|
1
|
|
|
#!/usr/local/bin/python |
|
|
|
|
|
|
2
|
|
|
# coding: utf-8 |
|
3
|
|
|
|
|
4
|
|
|
import re |
|
5
|
|
|
|
|
6
|
|
|
import jellyfish |
|
7
|
|
|
import requests |
|
8
|
|
|
from bs4 import BeautifulSoup as Soup |
|
9
|
|
|
|
|
10
|
|
|
from titlesearch.language.detection import matches_language |
|
11
|
|
|
from titlesearch.language.language_settings import * |
|
|
|
|
|
|
12
|
|
|
|
|
13
|
|
|
|
|
14
|
|
|
class VisualNovelDatabase(object): |
|
|
|
|
|
|
15
|
|
|
"""Module for extracting alternative language titles for titles from https://vndb.org""" |
|
16
|
|
|
|
|
17
|
|
|
ROOT_URL = 'https://vndb.org/' |
|
18
|
|
|
SEARCH_URL = 'https://vndb.org/v/all' |
|
19
|
|
|
KNOWN_LANGUAGES = [English, Japanese, Korean] |
|
|
|
|
|
|
20
|
|
|
|
|
21
|
|
|
@staticmethod |
|
22
|
|
|
def get_similar_titles(title: str) -> list: |
|
23
|
|
|
"""Main function for extracting alternate titles |
|
24
|
|
|
|
|
25
|
|
|
:type title: str |
|
26
|
|
|
:return: |
|
27
|
|
|
""" |
|
28
|
|
|
payload = { |
|
29
|
|
|
'q': title |
|
30
|
|
|
} |
|
31
|
|
|
|
|
32
|
|
|
results = [] |
|
33
|
|
|
|
|
34
|
|
|
link = requests.get(url=VisualNovelDatabase.SEARCH_URL, params=payload) |
|
35
|
|
|
# use html5lib here to generate a tbody tag from the table(not generated with html.parser) |
|
36
|
|
|
soup = Soup(link.text, 'html5lib') |
|
37
|
|
|
|
|
38
|
|
|
# if the match is above a certain percentage we won't get to the search result page but to the |
|
|
|
|
|
|
39
|
|
|
# detail page of the search result so we won't get more than 1 result |
|
40
|
|
|
current_url = soup.select_one('meta[property="og:url"]') |
|
41
|
|
|
# not set apparently in the search result page |
|
42
|
|
|
if current_url and re.match('{0:s}v\d+'.format(VisualNovelDatabase.ROOT_URL), current_url['content']): |
|
|
|
|
|
|
43
|
|
|
return [{ |
|
44
|
|
|
'title': title, |
|
45
|
|
|
'link': current_url['content'], |
|
46
|
|
|
'similarity': 1.00 |
|
47
|
|
|
}] |
|
48
|
|
|
|
|
49
|
|
|
title_links = soup.select('tbody td a') |
|
50
|
|
|
for search_result in title_links: |
|
51
|
|
|
results.append({ |
|
52
|
|
|
'title': search_result['title'], |
|
53
|
|
|
'link': VisualNovelDatabase.ROOT_URL + search_result['href'], |
|
54
|
|
|
'similarity': jellyfish.jaro_distance(search_result['title'].lower(), title.lower()) |
|
55
|
|
|
}) |
|
56
|
|
|
|
|
57
|
|
|
results.sort(key=lambda item: item['similarity'], reverse=True) |
|
58
|
|
|
return results |
|
59
|
|
|
|
|
60
|
|
|
@staticmethod |
|
61
|
|
|
def get_alternative_titles(title: str = '', link: str = '') -> dict: |
|
62
|
|
|
"""Get alternative titles for the given title. Preferring link over title argument |
|
63
|
|
|
|
|
64
|
|
|
:type title: str |
|
65
|
|
|
:type link: str |
|
66
|
|
|
:return: |
|
67
|
|
|
""" |
|
68
|
|
|
if title and not link: |
|
69
|
|
|
link = VisualNovelDatabase.get_similar_titles(title) |
|
70
|
|
|
if link: |
|
71
|
|
|
link = link[0]['link'] |
|
72
|
|
|
else: |
|
73
|
|
|
return VisualNovelDatabase.group_titles(title, []) |
|
74
|
|
|
|
|
75
|
|
|
link = requests.get(url=link) |
|
76
|
|
|
|
|
77
|
|
|
result_data = VisualNovelDatabase.parse_results(link.text) |
|
78
|
|
|
|
|
79
|
|
|
alternative_titles = [] |
|
80
|
|
|
if 'Aliases' in result_data: |
|
81
|
|
|
for alternative_title in result_data['Aliases'].split(', '): |
|
82
|
|
|
alternative_titles.append(alternative_title) |
|
83
|
|
|
if 'Original title' in result_data: |
|
84
|
|
|
alternative_titles.append(result_data['Original title']) |
|
85
|
|
|
|
|
86
|
|
|
return VisualNovelDatabase.group_titles(release_title=result_data['Title'], |
|
87
|
|
|
alternative_titles=alternative_titles) |
|
88
|
|
|
|
|
89
|
|
|
@staticmethod |
|
90
|
|
|
def parse_results(html_content: str) -> dict: |
|
91
|
|
|
"""Parse the result table of the HTML content into a dictionary |
|
92
|
|
|
|
|
93
|
|
|
:type html_content: str |
|
94
|
|
|
:return: |
|
95
|
|
|
""" |
|
96
|
|
|
soup = Soup(html_content, 'html.parser') |
|
97
|
|
|
result_data = {} |
|
98
|
|
|
|
|
99
|
|
|
# |
|
100
|
|
|
table_body = soup.select_one('div.vndetails table') |
|
101
|
|
|
rows = table_body.find_all('tr') |
|
102
|
|
|
for row in rows: |
|
103
|
|
|
cols = row.find_all('td') |
|
104
|
|
|
cols = [ele.text.strip() for ele in cols if ele.text.strip()] |
|
105
|
|
|
if cols[1:]: |
|
106
|
|
|
result_data[cols[0]] = cols[1] |
|
107
|
|
|
|
|
108
|
|
|
return result_data |
|
109
|
|
|
|
|
110
|
|
View Code Duplication |
@staticmethod |
|
|
|
|
|
|
111
|
|
|
def group_titles(release_title: str, alternative_titles: list) -> dict: |
|
112
|
|
|
"""Iterate through the supported languages and group the titles according to the detected languages |
|
|
|
|
|
|
113
|
|
|
|
|
114
|
|
|
:type release_title: str |
|
115
|
|
|
:type alternative_titles: list |
|
116
|
|
|
:return: |
|
117
|
|
|
""" |
|
118
|
|
|
grouped_titles = {} |
|
119
|
|
|
for language in VisualNovelDatabase.KNOWN_LANGUAGES: |
|
120
|
|
|
grouped_titles[language.__name__.lower()] = [] |
|
121
|
|
|
|
|
122
|
|
|
grouped_titles['english'] = [release_title] |
|
123
|
|
|
|
|
124
|
|
|
for title in alternative_titles: |
|
125
|
|
|
for language in VisualNovelDatabase.KNOWN_LANGUAGES: |
|
126
|
|
|
if matches_language(title, language) and title not in grouped_titles[language.__name__.lower()]: |
|
|
|
|
|
|
127
|
|
|
grouped_titles[language.__name__.lower()].append(title) |
|
128
|
|
|
continue |
|
129
|
|
|
|
|
130
|
|
|
return grouped_titles |
|
131
|
|
|
|
DaRealFreak /
TitleSearch
Loading history...
The coding style of this project requires that you add a docstring to this code element. Below, you find an example for methods:
If you would like to know more about docstrings, we recommend to read PEP-257: Docstring Conventions.