Mar. 16, 2019

Python Websocket服务端

用Python做了个Websocket的服务端，虽然知道相关的库应该早就有了，但是还是想自己做出来。

当然也参考了不少网上的教程，琢磨了好久的Websocket协议的格式。

内含一个WebsocketServer类，可以实现握手、建立连接、心跳（ping）、传输数据、上下文管理器。可以接收任意类型的数据，发送字符串（稍微改一下也可以发送二进制数据），支持数据分片（传输和发送都可以）。

More...

Feb. 10, 2019

将歌词处理成电子书

后端的最后一部分。

# coding=utf-8
import json
import re
class Lyric2Book:
support_format = ['txt', 'html']
support_ts = ['parallel', 'chunk']
html_template = """
<section>
<h2 class='header'>%(header)s</h2>
<div class='info'>
<div class='album'>%(album)s</div>
<div class='artists'>%(artists)s</div>
</div>
<div class='content'>
%(lyrics)s
</div>
</section>
"""
html_frame = """<!DOCTYPE html>
<html lang="zh-cn">
<head>
<meta charset="utf-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta name="referrer" content="never" />
<title>%s</title>
</head>
<body>
%s
</body>
</html>
"""
lyrics_template = "<div class='content-%(ver)d' >%(content)s</div>"
txt_template = "%(header)s\n专辑：%(album)s\n作者：%(artists)s\n\n%(lyrics)s\n\n"
txt_frame = "%s\n\n%s"
def __init__(self, file_format='html', title='Lyrics', typesetting='parallel'):
self.title = title
if file_format in Lyric2Book.support_format:
self.format = file_format
else:
raise Exception('Unsupported format: %s.' % file_format)
self.res = ""
self.data = ""
if typesetting in Lyric2Book.support_ts:
self.ts = typesetting
else:
raise Exception('Unsupported typesetting: %s' % typesetting)
def chunk(self, lyrics):
predlyric = []
# 传入的是多个版本的歌词
for item in lyrics:
lines = []
if item is not None:
# 分行处理
for line in item.split('\n'):
line = re.sub('\(\d+,\d+\)', '', line)
time_tag = ''.join(re.findall('\[.*?\]', line))
line = [time_tag, line.strip(time_tag)]
lines.append(line)
predlyric.append(lines)
output_section = ''
if self.format == 'html':
i = 1
for item in predlyric:
output_section += "<div class='content-%d'>\n" % i
i += 1
for line in item:
output_section += "<p><span class='timetag'>%s</span>%s</p> \n" % (re.sub('[\[\]]', '-', line[0]), line[1])
output_section += "</div>"
elif self.format == 'txt':
for item in predlyric:
output_section += ""
for line in item:
format_tag = '-%s- ' % re.sub('[\[\]]', '-', line[0]) if line[0] else ''
output_section += format_tag + line[1] + '\n'
return output_section
def parallel(self, lyrics):
unpredlyric = []
for item in lyrics:
lines = {}
if item is not None:
for line in item.split('\n'):
line = re.sub('\(\d+,\d+\)', '', line)
time_tag = ''.join(re.findall('\[.*?\]', line))
lines[time_tag] = line.strip(time_tag)
unpredlyric.append(lines)
predlyric = []
for group_tag in list(unpredlyric[0].keys()):
predlyric += [[group_tag] + [i.get(group_tag, unpredlyric[0][group_tag]) for i in unpredlyric]]
dparallels = predlyric
output_section = ''
if self.format == 'html':
for item in dparallels:
output_section += "<div class='content'>\n"
dtimetag = item[0]
dcontent = item[1::]
item_div = "<div class='timetag'>\n<p>{timetag}</p>\n</div>\n<div class='single-lyric'>\n{content}</div>\n"
output_spar = ''.join("<p class='ver-%d'>%s</p>\n" % (c[0], c[1]) for c in enumerate(dcontent, 1))
output_section += item_div.format(timetag=re.sub('[\[\]]', ' - ', dtimetag), content=output_spar) + '</div>'
elif self.format == 'txt':
for item in dparallels:
dtimetag = item[0]
dcontent = item[1::]
item_div = "{timetag}\n{content}\n"
output_spar = ''.join("%s\n" % c[1] for c in enumerate(dcontent))
output_section += item_div.format(timetag=re.sub('[\[\]]', ' - ', dtimetag), content=output_spar)
return output_section
def doconv(self, sections):
last_album = ''
output = ''
for item in sections:
header = item['name']
album = item['album']
artists = ','.join(item['artists'])
a = item['lyric']
if a is not None:
ly_res = {}
lyrics = [a['0'], a['1'], a['2']]
if self.ts == 'parallel':
ly_res = self.parallel(lyrics)
elif self.ts == 'chunk':
ly_res = self.chunk(lyrics)
if self.format == 'html':
if last_album != album:
last_album = album
output += '<h1>%s</h1>' % album
output += Lyric2Book.html_template % {'header': header, 'album': album, 'artists': artists, 'lyrics': ly_res}
elif self.format == 'txt':
if last_album != album:
last_album = album
output += album + '\n\n'
output += Lyric2Book.txt_template % {'header': header, 'album': album, 'artists': artists, 'lyrics': ly_res}
self.res = output
def output(self):
filename = self.title + '.' + self.format
with open(filename, 'w+', encoding='utf-8') as f:
res = self.res
if self.format == 'html':
res = Lyric2Book.html_frame % (self.title, res)
elif self.format == 'txt':
res = Lyric2Book.txt_frame % (self.title, res)
f.write(res)
if __name__ == '__main__':
with open('bbc.txt', 'r', encoding='utf-8') as file:
t = Lyric2Book(file_format='thjxt', title='BBC Documentary', typesetting='parallel')
content = json.loads(file.read())
t.doconv(content['result'])
t.output()

Feb. 6, 2019

寻找孤独

2019年的春节只有5天，二十八放假，初三上课。

放假前几分钟，我在想我能够做什么做什么，把计划中的事，未做完的事，未计划的事统统完成，大部分都是课外的事。

我确实完成了一些，但有些我却总也走不下去。歌词工具从十月份开始有想法，十一月份真正开始，到现在有三四个月了。借着这个想法我才真正熟悉了Python，后台逻辑其实一个月以前就结束了，一些改进的想法也决定搁置，没有再重构；前端最关键的部分也测试证明可行，然后我就不走了。

More...

Nov. 11, 2018

Python小工具：批量下载网易云音乐歌单/专辑中的歌词

最近想把网易云音乐中的英语听力下载到播放器中，方便随时听。但是只下载音频的话就少了些什么，最好连听力的歌词也下载进去。于是乎写了个小工具，可以批量下载歌单或者专辑中的音乐的歌词，可以省很多时间。

netease-lyric.py

#coding=utf-8
import requests
from bs4 import BeautifulSoup
import json
import re,os
from spider import *
def get_html(url):
headers = {
'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.77 Safari/537.36',
'Referer':'http://music.163.com',
'Host':'music.163.com'
}
try:
response = requests.get(url,headers=headers)
html = response.text
return html
except:
print('request error')
pass
def get_lyric(song_id):
url = 'http://music.163.com/api/song/lyric?' + 'id=' + str(song_id) + '&lv=1&kv=1&tv=-1'
html = get_html(url)
json_obj = json.loads(html)
print(json_obj)
lyric = json_obj['lrc']['lyric']
return lyric
def get_details(song_id):
url = 'http://music.163.com/api/song/detail/?' + 'id=' + str(song_id) + '&ids=%5B' + str(song_id) + '%5D'
html = get_html(url)
print(html)
json_obj = json.loads(html)
song_name = json_obj['songs'][0]['name']
song_artist = json_obj['songs'][0]['artists'][0]['name']
return song_name,song_artist
def output_file(song_id):
song_name,song_artist = get_details(song_id)
lyric = get_lyric(song_id)
file_name = song_name + ' - ' + song_artist + '.lrc'
#替换歌曲名中的斜杠
file_name = file_name.replace('/','／')
file = open(file_name,"w+",encoding="utf-8")
file.writelines(lyric)
file.close()
url = "https://music.163.com/album?id=2817001"
id_list = get_songlist(url)
print(id_list)
for iterm in id_list:
output_file(iterm)

抓取歌单中的歌曲信息，获得id spider.py

More...

Oct. 20, 2018

我与Esport S8四分之一决赛后【有的人不配赢】

首先恭喜IG击败LCK头号种子KT，打入四强。

然后，让我们用最饱满的热情恭喜夺冠热门RNG喜提八强，提前回家。

怎么说呢，IG赢KT我的确想到过，但是没有想到IG怎么做才能打赢KT。我曾经想过如果IG能赢KT我要吹IG一辈子，因为今天不仅是S8的四分之一决赛日，也是我的18岁生日。那么现在就是我兑现诺言的时候了。

More...

Oct. 3, 2018

使用Calibre制作书籍目录 Kindle上阅读是章节名始终固定在第一个的问题

最近遇到了几首很好听的歌曲，为了方便，就打算将歌词编辑出来放到Kindle。大概的步骤是先在Word上排好。然后用Calibre转换成Kindle能识别的格式。最初效果很好，抱着Kindle唱歌感觉非常有B格。后来同学也让我查几首歌的歌词放上去，这样歌曲从原来的两三首增加到八首，这样一页一页的翻找就比较费力，于是乎我用Calibre编辑功能为它自动生成了目录，一切和想象中的一样顺利，但是选择目录时却遇到了一件尴尬的事：

More...

Oct. 1, 2018