searxng/searx/engines/swisscows.py

117 lines
3.9 KiB
Python
Raw Normal View History

"""
Swisscows (Web, Images)
@website https://swisscows.ch
@provide-api no
@using-api no
@results HTML (using search portal)
@stable no (HTML can change)
@parse url, title, content
"""
from json import loads
from urllib import urlencode, unquote
import re
# engine dependent config
categories = ['general', 'images']
paging = True
language_support = True
2016-10-30 02:04:01 +00:00
supported_languages = ["ar-SA", "es-AR", "en-AU", "de-AT", "fr-BE", "nl-BE", "pt-BR", "bg-BG", "en-CA", "fr-CA",
"es-CL", "zh-CN", "hr-HR", "cs-CZ", "da-DK", "et-EE", "fi-FI", "fr-FR", "de-DE", "el-GR",
"zh-HK", "hu-HU", "en-IN", "en-IE", "he-IL", "it-IT", "ja-JP", "ko-KR", "lv-LV", "lt-LT",
"en-MY", "es-MX", "nl-NL", "en-NZ", "nb-NO", "en-PH", "pl-PL", "pt-PT", "ro-RO", "ru-RU",
"en-SG", "sk-SK", "sl-SI", "en-ZA", "es-ES", "sv-SE", "de-CH", "fr-CH", "zh-TW", "th-TH",
"tr-TR", "uk-UA", "en-GB", "en-US", "es-US"]
# search-url
base_url = 'https://swisscows.ch/'
search_string = '?{query}&page={page}'
# regex
2016-07-11 13:29:47 +00:00
regex_json = re.compile(r'initialData: {"Request":(.|\n)*},\s*environment')
regex_json_remove_start = re.compile(r'^initialData:\s*')
regex_json_remove_end = re.compile(r',\s*environment$')
regex_img_url_remove_start = re.compile(r'^https?://i\.swisscows\.ch/\?link=')
# do search-request
def request(query, params):
if params['language'] == 'all':
ui_language = 'browser'
region = 'browser'
2016-10-30 02:04:01 +00:00
elif params['language'].split('-')[0] == 'no':
region = 'nb-NO'
else:
region = params['language']
ui_language = params['language'].split('-')[0]
search_path = search_string.format(
query=urlencode({'query': query,
'uiLanguage': ui_language,
'region': region}),
page=params['pageno'])
# image search query is something like 'image?{query}&page={page}'
if params['category'] == 'images':
search_path = 'image' + search_path
params['url'] = base_url + search_path
return params
# get response from search-request
def response(resp):
results = []
json_regex = regex_json.search(resp.content)
# check if results are returned
if not json_regex:
return []
json_raw = regex_json_remove_end.sub('', regex_json_remove_start.sub('', json_regex.group()))
json = loads(json_raw)
# parse results
for result in json['Results'].get('items', []):
result_title = result['Title'].replace(u'\uE000', '').replace(u'\uE001', '')
# parse image results
if result.get('ContentType', '').startswith('image'):
img_url = unquote(regex_img_url_remove_start.sub('', result['Url']))
# append result
results.append({'url': result['SourceUrl'],
'title': result['Title'],
'content': '',
'img_src': img_url,
'template': 'images.html'})
# parse general results
else:
result_url = result['Url'].replace(u'\uE000', '').replace(u'\uE001', '')
result_content = result['Description'].replace(u'\uE000', '').replace(u'\uE001', '')
# append result
results.append({'url': result_url,
'title': result_title,
'content': result_content})
# parse images
for result in json.get('Images', []):
# decode image url
img_url = unquote(regex_img_url_remove_start.sub('', result['Url']))
# append result
results.append({'url': result['SourceUrl'],
'title': result['Title'],
'content': '',
'img_src': img_url,
'template': 'images.html'})
# return results
return results