python爬取網(wǎng)頁(yè)版QQ空間,生成各類(lèi)圖表
https://github.com/kuishou68/python
各類(lèi)圖表的實(shí)現(xiàn)效果
按照你的谷歌瀏覽器下載指定版本的驅(qū)動(dòng) http://chromedriver.storage.googleapis.com/index.html
驅(qū)動(dòng)跟兩個(gè)python腳本放入同目錄，我的版本是90.0.4430的，查看你自己的版本，下載后把我的chromedriver.exe替換掉！
這里用到了很多第三方包，鼠標(biāo)放在報(bào)紅的包名下，用Alt+Enter導(dǎo)包，如果失敗則在控制臺(tái)用下面的必殺技
pip install 包名 -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
qq空間txt.py
import timefrom selenium import webdriverfrom lxml import etree# 這里一定要設(shè)置編碼格式，防止后面寫(xiě)入文件時(shí)報(bào)錯(cuò)friend = ’1569339843’ # 朋友的QQ號(hào)，朋友的空間要求允許你能訪問(wèn)user = ’783533896’ # 你的QQ號(hào)pw = ’1323mkoNJI.@’ # 你的QQ密碼# 獲取瀏覽器驅(qū)動(dòng)chrome_driver = ’chromedriver.exe’driver = webdriver.Chrome(executable_path=chrome_driver)# 瀏覽器窗口最大化driver.maximize_window()# 瀏覽器地址定向?yàn)閝q登陸頁(yè)面driver.get('http://i.qq.com')# 所以這里需要選中一下frame，否則找不到下面需要的網(wǎng)頁(yè)元素driver.switch_to.frame('login_frame')time.sleep(3)# 自動(dòng)點(diǎn)擊賬號(hào)登陸方式driver.find_element_by_id('switcher_plogin').click()time.sleep(3)# 賬號(hào)輸入框輸入已知qq賬號(hào)driver.find_element_by_id('u').send_keys(user)time.sleep(5)# 密碼框輸入已知密碼driver.find_element_by_id('p').send_keys(pw)time.sleep(5)# 自動(dòng)點(diǎn)擊登陸按鈕driver.find_element_by_id('login_button').click()time.sleep(5)# 讓webdriver操縱當(dāng)前頁(yè)driver.switch_to.default_content()time.sleep(5)# 跳到說(shuō)說(shuō)的url, friend你可以任意改成你想訪問(wèn)的空間driver.get('http://user.qzone.qq.com/' + friend + '/311')time.sleep(5)next_num = 0 # 初始“下一頁(yè)”的idwhile True: # 下拉滾動(dòng)條，使瀏覽器加載出動(dòng)態(tài)加載的內(nèi)容， # 我這里是從1開(kāi)始到6結(jié)束 分5 次加載完每頁(yè)數(shù)據(jù) for i in range(1, 6):height = 20000 * i # 每次滑動(dòng)20000像素strWord = 'window.scrollBy(0,' + str(height) + ')'driver.execute_script(strWord)time.sleep(4) # 很多時(shí)候網(wǎng)頁(yè)由多個(gè)<frame>或<iframe>組成，webdriver默認(rèn)定位的是最外層的frame， # 所以這里需要選中一下說(shuō)說(shuō)所在的frame，否則找不到下面需要的網(wǎng)頁(yè)元素 driver.switch_to.frame('app_canvas_frame') selector = etree.HTML(driver.page_source) divs = selector.xpath(’//*[@id='msgList']/li/div[3]’) # 這里使用 a 表示內(nèi)容可以連續(xù)不清空寫(xiě)入 with open(’qq_word.txt’, ’a’, encoding='utf-8') as f:for div in divs: qq_name = div.xpath(’./div[2]/a/text()’) qq_content = div.xpath(’./div[2]/pre/text()’) qq_time = div.xpath(’./div[4]/div[1]/span/a/text()’) qq_praise = div.xpath(’./div[4]/div[2]/span/span/a[2]/text()’) qq_comment = div.xpath(’./div[4]/div[2]/a[3]/text()’) qq_name = qq_name[0] if len(qq_name) > 0 else ’’ qq_content = qq_content[0] if len(qq_content) > 0 else ’’ qq_content = qq_content.replace(’n’, ’ ’) qq_time = qq_time[0] if len(qq_time) > 0 else ’’ qq_praise = qq_praise[0] if len(qq_praise) > 0 else ’’ qq_comment = qq_comment[0] if len(qq_comment) > 0 else ’’ print(qq_name, qq_time, qq_content, qq_praise, qq_comment) f.write(qq_content + 'n') # 當(dāng)已經(jīng)到了尾頁(yè)，“下一頁(yè)”這個(gè)按鈕就沒(méi)有id了，可以結(jié)束了 if driver.page_source.find(’pager_next_’ + str(next_num)) == -1:break # 找到“下一頁(yè)”的按鈕，因?yàn)橄乱豁?yè)的按鈕是動(dòng)態(tài)變化的，這里需要?jiǎng)討B(tài)記錄一下 driver.find_element_by_id(’pager_next_’ + str(next_num)).click() # “下一頁(yè)”的id next_num += 1 # 因?yàn)樵谙乱粋€(gè)循環(huán)里首先還要把頁(yè)面下拉，所以要跳到外層的frame上 driver.switch_to.parent_frame()# 關(guān)閉瀏覽器driver.quit()
各種圖表的生成
import pandas as pdfrom pyecharts.charts import Barfrom pyecharts.charts import Pieimport pyecharts.options as optsfrom pyecharts.charts import Lineimport redf_excel = pd.read_excel(’qq_excel.xlsx’) # 默認(rèn)讀取sheet=0 Pandas DataFrame’def getTimeStr(row): item = row[’時(shí)間’] if pd.isnull(item) | pd.isna(item):return data = item.split(’年’)[0] return data# 按年統(tǒng)計(jì)說(shuō)說(shuō)數(shù)量def readCount(result, row): timeData = getTimeStr(row) if timeData == None: return if timeData in result.keys():result[timeData] += 1 else:result[timeData] = 1# 按年統(tǒng)計(jì)說(shuō)說(shuō)點(diǎn)贊數(shù)def readThumb(result, row): item = row[’贊’] if pd.isnull(item):return # data = re.match(r’贊((d+).*’, item, re.M | re.I) if len(item.split('(')) <= 1:return data = item.split('(')[1].split(')')[0] timeData = getTimeStr(row) if timeData == None: return if timeData in result.keys():result[timeData] += int(data) else:result[timeData] = int(data)# 按年統(tǒng)計(jì)說(shuō)說(shuō)評(píng)論數(shù)def readComment(result, row): item = row[’評(píng)論’] if pd.isnull(item):return # data = re.match(r’贊((d+).*’, item, re.M | re.I) if len(item.split('(')) <= 1:return data = item.split('(')[1].split(')')[0] timeData = getTimeStr(row) if timeData == None: return if timeData in result.keys():result[timeData] += int(data) else:result[timeData] = int(data)def readExcel(df_excel): count = {} result = {} thumb = {} comment = {} for index, row in df_excel.iterrows():readCount(count, row)readThumb(thumb, row)readComment(comment, row) result[’count’] = count result[’thumb’] = thumb result[’comment’] = comment return resultdef getKeyAndVal(keyWord): data = readExcel(df_excel).get(keyWord) key = [] value = [] for item in data.keys():key.append(item)value.append(data[item]) key.reverse() value.reverse() return [key, value]# 統(tǒng)計(jì)每年發(fā)表說(shuō)說(shuō)次數(shù)柱狀圖def paintBar(): count = readExcel(df_excel).get(’count’) # V1 版本開(kāi)始支持鏈?zhǔn)秸{(diào)用 data = getKeyAndVal(’count’) print(data[0]) d = (Bar() .add_xaxis(data[0]) .add_yaxis('每年發(fā)表說(shuō)說(shuō)總數(shù)', data[1]) .render('每年發(fā)表說(shuō)說(shuō)總數(shù)柱狀圖.html') )paintBar()# 統(tǒng)計(jì)點(diǎn)贊和評(píng)論折線圖def paintLine(): commentData = getKeyAndVal(’comment’) thumbData = getKeyAndVal(’thumb’) xaxis_data = commentData[0] commentValue = commentData[1] thumbValue = thumbData[1] d = (Line() .add_xaxis(xaxis_data=xaxis_data) .add_yaxis('每年評(píng)論數(shù)', y_axis=commentValue) .add_yaxis('每年點(diǎn)贊數(shù)', y_axis=thumbValue) .render('每年點(diǎn)贊和評(píng)論折現(xiàn)圖.html') # 輸出圖形 )paintLine()
其他代碼自行下載項(xiàng)目查看
以上就是python爬取網(wǎng)頁(yè)版QQ空間,生成各類(lèi)圖表的詳細(xì)內(nèi)容，更多關(guān)于python 爬取QQ空間的資料請(qǐng)關(guān)注好吧啦網(wǎng)其它相關(guān)文章！
網(wǎng)公網(wǎng)安備