nas/Auto_Bangumi
1
0
Fork 0
mirror of https://github.com/EstrellaXD/Auto_Bangumi.git synced 2026-02-03 02:04:06 +08:00
Code Issues Packages Projects Releases Wiki Activity

修正目录

Browse Source
This commit is contained in:
EstrellaXD
2022-06-06 00:20:42 +08:00
parent ccfc23c74f
commit 291213d120
21 changed files with 7 additions and 40870 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.dockerignore
View File

@@ -20,4 +20,4 @@ coverage.xml
.hypothesis
auto_bangumi/const_dev.py
config/bangumi.json
config/bangumi.json/config/bangumi.json

1
.gitignore vendored
View File

@@ -163,3 +163,4 @@ cython_debug/
/auto_bangumi/const_dev.py
/config/bangumi.json
/auto_bangumi/tester.py
!/source/names.txt

4
.idea/Bangumi_Auto_Rename.iml generated
View File

@@ -1,7 +1,9 @@
<?xml version="1.0" encoding="UTF-8"?>
<module type="PYTHON_MODULE" version="4">
<component name="NewModuleRootManager">
<content url="file://$MODULE_DIR$" />
<content url="file://$MODULE_DIR$">
<sourceFolder url="file://$MODULE_DIR$/auto_bangumi" isTestSource="false" />
</content>
<orderEntry type="jdk" jdkName="Python 3.10" jdkType="Python SDK" />
<orderEntry type="sourceFolder" forTests="false" />
</component>

889
auto_bangumi/RssFilter/RSSFilter.py
View File

@@ -1,889 +0,0 @@
import re
import json
import zhconv
import logging
from fliter_base import *
logger = logging.getLogger(__name__)
handler = logging.FileHandler(
filename="RssFilter/rename_log.txt", mode="w", encoding="utf-8"
)
handler.setFormatter(
logging.Formatter(
"%(asctime)s - %(filename)s[line:%(lineno)d] - %(levelname)s: %(message)s"
)
)
logger.level = logging.WARNING
logger.addHandler(handler)
const = {
"group_character":[
"字幕社",
"字幕组",
"字幕屋",
"发布组",
"连载组",
"动漫",
"国漫",
"汉化",
"raw",
"works",
"工作室",
"压制",
"合成",
"制作",
"搬运",
"委员会",
"家族",
"译制",
"动画",
"研究所",
"sub",
"翻译",
"联盟",
"dream",
"-rip",
"neo",
"team",
"百合组",
"慕留人",
"行动组",
],
"group_char" : [
"dmhy",
"澄空学园",
"c.c动漫",
"vcb",
"amor",
"moozzi2",
"skytree",
"sweetsub",
"pcsub",
"ahu-sub",
"f宅",
"captions",
"dragsterps",
"onestar",
"lolihouse",
"天空树",
"妇联奶子",
"不够热",
"烤肉同好",
"卡通",
"时雨初空",
"nyaa",
"ddd",
"koten",
"reinforce",
"届恋对邦小队",
"cxraw",
"witex.io",
]
}
const["all_charactor"] = const["group_character"] + const["group_char"]
class RSSInfoCleaner:
class Name:
def __init__(self) -> None:
self.raw = None
self.conv = None
self.zh = None
self.en = None
self.jp = None
self.clean = None
class Info:
def __init__(self) -> None:
self.group = None
self.season = None
self.episode = None
self.vision = None
class Tag:
def __init__(self) -> None:
self.dpi = None
self.ass = None
self.lang = None
self.type = None
self.code = None
self.source = None
def __init__(self, file_name):
self.name = RSSInfoCleaner.Name()
self.info = RSSInfoCleaner.Info()
self.tag = RSSInfoCleaner.Tag()
self.file_name = file_name
self.name.raw = file_name # 接收文件名参数
self.clean() # 清理广告等杂质
# 匹配特征等
with open("RssFilter/clean_rule.json", encoding="utf-8") as file_obj:
rule_json = json.load(file_obj)[0]["group_name"]
const["group_rule"] = [zhconv.convert(x, "zh-cn") for x in rule_json]
self.file_info = {}
self.pre_analyse = None
# 匹配字幕组特征
self.recognize_group()
self.info.group = self.get_group()
self.tag.dpi = self.get_dpi()
self.info.season = self.get_season()
self.info.episode = self.get_episode()
self.info.vision = self.get_vision()
self.tag.lang = self.get_language()
self.tag.ass = self.get_ass()
self.tag.type = self.get_type()
self.tag.code = self.get_code()
self.tag.source = self.get_source()
self.name.clean = self.get_clean_name()
self.zh_list = []
self.jp_list = []
self.en_list = []
self.get_title()
# 清理原链接（中文字符替换为英文）
def clean(self):
file_name = zhconv.convert(self.name.raw, "zh-cn")
# 去广告
file_name = re.sub(
"[（(\[【]?(字幕)?[\u4e00-\u9fa5、]{0,3}(新人|招募?新?)[\u4e00-\u9fa5、]{0,8}[）)\]】]?",
"",
file_name,
)
# 除杂
file_name = re.sub(
"[（(\[【]?★?((网飞)?\d{4}年[春夏秋冬]?)?[\d一二三四五六七八九十]{1,2}月新?番?★?[）)\]】]?",
"",
file_name,
)
# 除杂x2
file_name = re.sub("[（(\[【 ](2\d{3})[）)\]】 ]", " ", file_name)
# 除杂x3
file_name = re.sub(
"[（(\[【]?((网飞)?2(\d{3}[年.][春夏秋冬]?)\d{1,2}\.?\d{1,2})[）)\]】]?", "", file_name
)
# 除杂x4
file_name = re.sub("[（(\[【]检索.*[）)\]】]?", "", file_name)
strip = [
"特效歌词",
"复制磁连",
"兼容",
"配音",
"网盘",
"\u200b",
"[PSV&PC]",
"Rv40",
"R10",
"Fin]",
"Fin ",
"[mkv]",
"[]",
"★",
"☆",
]
file_name = del_rules(file_name, strip)
# xx_xx_xx
f_res = re.search("]?(([a-zA-Z:.。,，!！]{1,10})[_\[ ]){2,}", file_name)
if f_res is not None:
file_name = file_name.replace(
f_res.group(), "%s/" % f_res.group().replace("_", " ")
)
# 中文_英文名_
f_res = re.search("_[a-zA-Z_ \-·、.。，!！]*[_）)\]】]", file_name)
# !!!重要
if f_res is not None:
file_name = file_name.replace(
f_res.group(), "/%s/" % f_res.group().strip("_")
)
# 日文.英文名
f_res = re.search(
"([\u4e00-\u9fa5\u3040-\u31ff\d:\-·、.。，!！]{1,20}\.)([a-zA-Z\d:\-.。,，!！]{1,20} ?){2,}",
file_name,
)
if f_res is not None:
file_name = file_name.replace(
f_res.group(1), "%s/" % f_res.group(1).strip(".")
)
self.name.raw = (
str(file_name)
.replace("：", ":")
.replace("【", "[")
.replace("】", "]")
.replace("-", "-")
.replace("（", "(")
.replace("）", ")")
.replace("＆", "&")
.replace("X", "x")
.replace("×", "x")
.replace("Ⅹ", "x")
.replace("__", "/")
)
# 检索字幕组特征
def recognize_group(self):
rule = const["group_rule"]
# 字幕组（特例）特征优先级大于通用特征
character = const["all_charactor"]
# !强规则，人工录入标准名，区分大小写，优先匹配
for char in rule:
if ("&" + char) in self.file_name or (char + "&") in self.file_name:
self.pre_analyse = (
re.search(
"[（(\[【]?(.*?(&%s|%s&).*?)[）)\]】]?" % (char, char),
self.file_name,
)
.group(1)
.lower()
)
return "enforce"
else:
if char in self.file_name:
self.pre_analyse = char.lower()
return "enforce"
# 如果文件名以 [字幕组名] 开头
if self.name.raw[0] == "[":
str_split = self.name.raw.lower().split("]")
# 检索特征值是否位于文件名第1、2、最后一段
for char in character:
if (
char in str_split[0]
or char in str_split[1]
or char in str_split[-1]
):
self.pre_analyse = char
return "success"
# 文件名是否为 [字幕组名&字幕组名&字幕组名] ，求求了，一集的工作量真的需要三个组一起做吗
if "&" in str_split[0]:
# 限制匹配长度，防止出bug
self.pre_analyse = (
str_split[0][1:] if len(str_split[0][1:]) < 15 else None
)
return "special"
# 再匹配不上我就麻了
self.pre_analyse = None
return False
# 文件名以 -字幕组名 结尾
elif "-" in self.name.raw:
for char in character:
if char in self.name.raw.lower().split("-")[-1]:
self.pre_analyse = self.name.raw.lower().split("-")[-1]
return "reserve"
self.pre_analyse = None
return False
# 文件名以空格分隔 字幕组名为第一段
else:
first_str = self.name.raw.lower().split(" ")[0]
for char in character:
if char in first_str:
self.pre_analyse = first_str
return "blank"
self.pre_analyse = None
return False
# 获取字幕组名
def get_group(self):
# 是否匹配成功（哪种方式匹配成功）
status = self.recognize_group()
# 检索到的特征值
res_char = self.pre_analyse
# 分别对应 1、强制匹配 2、文件名为 [字幕组名&字幕组名&字幕组名]
# 3、字幕组在结尾，这种情况已经识别出关键词 4、文件名以空格分隔 字幕组名为第一段
if status in ["enforce", "special", "reserve", "blank"]:
return res_char
# 大部分情况
elif status == "success":
# 如果是 [字幕组名] ，这么标准的格式直接else送走吧，剩下的匹配一下
if "[%s]" % res_char not in self.name.raw.lower():
if self.name.raw[0] == "[":
try:
# 以特征值为中心，匹配最近的中括号，八成就这个了
gp = get_gp(res_char, self.name.raw.lower())
return gp
except Exception as e:
logger.warning(
"bug -- res_char:%s,%s,%s"
% (res_char, self.name.raw.lower(), e)
)
else:
return res_char
# 再见
return None
# 扒了6W数据，硬找的参数，没啥说的
def get_dpi(self):
file_name = self.name.raw
dpi_list = [
"4k",
"2160p",
"1440p",
"1080p",
"1036p",
"816p",
"810p",
"720p",
"576p",
"544P",
"540p",
"480p",
"1080i",
"1080+",
"360p",
"3840x2160",
"1920x1080",
"1920x1036",
"1920x804",
"1920x800",
"1536x864",
"1452x1080",
"1440x1080",
"1280x720",
"1272x720",
"1255x940",
"1024x768",
"1024X576",
"960x720",
"948x720",
"896x672",
"872x480",
"848X480",
"832x624",
"704x528",
"640x480",
"mp4_1080",
"mp4_720",
]
for i in dpi_list:
dpi = str(file_name).lower().find(i)
if dpi > 0:
return [str(i)]
return None
# 获取语种
def get_language(self):
file_name = self.name.raw
lang = []
# 中文标示
try:
lang.append(
re.search(
"[（(\[【 ]((tvb)?([粤简繁英俄][日中文体&/]?[_&]?){1,5})[）)\]】]?",
str(file_name),
)
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
# 中文标示
try:
lang.append(
re.search("[（(\[【]?[粤中简繁英俄日文体](双?(语|字幕))[）)\]】]?", str(file_name))
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
# 英文标示
try:
lang = lang + re.search(
"[（(\[【]?(((G?BIG5|CHT|CHS|GB|JPN?|CN)[/ _]?){1,3})[）)\]】]?",
str(file_name),
).group(1).lower().strip(" ").split(" ")
except Exception as e:
logger.info(e)
if lang:
return lang
else:
return None
# 文件种类
def get_type(self):
file_name = self.name.raw
type_list = []
# 英文标示
try:
type_list.append(
re.search(
"[（(\[【]?(((mp4|mkv|mp3)[ -]?){1,3})[）)\]】]?",
str(file_name).lower(),
)
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
if type_list:
return type_list
else:
return None
# 编码格式
def get_code(self):
file_name = self.name.raw
code = []
# 视频编码
try:
code = code + re.search(
"[（(\[【]?([ _-]?([xh]26[45]|hevc|avc)){1,5}[ ）)\]】]?",
str(file_name).lower(),
).group(1).split(" ")
except Exception as e:
logger.info(e)
# 位深
try:
code = code + re.search(
"[（(\[【]?[ _-]?((10|8)[ -]?bit)[ ）)\]】]?", str(file_name).lower()
).group(1).split(" ")
except Exception as e:
logger.info(e)
# 音频编码
try:
code = code + re.search(
"[（(\[【]?(([ _-]?((flac(x\d)?|aac|mp3|opus)(x\d)?)){1,5})[ ）)\]】]?",
str(file_name).lower(),
).group(3).split(" ")
except Exception as e:
logger.info(e)
if code:
return code
else:
return None
# 来源
def get_source(self):
file_name = str(self.name.raw).lower()
type_list = []
# 英文标示
for _ in range(3):
try:
res = (
re.search(
"[（(\[【]?((bd|dvd|hd|remux|(viu)?tvb?|ani-one|bilibili|网飞(动漫)|b-?global|baha|web[ /-]?(dl|rip))[ -]?(b[o0]x|iso|mut|rip)?)[）)\]】]?",
file_name,
)
.group(1)
.lower()
.strip(" ")
)
if res not in type_list:
type_list.append(res)
except Exception as e:
logger.info(e)
for res in type_list:
file_name = file_name.replace(res, "")
if type_list:
return type_list
else:
return None
# 获取季度
def get_season(self):
file_name = self.name.raw.lower()
season = []
# 中文标示
try:
season.append(
re.search(" ?(第?(\d{1,2}|[一二三])(部|季|季度|丁目))", str(file_name))
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
# 英文标示
try:
season.append(
re.search(
"((final ?)?(season|[ \[]s) ?\d{1,2}|\d{1,2}-?choume)",
str(file_name),
)
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
if season:
return season
else:
return None
# 获取集数
def get_episode(self):
file_name = self.name.raw.lower()
episode = []
# _集，国漫
try:
episode.append(
re.search("(_((\d+集-)?\d+集)|[ (\[第]\d+-\d+ ?)", str(file_name)).group(1)
)
return episode
except Exception as e:
logger.info(e)
# [10 11]集点名批评这种命名方法，几个国漫的组
try:
episode.append(
re.search(
"[\[( ](\d{1,3}[- &_]\d{1,3}) ?(fin| Fin|\(全集\))?[ )\]]",
str(file_name),
).group(1)
)
return episode
except Exception as e:
logger.info(e)
# 这里匹配ova 剧场版 不带集数的合集 之类的
try:
episode.append(
re.search(
"[\[ 第](_\d{1,3}集|ova|剧场版|全|OVA ?\d{0,2}|合|[一二三四五六七八九十])[集话章 \]\[]",
str(file_name),
).group(1)
)
return episode
except Exception as e:
logger.info(e)
# 标准单集 sp单集
try:
episode.append(
re.search(
"[\[ 第e]((sp|(数码)?重映)?(1?\d{1,3}(\.\d)?|1?\d{1,3}\(1?\d{1,3}\)))(v\d)?[集话章 \]\[]",
str(file_name),
).group(1)
)
return episode
except Exception as e:
logger.info(e)
# xx-xx集
try:
episode.append(
re.search(
"[\[ 第(]((合集)?\\\)?(\d{1,3}[ &]\d{1,3})(话| |]|\(全集\)|全集|fin)",
str(file_name),
).group(1)
)
return episode
except Exception as e:
logger.info(e)
return None
# 获取版本
def get_vision(self):
file_name = self.name.raw.lower()
vision = []
# 中文
try:
vision.append(
re.search(
"[（(\[【]?(([\u4e00-\u9fa5]{0,5}|v\d)((版本?|修[复正]|WEB限定)|片源?|内详|(特别篇))(话|版|合?集?))[）)\]】]?",
str(file_name),
).group(1)
)
except Exception as e:
logger.info(e)
# 英文
try:
vision.append(
re.search(
"[（(\[【 ]\d{1,2}((v\d)((版本?|修复?正?版)|片源?|内详)?)[）)\]】]",
str(file_name),
).group(1)
)
except Exception as e:
logger.info(e)
# [v2]
try:
vision.append(re.search("[（(\[【 ](v\d)[）)\]】]", str(file_name)).group(1))
except Exception as e:
logger.info(e)
if vision:
return vision
else:
return None
# 获取字幕类型
def get_ass(self):
file_name = self.name.raw.lower()
ass = []
# 中文标示
try:
ass.append(
re.search(
"[（(\[【]?(附?([内外][挂嵌封][+&]?){1,2}(字幕|[简中日英]*音轨)?)[）)\]】]?",
str(file_name),
).group(1)
)
except Exception as e:
logger.info(e)
# 英文标示
try:
ass.append(
re.search(
"[ （(\[【+](([ +]?(ass|pgs|srt)){1,3})[）)\]】]?", str(file_name)
)
.group(1)
.strip(" ")
)
except Exception as e:
logger.info(e)
if ass:
return ass
else:
return None
# 对以/分隔的多个翻译名，进行简单提取
def easy_split(self, clean_name, zh_list, en_list, jp_list):
if "/" in clean_name:
n_list = clean_name.split("/")
for k_i in n_list:
if has_jp(k_i):
jp_list.append(k_i.strip(" "))
else:
if has_zh(k_i) is False:
en_list.append(k_i.strip(" "))
elif has_en(k_i) is False:
zh_list.append(k_i.strip(" "))
elif has_zh(k_i) and has_en(k_i):
# 如果还是同时包含中英文的情况，递龟一下
if " " not in k_i:
res = re.search(k_i, self.name.raw.lower())
if res is not None:
zh_list.append(res.group())
else:
k_i = add_separator(k_i)
self.easy_split(k_i, zh_list, en_list, jp_list)
else:
self.easy_split(k_i, zh_list, en_list, jp_list)
else:
k_list = clean_name.split(" ")
for k_i in k_list:
if has_jp(k_i):
jp_list.append(k_i.strip(" "))
else:
if has_zh(k_i) is False:
en_list.append(k_i.strip(" "))
elif has_en(k_i) is False:
zh_list.append(k_i.strip(" "))
elif has_zh(k_i) and has_en(k_i):
res = re.search(k_i, self.name.raw.lower())
if res is not None:
zh_list.append(res.group())
# 混合验证
def all_verity(self, raw_name):
self.zh_list = (
re_verity(self.zh_list, raw_name) if self.zh_list is not None else None
)
self.en_list = (
re_verity(self.en_list, raw_name) if self.en_list is not None else None
)
self.jp_list = (
re_verity(self.jp_list, raw_name) if self.jp_list is not None else None
)
# 汇总信息
def get_clean_name(self):
# 获取到的信息
info = {
"group": self.info.group,
"dpi": self.tag.dpi,
"season": self.info.season,
"episode": self.info.episode,
"vision": self.info.vision,
"lang": self.tag.lang,
"ass": self.tag.ass,
"type": self.tag.type,
"code": self.tag.code,
"source": self.tag.source,
}
# 字母全部小写
clean_name = self.name.raw.lower()
# 去除拿到的有效信息
for k, v in info.items():
if v is not None:
if type(v) is list:
for i in v:
clean_name = (
clean_name.replace(i, "") if i is not None else clean_name
)
else:
clean_name = clean_name.replace(v, "")
# 除杂
x_list = [
"pc&psp",
"pc&psv",
"movie",
"bangumi.online",
"donghua",
"[_]",
"仅限港澳台地区",
"话全",
"第话",
"第集",
"全集",
"字幕",
"话",
"集",
"粤",
"+",
"@",
]
for i in x_list:
clean_name = clean_name.replace(i, "")
# 去除多余空格
clean_name = re.sub(" +", " ", clean_name).strip(" ").strip("-").strip(" ")
# 去除空括号
# !!! 不能删
clean_name = clean_name.replace("][", "/")
xx = re.search(
"[\u4e00-\u9fa5\u3040-\u31ff ]([(\[。_])[\u4e00-\u9fa5\a-z]", clean_name
)
if xx is not None:
clean_name = clean_name.replace(xx.group(1), "/")
clean_name = re.sub("([(\[] *| *[)\]])", "", clean_name)
clean_name = re.sub("(/ */)", "/", clean_name)
clean_name = re.sub(" +- +", "/", clean_name).strip("_").strip("/").strip(" ")
return clean_name
# 提取标题
def get_title(self):
self.name.zh, self.name.en, self.name.jp = None, None, None
# 国漫筛选
if "国漫" in self.name.raw:
zh = re.search(
"-?([\u4e00-\u9fa5]{2,10})_?", self.name.raw.replace("[国漫]", "")
)
if zh is not None:
self.name.zh = clean_list([zh.group()])
return
if "/" not in self.name.clean:
if has_jp(self.name.clean) is False:
if has_zh(self.name.clean) is False:
en = re.search(self.name.clean, self.name.raw.lower())
if en is not None:
self.name.en = clean_list([en.group()])
return
elif (
re.search(
"(^[\u4e00-\u9fa5\u3040-\u31ff\d:\-·?？、.。，!]{1,20}[a-z\d]{,3} ?！?)([a-z\d:\-.。,，!！ ]* ?)",
self.name.clean,
)
is not None
):
res = re.search(
"(^[\u4e00-\u9fa5\u3040-\u31ff\d:\-·?？、.。，!]{1,20}[a-z\d]{,3} ?！?)[._&]?([a-z\d:\-.。,，!！ ]* ?)",
self.name.clean,
)
zh = res.group(1)
en = res.group(2)
zh = re.search(zh, self.name.raw.lower())
if zh is not None:
self.name.zh = clean_list([zh.group()])
en = re.search(en, self.name.raw.lower())
if en is not None:
self.name.en = clean_list([en.group()])
return
# 英中
elif (
re.search(
"(^([a-z\d:\-_.。,，!！ ]* ?) ?)[._&]?([\u4e00-\u9fa5\u3040-\u31ffa-z\d:\-_·?？、.。，!！ ]{1,20})",
self.name.clean,
)
is not None
):
res = re.search(
"(^([a-z\d:\-_.。,，!！ ]* ?) ?)[._&]?([\u4e00-\u9fa5\u3040-\u31ffa-z\d:\-_·?？、.。，!！ ]{1,20})",
self.name.clean,
)
zh = res.group(3)
en = res.group(1)
zh = re.search(zh, self.name.raw.lower())
if zh is not None:
self.name.zh = clean_list([zh.group()])
en = re.search(en, self.name.raw.lower())
if en is not None:
self.name.en = clean_list([en.group()])
return
elif len(re.findall("[a-zA-Z]", self.name.clean.lower())) < 10:
zh = re.search(self.name.clean, self.name.raw.lower())
if zh is not None:
self.name.zh = clean_list([zh.group()])
return
if debug > 0:
print("初筛:\r\n%s\r\n%s\r\n%s" % (self.zh_list, self.en_list, self.jp_list))
if (has_zh(self.name.clean) or has_jp(self.name.clean)) and has_en(
self.name.clean
):
self.name.clean = add_separator(self.name.clean)
self.easy_split(self.name.clean, self.zh_list, self.en_list, self.jp_list)
if debug > 0:
print("二筛:\r\n%s\r\n%s\r\n%s" % (self.zh_list, self.en_list, self.jp_list))
# 结果反代入原名验证
self.all_verity([self.name.raw, self.name.clean])
# 去除正确结果后，重新识别其他部分
if self.jp_list:
temp_name = del_rules(self.name.clean, self.jp_list)
self.easy_split(temp_name, self.zh_list, self.en_list, self.jp_list)
if self.zh_list and self.en_list == []:
temp_name = del_rules(self.name.clean, self.zh_list)
self.easy_split(temp_name, self.zh_list, self.en_list, self.jp_list)
elif self.zh_list == [] and self.en_list:
temp_name = del_rules(self.name.clean, self.en_list)
self.easy_split(temp_name, self.zh_list, self.en_list, self.jp_list)
while "" in self.en_list:
self.en_list.remove("")
if debug > 0:
print("三筛:\r\n%s\r\n%s\r\n%s" % (self.zh_list, self.en_list, self.jp_list))
# 一步一验
self.all_verity([self.name.raw, self.name.clean])
for _ in range(5):
# 拼合碎片
splicing(self.zh_list, self.zh_list, self.name.clean)
splicing(self.en_list, self.en_list, self.name.clean)
splicing(self.jp_list, self.jp_list, self.name.clean)
try:
# 拼合中英文碎片
for i in self.en_list:
for j in self.zh_list:
res = re.search("%s +%s" % (i, j), self.name.raw.lower())
if res is not None:
self.en_list.remove(i)
self.zh_list.append(res.group())
except Exception as e:
logger.info(e)
if debug > 0:
print("拼合:\r\n%s\r\n%s\r\n%s" % (self.zh_list, self.en_list, self.jp_list))
# 再次验证，这里只能验raw名
self.all_verity(self.name.raw)
# 灌装
self.name.zh = clean_list(self.zh_list)
bug_list = ["不白吃话山海经"]
for i in bug_list:
if i in self.name.raw.lower():
if has_zh(i):
self.name.zh = [i]
self.name.en = clean_list(self.en_list)
self.name.jp = clean_list(self.jp_list)
if __name__ == "__main__":
debug = 0
# mikan/dmhy 获取数据，dmhy 最多1w行，mikan最多3w行
# 数据序号，向下x条
num = debug if debug > 1 else 800
row = 1 if debug else 200
name_list = read_data("mikan", num, row)
for i in range(0, len(name_list)):
info = RSSInfoCleaner(name_list[i])
title = info.Name
print("%s:%s" % (num + i, name_list[i]))
print("raw_name:%s" % title.raw)
print("clean_name:%s" % title.clean)
print("zh:%s" % title.zh)
print("en:%s" % title.en)
print("jp:%s" % title.jp)
print()

0
auto_bangumi/RssFilter/__init__.py
View File

74
auto_bangumi/RssFilter/clean_rule.json
View File

@@ -1,74 +0,0 @@
[
{
"group_name": [
"Lilith-Raws x WitEx.io",
"极影字幕社+辉夜汉化组",
"Lilith-Raws",
"NC-Raws",
"Skymoon-Raws",
"天月搬运组",
"肥猫压制",
"LoliHouse",
"猎户不鸽发布组",
"NaN-Raws",
"猎户随缘发布组",
"桜都字幕组",
"澄空学园&雪飘工作室",
"千夏字幕组",
"百冬练习组",
"IET字幕组",
"离谱Sub",
"酷漫404",
"星空字幕组",
"轻之国度字幕组",
"枫叶字幕组",
"雪飘工作室",
"豌豆字幕组",
"云光字幕组",
"悠哈璃羽字幕社",
"桜都字幕组",
"ANi",
"❀拨雪寻春❀",
"极彩字幕组",
"悠哈璃羽字幕社",
"爱恋&漫猫字幕组",
"MingY",
"VCB-Studio",
"喵萌奶茶屋",
"爱恋字母社",
"诸神字幕组",
"驯兽师联盟",
"夏沐字幕组",
"动漫国字幕组",
"百冬练习组s",
"SweetSub&圆环记录攻略组",
"动漫萌",
"极影字幕社",
"喵萌Production",
"喵萌Production&LoliHouse",
"60yrs ago",
"50yrs ago",
"41yrs ago",
"40yrs ago",
"30yrs ago",
"20yrs ago",
"s5291s",
"xyx98",
"nvacg",
"RHxDymy",
"PoInSu",
"EMe",
"DHR百合組",
"雪飘工作室",
"喵萌美食殿",
"GalaxyRailroad-888",
"APTX4869",
"SummerHuo&ZJCONAN",
"nyaa",
"AngelEcho",
"逆时针环游",
"MCE汉化组"
],
"name_position": 1
}
]

10001
auto_bangumi/RssFilter/dmhy.csv
View File

File diff suppressed because one or more lines are too long

225
auto_bangumi/RssFilter/fliter_base.py
View File

@@ -1,225 +0,0 @@
import re
import logging
import csv
logger = logging.getLogger(__name__)
def read_data(file_name, start, rows):
if file_name == "mikan":
with open('RssFilter/mikan.csv', 'r', encoding='utf-8') as csv_file:
reader = csv.reader(csv_file)
raw_data = [row[3] for row in reader][start:start + rows]
return raw_data
elif file_name == "dmhy":
with open('RssFilter/dmhy.csv', 'r', encoding='utf-8') as csv_file:
reader = csv.reader(csv_file)
raw_data = [row[4] for row in reader][start + 1:start + rows + 1]
return raw_data
# 以 / 代替空格分隔中英文名
def add_separator(clean_name):
try:
if '\u4e00' <= clean_name[0] <= '\u9fff':
try:
res = re.search(
"(^[\u4e00-\u9fa5\u3040-\u31ff\d: \-·、.。，!！]{1,20}[ -_]{1,5})([a-z\d:\-.。,，!！]{1,20} ?){2,}",
clean_name).group(1)
clean_name = clean_name.replace(res, res.strip(" ") + "/")
except Exception as e:
logger.exception(e)
else:
try:
res = re.search(
"^(([a-z\d:\-.。,，!！]{1,20} ?){2,}[ -_]{1,5})[\u4e00-\u9fa5\u3040-\u31ff\d: \-·、.。,，!！]{1,20}",
clean_name).group(1)
clean_name = clean_name.replace(res, res.strip(" ") + "/")
except Exception as e:
logger.exception(e)
except Exception as e:
logger.exception(e)
clean_name = re.sub("(/ */)", "/", clean_name)
return clean_name
# 拼合碎片
def splicing(frag_list, name_list, raw_name):
try:
for i in range(0, len(name_list) - 1):
if name_list[i] in name_list[i + 1] and name_list[i] != name_list[i + 1]:
name_list.remove(name_list[i])
elif name_list[i + 1] in name_list[i] and name_list[i] != name_list[i + 1]:
name_list.remove(name_list[i + 1])
except Exception as e:
logger.info(e)
min_list = sorted(name_list, key=lambda i: len(i), reverse=False)
for i in range(0, len(min_list) - 1):
# 处理中英文混合名
if frag_list is not None and len(frag_list) > 1:
fragment = min_list[i]
try:
if fragment in raw_name.lower():
for piece_name in name_list:
try:
r_name = re.search("(%s {0,3}%s|%s {0,5}%s)" % (fragment, piece_name, piece_name, fragment),
raw_name.lower())
if r_name is not None:
frag_list.remove(fragment)
name_list.remove(piece_name)
name_list.append(r_name.group())
except Exception as e:
logger.warning("bug--%s" % e)
logger.warning("piece_name:%s,fragment:%s" % (piece_name, fragment))
except Exception as e:
logger.exception(e)
# 清理列表
def clean_list(raw_list):
if raw_list is not None:
# 去除碎片和杂质
raw_list = [x.strip("-").strip(" ") for x in raw_list if len(x) > 1]
# 小碎片归并
for _ in range(len(raw_list)):
if raw_list is not None and len(raw_list) > 1:
try:
for i in range(0, len(raw_list) - 1):
if raw_list[i] in raw_list[i + 1] and raw_list[i] != raw_list[i + 1]:
raw_list.remove(raw_list[i])
elif raw_list[i + 1] in raw_list[i] and raw_list[i] != raw_list[i + 1]:
raw_list.remove(raw_list[i + 1])
except Exception as e:
logger.info(e)
if raw_list is not None and len(raw_list) > 1:
try:
for i in range(0, len(raw_list)):
up_list = sorted(raw_list, key=lambda i: len(i), reverse=False)
if up_list[i] in up_list[-1] and up_list[i] != up_list[-1]:
raw_list.remove(up_list[i])
except Exception as e:
logger.info(e)
if raw_list:
return set(raw_list)
else:
return None
# 粗略识别失败，re强制匹配
def extract_title(raw_name):
title = {
"zh": None,
"en": None,
}
clean_name = raw_name
if has_en(clean_name) and has_zh(clean_name):
# 中英
try:
res = re.search("(([\u4e00-\u9fa5]{2,12}[ /:]{0,3}){1,5}) {0,5}(( ?[a-z':]{1,15}){1,15})", clean_name)
title["zh"] = res.group(1).strip(" ")
title["en"] = res.group(3).strip(" ")
except Exception as e:
logger.info(e)
# 本程序依赖此bug运行，这行不能删
if title["zh"] is None:
# 中英
try:
res = re.search(
"(([\u4e00-\u9fa5a]{1,12}[ /:]{0,3}){1,5})[&/ (]{0,5}(( ?[a-z':]{1,15}){1,15})[ )/]{0,3}",
clean_name)
title["zh"] = res.group(1).strip(" ")
title["en"] = res.group(3).strip(" ")
except Exception as e:
logger.info(e)
# 英中
try:
res = re.search(
"(([ a-z'.:]{1,20}){1,8})[&/ (]{0,5}(([\u4e00-\u9fa5a]{2,10}[a-z]{0,3} ?){1,5})[ )/]{0,3}",
clean_name)
title["en"] = res.group(1).strip(" ")
title["zh"] = res.group(3).strip(" ")
except Exception as e:
logger.info(e)
else:
if has_zh(clean_name):
# 中文
try:
res = re.search("(([\u4e00-\u9fa5:]{2,15}[ /]?){1,5}) *", clean_name)
title["zh"] = res.group(1).strip(" ")
except Exception as e:
logger.info(e)
elif has_en(clean_name):
# 英文
try:
res = re.search("(([a-z:]{2,15}[ /]?){1,15}) *", clean_name)
title["en"] = res.group(1).strip(" ")
except Exception as e:
logger.info(e)
for k, v in title.items():
if v is not None and "/" in v:
zh_list = v.split("/")
title[k] = zh_list[0].strip(" ")
return title
def del_rules(raw_name, rule_list):
for i in rule_list:
raw_name = raw_name.replace(i, "")
return raw_name
# 获取字符串出现位置
def get_str_location(char, target):
locate = []
for index, value in enumerate(char):
if target == value:
locate.append(index)
return locate
# 匹配某字符串最近的括号
def get_gp(char, string):
begin = [x for x in get_str_location(string, "[") if int(x) < int(string.find(char))][-1] + 1
end = [x for x in get_str_location(string, "]") if int(x) > int(string.find(char))][0]
return string[begin:end]
def has_en(str):
my_re = re.compile(r'[a-z]', re.S)
res = re.findall(my_re, str)
if len(res):
return True
else:
return False
def has_zh(str):
my_re = re.compile(r'[\u4e00-\u9fa5]', re.S)
res = re.findall(my_re, str)
if len(res):
return True
else:
return False
def has_jp(str):
my_re = re.compile(r'[\u3040-\u31ff]', re.S)
res = re.findall(my_re, str)
if len(res):
return True
else:
return False
# 单list验证
def re_verity(raw_list, raw_name):
correct_list = []
for c_res in raw_list:
if type(raw_name) is list:
if c_res in raw_name[0].lower() and c_res in raw_name[1].lower():
correct_list.append(c_res)
else:
if c_res in raw_name.lower():
correct_list.append(c_res)
return correct_list

29512
auto_bangumi/RssFilter/mikan.csv
View File

File diff suppressed because it is too large Load Diff

1
auto_bangumi/app.py
View File

@@ -7,7 +7,6 @@ from conf import settings
from argument_parser import parse
from log import setup_logger
from utils import json_config
from env_bool import init_switch
from core.rss_collector import RSSCollector
from core.download_client import DownloadClient

1
auto_bangumi/const.py
View File

@@ -10,7 +10,6 @@ DEFAULT_SETTINGS = {
"method": "pn",
"enable_group_tag": True,
"info_path": "/config/bangumi.json",
"rule_path": "/config/rule_beta.json",
"not_contain": "720",
"get_rule_debug": False,
"rule_url": "https://raw.githubusercontent.com/EstrellaXD/Bangumi_Auto_Collector/main/AutoBangumi/config/rule.json",

1
auto_bangumi/core/download_client.py
View File

@@ -7,7 +7,6 @@ from downloader import getClient
from downloader.exceptions import ConflictError
from conf import settings
from utils import json_config
from core.eps_complete import FullSeasonGet

6
auto_bangumi/core/eps_complete.py
View File

@@ -2,7 +2,6 @@ import os.path
import re
import requests
from qbittorrentapi import Client
from bs4 import BeautifulSoup
import logging
@@ -58,8 +57,3 @@ class FullSeasonGet:
return downloads
if __name__ == "__main__":
a = FullSeasonGet("Lilith-Raws", "Shijou Saikyou no Daimaou", "S01")
a.run()
for torrent in a.torrents:
logger.debug(torrent["url"])

9
auto_bangumi/core/renamer.py
View File

@@ -8,7 +8,6 @@ from core.download_client import DownloadClient
logger = logging.getLogger(__name__)
class Renamer:
def __init__(self, downloadClient: DownloadClient):
self.client = downloadClient
@@ -30,13 +29,13 @@ class Renamer:
return new_name
def rename_pn(self, name):
n = re.split(r"\[|\]", name)
n = re.split(r"[\[\]]", name)
file_name = name.replace(f"[{n[1]}]", "")
for rule in self.rules:
matchObj = re.match(rule, file_name, re.I)
if matchObj is not None:
new_name = re.sub(
r"\[|\]",
r"[\[\]]",
"",
f"{matchObj.group(1).strip()} E{matchObj.group(2)}{n[-1]}",
)
@@ -71,7 +70,3 @@ class Renamer:
self.client.delete_torrent(info.hash)
self.print_result(torrent_count, rename_count)
if __name__ == "__main__":
rename = Renamer()
rename.rename_pn("[Lilith-Raws] Shokei Shoujo no Virgin Road - 02 [Baha][WEB-DL][1080p][AVC AAC][CHT][MP4]")

1
auto_bangumi/core/rss_collector.py
View File

@@ -1,5 +1,4 @@
# -*- coding: UTF-8 -*-
import os
import logging
import requests
from bs4 import BeautifulSoup

1
auto_bangumi/downloader/qb_downloader.py
View File

@@ -3,7 +3,6 @@ import time
from qbittorrentapi import Client, LoginFailed
from qbittorrentapi.exceptions import Conflict409Error
from tomlkit import item
from conf import settings

22
auto_bangumi/env_bool.py
View File

@@ -1,22 +0,0 @@
from conf import settings
bool_group = [
settings.enable_group_tag,
settings.get_rule_debug,
settings.debug_mode,
settings.enable_eps_complete,
settings.season_one_tag
]
def init_switch():
if settings.sleep_time is str:
settings.sleep_time = float(settings.sleep_time)
for switch in bool_group:
if switch is str:
switch = switch.lower() in ("true", "t", "i")
if __name__ == "__main__":
settings.init()
print(type(settings.debug_mode))

59
config/rule.json
View File

@@ -1,59 +0,0 @@
[
{
"group_name": [
"Lilith-Raws",
"NC-Raws",
"Skymoon-Raws",
"天月搬运组",
"LoliHouse",
"猎户不鸽发布组",
"NaN-Raws",
"猎户随缘发布组",
"桜都字幕组",
"澄空学园&雪飘工作室",
"千夏字幕组",
"IET字幕组",
"离谱Sub",
"酷漫404",
"星空字幕组",
"轻之国度字幕组",
"枫叶字幕组",
"雪飘工作室",
"豌豆字幕组",
"云光字幕组",
"悠哈璃羽字幕社",
"桜都字幕组",
"ANi",
"❀拨雪寻春❀",
"极彩字幕组",
"悠哈璃羽字幕社",
"爱恋&漫猫字幕组",
"MingY",
"VCB-Studio",
"虹咲学园烤肉同好会"
],
"name_position": 1
},
{
"group_name": [
"喵萌奶茶屋",
"爱恋字母社",
"诸神字幕组",
"驯兽师联盟",
"夏沐字幕组",
"幻樱",
"动漫国字幕组",
"SweetSub&圆环记录攻略组",
"动漫萌",
"极影字幕社",
"喵萌Production"
],
"name_position": 2
},
{
"group_name": [
"爱恋字母社"
],
"name_position": 3
}
]

67
config/rule_beta.json
View File

@@ -1,67 +0,0 @@
[
{
"group_name": [
"Lilith-Raws",
"NC-Raws",
"Skymoon-Raws",
"天月搬运组",
"LoliHouse",
"猎户不鸽发布组",
"NaN-Raws",
"猎户随缘发布组",
"桜都字幕组",
"澄空学园&雪飘工作室",
"千夏字幕组",
"IET字幕组",
"离谱Sub",
"酷漫404",
"星空字幕组",
"轻之国度字幕组",
"枫叶字幕组",
"雪飘工作室",
"豌豆字幕组",
"云光字幕组",
"悠哈璃羽字幕社",
"桜都字幕组",
"ANi",
"❀拨雪寻春❀",
"极彩字幕组",
"悠哈璃羽字幕社",
"爱恋&漫猫字幕组",
"MingY",
"VCB-Studio",
"虹咲学园烤肉同好会",
"喵萌奶茶屋",
"爱恋字母社",
"诸神字幕组",
"驯兽师联盟",
"夏沐字幕组",
"幻樱",
"动漫国字幕组",
"SweetSub&圆环记录攻略组",
"动漫萌",
"极影字幕社",
"喵萌Production",
"Lolihouse",
"PCSUB",
"桜都字幕組",
"LowPower-Raws",
"雪飄工作室"
],
"type": 1
},
{
"group_name": [
"爱恋字母社",
"風車字幕組",
"幻櫻字幕組",
"幻櫻字幕組",
"Dymy字幕組",
"動漫國字幕組",
"c.c動漫",
"动漫国字幕组",
"c.c动漫"
],
"type": 2
}
]

1
requirements.txt
View File

@@ -3,5 +3,4 @@ bs4
requests
lxml
zhconv
tomlkit

0
auto_bangumi/bangumi_parser/names.txt → source/names.txt
View File