code-image-backup-public/GetDanMu
1
0
Fork 0
mirror of https://github.com/xhlove/GetDanMu.git synced 2025-12-16 16:15:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

增加爱奇艺链接方式解析

Browse Source
This commit is contained in:
xhlove
2020-01-05 14:49:33 +08:00
parent 858fb65ede
commit 78083c2295
11 changed files with 239 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
GetDanMu.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 12:59:11
# 上次编辑时间 : 2020-01-04 20:14:39
# 创建日期: 2020-01-04 19:14:39
# 上次编辑时间: 2020-01-05 14:47:53
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
@@ -13,10 +13,7 @@ from argparse import ArgumentParser
from sites.qq import main as qq
from sites.iqiyi import main as iqiyi
from basic.ass import get_ass_head, check_font
from pfunc.dump_to_ass import write_lines_to_file
from methods.assbase import ASS
from methods.sameheight import SameHeight
from pfunc.cfunc import check_url_site
# -------------------------------------------
# 基本流程
@@ -32,7 +29,7 @@ def main():
parser = ArgumentParser(description="视频网站弹幕转换/下载工具，任何问题请联系vvtoolbox.dev@gmail.com")
parser.add_argument("-f", "--font", default="微软雅黑", help="指定输出字幕字体")
parser.add_argument("-fs", "--font-size", default=28, help="指定输出字幕字体大小")
parser.add_argument("-s", "--site", default="qq", help="指定网站")
parser.add_argument("-s", "--site", default="", help="指定网站")
parser.add_argument("-cid", "--cid", default="", help="下载cid对应视频的弹幕（腾讯视频合集）")
parser.add_argument("-vid", "--vid", default="", help="下载vid对应视频的弹幕，支持同时多个vid，需要用逗号隔开")
parser.add_argument("-aid", "--aid", default="", help="下载aid对应视频的弹幕（爱奇艺合集）")
@@ -41,20 +38,15 @@ def main():
parser.add_argument("-y", "--y", action="store_true", help="覆盖原有弹幕而不提示")
args = parser.parse_args()
# print(args.__dict__)
font_path, font_style_name = check_font(args.font)
ass_head = get_ass_head(font_style_name, args.font_size)
if args.url != "":
args.site = check_url_site(args.url)
if args.site == "":
args.site = input("请输入站点（qq/iqiyi/youku）：\n")
if args.site == "qq":
subtitles = qq(args)
if args.site == "iqiyi":
subtitles = iqiyi(args)
for file_path, comments in subtitles.items():
get_xy_obj = SameHeight("那就写这一句作为初始化测试吧！", font_path=font_path, font_size=args.font_size)
subtitle = ASS(file_path, get_xy_obj, font=font_style_name)
for comment in comments:
subtitle.create_new_line(comment)
write_lines_to_file(ass_head, subtitle.lines, file_path)
if __name__ == "__main__":
# 打包 --> pyinstaller GetDanMu.spec
main()

28
README.md
View File

@@ -1,2 +1,30 @@
<!--
* @作者: weimo
* @创建日期: 2020-01-04 18:45:58
* @上次编辑时间 : 2020-01-05 14:42:42
* @一个人的命运啊,当然要靠自我奋斗,但是...
-->
# GetDanMu
转换/下载各类视频的弹幕
## 网站支持
| Site | URL | 单集? | 合集? | 综艺合集? |
| :--: | :-- | :-----: | :-----: | :-----: |
| **腾讯视频** | <https://v.qq.com/> |✓|✓| |
| **爱奇艺** | <https://www.iqiyi.com/> |✓|✓|✓|
## 可能存在的问题
- 下载进度接近100%时暂时没有反应
这是因为在全部弹幕获取完后一次性处理所致，对于时间过长和弹幕过多的视频，处理耗时较多，属于正常现象。
- 命令组合未达到预期效果
当前的逻辑并不完善，如果出现这种现象请反馈给我。
# 更新日志
## 2020/1/5
- 增加了通过链接下载爱奇艺视频弹幕的方法，支持综艺合集。
- 增加通过链接判断网站

6
basic/ass.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 13:05:23
# 上次编辑时间 : 2020-01-04 15:52:11
# 创建日期: 2020-01-04 19:14:46
# 上次编辑时间: 2020-01-05 14:45:55
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''

9
basic/vars.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 13:16:18
# 上次编辑时间 : 2020-01-04 16:08:34
# 创建日期: 2020-01-04 19:14:35
# 上次编辑时间: 2020-01-05 14:46:15
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
qqlive = {
@@ -12,6 +12,9 @@ qqlive = {
iqiyiplayer = {
"User-Agent":"Qiyi List Client PC 7.2.102.1343"
}
chrome = {
"User-Agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.88 Safari/537.36"
}
fonts = {
"微软雅黑":"msyh.ttc",
"微软雅黑粗体":"msyhbd.ttc",

6
methods/assbase.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 13:01:04
# 上次编辑时间 : 2020-01-04 15:42:02
# 创建日期: 2020-01-04 19:14:32
# 上次编辑时间: 2020-01-05 14:46:27
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''

7
methods/sameheight.py
View File

@@ -1,12 +1,13 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2019-12-25 20:35:43
# 上次编辑时间 : 2019-12-25 23:23:32
# 创建日期: 2020-01-04 19:14:47
# 上次编辑时间: 2020-01-05 14:46:51
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
from PIL.ImageFont import truetype
class SameHeight(object):

24
pfunc/cfunc.py Normal file
View File

@@ -0,0 +1,24 @@
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-05 12:45:18
# 上次编辑时间: 2020-01-05 14:44:42
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
from urllib.parse import urlparse
def check_url_site(url):
return urlparse(url).netloc.split(".")[-2]
def check_url_locale(url):
flag = {
"cn":"zh_cn",
"tw":"zh_tw",
"intl":"intl"
}
if urlparse(url).netloc.split(".")[0] == "tw":
return flag["tw"]
else:
return flag["cn"]

19
pfunc/dump_to_ass.py
View File

@@ -1,14 +1,27 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 19:17:44
# 上次编辑时间 : 2020-01-04 19:30:24
# 上次编辑时间: 2020-01-05 14:45:03
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
import os
from basic.ass import get_ass_head, check_font
from methods.assbase import ASS
from methods.sameheight import SameHeight
def write_one_video_subtitles(file_path, comments, args):
# 对于合集则每次都都得检查一次 也可以放在上一级 放在这里 考虑后面可能特殊指定字体的情况
font_path, font_style_name = check_font(args.font)
ass_head = get_ass_head(font_style_name, args.font_size)
get_xy_obj = SameHeight("那就写这一句作为初始化测试吧！", font_path=font_path, font_size=args.font_size)
subtitle = ASS(file_path, get_xy_obj, font=font_style_name)
for comment in comments:
subtitle.create_new_line(comment)
write_lines_to_file(ass_head, subtitle.lines, file_path)
def write_lines_to_file(ass_head, lines, file_path):
with open(file_path, "a+", encoding="utf-8") as f:
f.write(ass_head + "\n")

134
pfunc/request_info.py
View File

@@ -1,17 +1,18 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2020-01-04 13:15:25
# 上次编辑时间 : 2020-01-04 17:47:16
# 创建日期: 2020-01-04 19:14:43
# 上次编辑时间: 2020-01-05 14:47:16
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
import re
import json
import requests
from basic.vars import qqlive, iqiyiplayer
from time import localtime
from pfunc.cfunc import check_url_locale
from basic.vars import qqlive, iqiyiplayer, chrome
# 放一些仅通过某个id获取另一个/多个id的方法
@@ -26,7 +27,7 @@ def get_danmu_target_id_by_vid(vid: str):
try:
r = requests.get(api_url, params=params, headers=qqlive).content.decode("utf-8")
except Exception as e:
print("error info -->", e)
print("target_id requests error info -->", e)
return None
data = json.loads(r.lstrip("QZOutputJson=").rstrip(";"))
target_id = None
@@ -53,7 +54,7 @@ def get_all_vids_by_cid(cid):
try:
nomal_ids = json.loads(data["results"][0]["fields"]["nomal_ids"])
except Exception as e:
print("error info -->", e)
print("load nomal_ids error info -->", e)
return None
# F 2是免费 7是会员 0是最新正片之前的预告 4是正片之后的预告
vids = [item["V"] for item in nomal_ids if item["F"] in [2, 7]]
@@ -63,23 +64,132 @@ def get_all_vids_by_cid(cid):
#-------------------------------------------iqiyi--------------------------------------------
def get_vinfos(aid):
def get_vinfos(aid, locale="zh_cn"):
api_url = "http://cache.video.iqiyi.com/avlist/{}/0/".format(aid)
if locale != "zh_cn":
api_url += "?locale=" + locale
try:
r = requests.get(api_url, headers=iqiyiplayer).content.decode("utf-8")
r = requests.get(api_url, headers=chrome, timeout=5).content.decode("utf-8")
except Exception as e:
print("error info -->", e)
print("get_vinfos requests error info -->", e)
return None
data = json.loads(r[len("var videoListC="):])
try:
vlist = data["data"]["vlist"]
except Exception as e:
print("error info -->", e)
print("get_vinfos load vlist error info -->", e)
return None
vinfos = [[v["shortTitle"] + "_" + str(v["timeLength"]), v["timeLength"], ["id"]] for v in vlist]
return vinfos
def matchit(patterns, text):
ret = None
for pattern in patterns:
match = re.match(pattern, text)
if match:
ret = match.group(1)
break
return ret
def duration_to_sec(duration):
return sum(x * int(t) for x, t in zip([3600, 60, 1][2 - duration.count(":"):], duration.split(":")))
def get_year_range(aid, locale="zh_cn"):
# 获取第一个和最新一个视频的年份，生成列表返回，遇到任何错误则返回当前年份
year_start = year_end = localtime().tm_year
api_url = "http://pcw-api.iqiyi.com/album/album/baseinfo/{}".format(aid)
if locale != "zh_cn":
api_url += "?locale=" + locale
try:
r = requests.get(api_url, headers=chrome, timeout=5).content.decode("utf-8")
except Exception as e:
print("error info -->", e)
return list(range(year_start, year_end + 1))
data = json.loads(r)["data"]
if data.get("firstVideo"):
year_start = int(data["firstVideo"]["period"][:4])
if data.get("latestVideo"):
year_end = int(data["latestVideo"]["period"][:4])
return list(range(year_start, year_end + 1))
def get_vinfo_by_tvid(tvid, locale="zh_cn"):
api_url = "https://pcw-api.iqiyi.com/video/video/baseinfo/{}".format(tvid)
if locale != "zh_cn":
api_url += "?locale=" + locale
try:
r = requests.get(api_url, headers=chrome, timeout=5).content.decode("utf-8")
except Exception as e:
print("error info -->", e)
return
data = json.loads(r)["data"]
if data.__class__ != dict:
return None
name = data["name"]
duration = data["durationSec"]
return [name + "_" + str(duration), duration, tvid]
def get_vinfos_by_year(aid, years: list, cid=6, locale="zh_cn"):
api_url = "https://pcw-api.iqiyi.com/album/source/svlistinfo?cid={}&sourceid={}&timelist={}".format(cid, aid, ",".join([str(_) for _ in years.copy()]))
if locale != "zh_cn":
api_url += "&locale=" + locale
try:
r = requests.get(api_url, headers=chrome, timeout=5).content.decode("utf-8")
except Exception as e:
print("get_vinfos_by_year error info -->", e)
return None
data = json.loads(r)["data"]
vinfos = []
for year in years:
if year.__class__ != str:
year = str(year)
if data.get(year) is None:
continue
for ep in data[year]:
sec = duration_to_sec(ep["duration"])
vinfos.append([ep["shortTitle"] + "_" + str(sec), sec, ep["tvId"]])
return vinfos
def get_vinfos_by_url(url):
pass
locale = check_url_locale(url)
patterns = [".+?/w_(\w+?).html", ".+?/v_(\w+?).html", ".+?/a_(\w+?).html", ".+?/lib/m_(\w+?).html"]
isw, isep, isas, isms = [re.match(pattern, url) for pattern in patterns]
if isw is None and isep is None and isas is None and isms is None:
return None
try:
r = requests.get(url, headers=chrome, timeout=5).content.decode("utf-8")
except Exception as e:
print("get_vinfos_by_url error info -->", e)
return None
cid_patterns = ["[\s\S]+?\.cid.+?(\d+)", "[\s\S]+?cid: \"(\d+)\"", "[\s\S]+?channelID.+?\"(\d+)\""]
cid = matchit(cid_patterns, r)
aid_patterns = ["[\s\S]+?aid:'(\d+)'", "[\s\S]+?albumid=\"(\d+)\"", "[\s\S]+?movlibalbumaid=\"(\d+)\"", "[\s\S]+?data-score-tvid=\"(\d+)\""]
aid = matchit(aid_patterns, r)
tvid_patterns = ["[\s\S]+?\"tvid\":\"(\d+)\"", "[\s\S]+?\['tvid'\].+?\"(\d+)\""]
tvid = matchit(tvid_patterns, r)
if cid is None:
cid = ""
elif cid == "6" and isas or isms:#对于综艺合集需要获取年份
# year_patterns = ["[\s\S]+?datePublished.+?(\d\d\d\d)-\d\d-\d\d", "[\s\S]+?data-year=\"(\d+)\""]
# year = matchit(year_patterns, r)
# if year is None:
# years = [localtime().tm_year]
# else:
# years = [year]
years = get_year_range(aid, locale=locale)
else:
pass#暂时没有其他的情况计划特别处理
if isep or isw:
if tvid is None:
return
return get_vinfo_by_tvid(tvid, locale=locale)
if isas or isms:
if aid is None:
return
if cid == "6":
return get_vinfos_by_year(aid, years, locale=locale)
else:
return get_vinfos(aid, locale=locale)
#-------------------------------------------iqiyi--------------------------------------------

43
sites/iqiyi.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2019-12-18 09:48:36
# 上次编辑时间 : 2020-01-04 17:54:46
# 创建日期: 2020-01-04 19:14:41
# 上次编辑时间: 2020-01-05 14:45:17
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
@@ -15,24 +15,10 @@ from xmltodict import parse
from basic.vars import iqiyiplayer
from basic.ass import check_content
from pfunc.dump_to_ass import check_file
from pfunc.request_info import get_vinfos
from pfunc.dump_to_ass import check_file, write_one_video_subtitles
from pfunc.request_info import get_vinfos, get_vinfos_by_url, get_vinfo_by_tvid
def get_vinfo_by_tvid(tvid):
api_url = "https://pcw-api.iqiyi.com/video/video/baseinfo/{}".format(tvid)
try:
r = requests.get(api_url, headers=iqiyiplayer).content.decode("utf-8")
except Exception as e:
print("error info -->", e)
return
data = json.loads(r)["data"]
if data.__class__ != dict:
return None
name = data["name"]
duration = data["durationSec"]
return [name + "_" + str(duration), duration, tvid]
def get_danmu_by_tvid(name, duration, tvid):
# http://cmts.iqiyi.com/bullet/41/00/10793494100_300_3.z
if tvid.__class__ == int:
@@ -50,7 +36,11 @@ def get_danmu_by_tvid(name, duration, tvid):
except Exception as e:
print("error info -->", e)
continue
raw_xml = decompress(bytearray(r), 15+32).decode('utf-8')
try:
raw_xml = decompress(bytearray(r), 15+32).decode('utf-8')
except Exception as e:
index += 1
continue
try:
entry = parse(raw_xml)["danmu"]["data"]["entry"]
except Exception as e:
@@ -62,6 +52,8 @@ def get_danmu_by_tvid(name, duration, tvid):
if entry.__class__ != list:
entry = [entry]
for comment in entry:
if comment.get("list") is None:
continue
bulletInfo = comment["list"]["bulletInfo"]
if bulletInfo.__class__ != list:
bulletInfo = [bulletInfo]
@@ -89,10 +81,12 @@ def main(args):
vi = get_vinfos(args.aid)
if vi:
vinfos += vi
# if args.url:
# vi = get_vinfos_by_url(args.url)
# if vi:
# vinfos += vi
if args.tvid == "" and args.aid == "" and args.url == "":
args.url = input("请输入iqiyi链接：\n")
if args.url:
vi = get_vinfos_by_url(args.url)
if vi:
vinfos += vi
subtitles = {}
for name, duration, tvid in vinfos:
print(name, "开始下载...")
@@ -101,5 +95,6 @@ def main(args):
print("跳过{}".format(name))
return
comments = get_danmu_by_tvid(name, duration, tvid)
write_one_video_subtitles(file_path, comments, args)
subtitles.update({file_path:comments})
return subtitles

9
sites/qq.py
View File

@@ -1,9 +1,9 @@
#!/usr/bin/env python
#!/usr/bin/env python3.7
# coding=utf-8
'''
# 作者: weimo
# 创建日期: 2019-12-18 09:37:15
# 上次编辑时间 : 2020-01-04 17:53:28
# 创建日期: 2020-01-04 19:14:37
# 上次编辑时间: 2020-01-05 14:47:36
# 一个人的命运啊,当然要靠自我奋斗,但是...
'''
@@ -14,7 +14,7 @@ import requests
from basic.vars import qqlive
from basic.ass import check_content
from pfunc.dump_to_ass import check_file
from pfunc.dump_to_ass import check_file, write_one_video_subtitles
from pfunc.request_info import get_all_vids_by_cid as get_vids
from pfunc.request_info import get_danmu_target_id_by_vid as get_target_id
@@ -158,6 +158,7 @@ def main(args):
subtitles = {}
for vinfo in vinfos:
comments, file_path = get_one_subtitle_by_vinfo(vinfo, args.font, args.font_size, args.y)
write_one_video_subtitles(file_path, comments, args)
subtitles.update({file_path:comments})
return subtitles