cids:
            urls.extend(getCidUrls(cid))
        with futures.ThreadPoolExecutor(32) as executor:
            executor.map(self.count, urls)
        for index, data in enumerate(
                sorted(self.obj["data"].items(), key=lambda d: d[1], reverse=True)[
                :50]):
            print('{index}:{data}'.format(index=index+1, data=data))

    # 获取番剧合集弹幕排行榜
    def count(self, url, desc=None):
        bytes = requests.get(url).content
        comment_selector = etree.HTML(bytes)
        if not desc is None:
            print(desc)
        print("url=%s" % url)
        for comment in comment_selector.xpath('//i//d/text()'):
            if comment in self.obj["data"]:
                with self.lock:
                    self.obj["data"][comment] = self.obj["data"][comment] + 1
            else:
                with self.lock:
                    self.obj["data"][comment] = 1
            if not self.obj["flag"]:
                for keyword in self.keywords:
                    if keyword in comment:
                        self.obj["flag"] = True


# 根据cid获取历史弹幕地址
def getCidUrls(cid):
    urls = []
    url = "https://comment.bilibili.com/rolldate,%d" % cid
    req = requests.get(url)
    if len(req.text) > 0:
        for i in json.loads(req.text):
            urls.append("https://comment.bilibili.com/dmroll,%s,%d" % (i['timestamp'], cid))
    else:
        urls.append("https://comment.bilibili.com/%d.xml" % cid)
    return urls


# 下载历史弹幕
def parseXmlByHistory(path, cid,size=None):
    dlist = set()
    flag=parseXml(getCidUrls(cid), dlist, size)
    if size is None or not size is None and flag:
        f = open(path, 'wb')
        f.write(b'<?xml version="1.0" encoding="UTF-8"?><i>')
        for i in dlist:
            f.write(('\r\n' + i).encode())
        f.write(b'\r\n</i>')
        f.close()

def xml(url):
    bytes = requests.get(url).content
    return etree.HTML(bytes)

def parseXml(urls, dlist, size=None):
    if isinstance(urls, str):
        urls = [urls]
    if not size is None:
        size = float(size.strip('%')) / 100.0
    for url in urls:
        comment_selector = xml(url)
        list = comment_selector.xpath('//i//d/text()')
        maxlimit = int(comment_selector.xpath('//i//maxlimit/text()')[0])

        if len(list) > 0:
            print('弹幕数：{list},最大弹幕数：{maxlimit},弹幕池填充：{p}'.format(list=len(list), maxlimit=maxlimit,
                                                                 p='%.2f%%' % (len(list) / maxlimit * 100)))
            for element in comment_selector.xpath('//i//d'):
                if len(element.xpath("text()")) > 0:
                    fstr = '<d p="{p}">{content}</d>'.format(p=str(element.xpath("@p")[0]),
                                                         content=str(element.xpath("text()")[0]))
                    dlist.add(fstr)

                currentSize = len(dlist) / maxlimit
                print('填充率：{l}'.format(l='%.2f%%' % (currentSize * 100)))
                if not size is None and currentSize >= size:
                    return True
    return False

if __name__ == '__main__':
    # parseXmlByHistory('10815558.xml', 10815558)
    # Comment('').getCommentSort(episodeIdToCid(172095))
    cids=[653701,653702,653703,653704,653705,653738,653739,653740,653741,653742,653743,655269,655270,655271,655272,655273,655274,655275,655276,655277,655278,655279,1153426,1160531,1160532,1160533,1160534,1166828,3462691,3526913,4566294,4566295,4566296,4566297,4566298,4566299,4997067,4997068,4997069,4997070,4997071,4997072,4997073,4997074,4997075,4997076,4997077,4997078,4997079,4997080,4997081,4997082,4997083,4997084,4997085,4997086,4997087,4997088,4997089,5679996,7091113,7091114,7091115,7091116,7091117,7091118,7091119,7091120,7102322,7102323,7102324,7105422,7105423,7105424,7105425,7105426,7105427,7105428,7105429,7105430,7105431,7105432,7111387,7862786,7862787,7862788,7862789,7862790,7862791,7862792,7862793,7862794,7862795,7862796,8097054,8462197,8462198,8463198,8463856,8464066,8465726,8466012,8466263,8466715,8467897,8468074,8469339,8470303,8470792,8471514,8471910,8472168,8472542,8473404,8474331,8476130,8974866,9006609,9914921,9914922,9930188,9930189,9930190,9935817,9935818,9935819,9946753,10240252,10240253,10240254,10240255,10240256,10240257,10240258,10240259,10240260,10240261,10240262,12029055,12107619,13109117,13109118,13109119,13109120,13109121,13109122,13109123,13109124,13109125,13109126,13109127,13109128,13109129,13109130,13109131,13109132,13109133,13109134,13109135,13109136,13109137,13109138,13109139,13109140,13109141,13109142,13109143,13109144,13109145,13109146,13109147,13109148,13109149,13109150,13109151,13109152,13109153,13109154,13109155,13280053,13386513,13386514,13386515,13386516,13386517,13386518,13386519,13386520,13386521,13386522,13386523,13494289,14369143,14842278]

    cids=[6675884,6675885,6675886,6722497,6722498,6722499,6722500,6761162,6761163,6761164,6761165,7008191,7008192,7008193,7008194,7008195,7008196,7008197,7008198,7008199,7008200,7008201,7018471,7018472,7018473,7018474,7018475,7018476,7018477,7018478,7018479,7018480,7018481,7636499,7636500,7636501,7636502,7636503,7636504,7636505,7636506,7636507,7636508,7636509,9019798,9019799,9019800,9019801,9019802,9019803,9019804,9019805,9019806,9019807,9019808,9019809,9832709]
    for cid in cids:
        comment_selector=xml('https://comment.bilibili.com/{cid}.xml'.format(cid=cid))
        print('cid:{cid},弹幕数：{length}'.format(cid=cid,length=len(comment_selector.xpath('//i//d/text()'))))
        # parseXmlByHistory('{cid}.xml'.format(cid=cid),cid)

    # path = 'D:/QQ/1029559041/FileRecv/tmp001.xlsx'
    # ExcelFile = xlrd.open_workbook(path)
    #
    #
    # for sheetName in ExcelFile.sheet_names():
    #     sheet = ExcelFile.sheet_by_name(sheetName)
    #     for row in sheet.get_rows():
    #         try:
    #             row[5].value.index('日剧')
    #             cid = int(row[2].value)
    #             print('aid={aid},cid={cid},title={title}'.format(aid=int(row[1].value),cid=cid,title=row[5].value))
    #             parseXmlByHistory('{cid}.xml'.format(cid=cid),cid,'50%')
    #         except BaseException as e:
    #             repr(e)