diff options
Diffstat (limited to 'searx/engines')
| -rw-r--r-- | searx/engines/__init__.py | 16 | ||||
| -rw-r--r-- | searx/engines/microsoft_academic.py | 77 |
2 files changed, 8 insertions, 85 deletions
diff --git a/searx/engines/__init__.py b/searx/engines/__init__.py index 7e8336e01..ca3b5d4a8 100644 --- a/searx/engines/__init__.py +++ b/searx/engines/__init__.py @@ -13,7 +13,7 @@ usage:: import sys import copy -from typing import List +from typing import Dict, List, Optional from os.path import realpath, dirname from babel.localedata import locale_identifiers @@ -67,10 +67,10 @@ class Engine: # pylint: disable=too-few-public-methods timeout: float -# Defaults for the namespace of an engine module, see :py:func:`load_engine`` +# Defaults for the namespace of an engine module, see :py:func:`load_engine` categories = {'general': []} -engines = {} +engines: Dict[str, Engine] = {} engine_shortcuts = {} """Simple map of registered *shortcuts* to name of the engine (or ``None``). @@ -81,7 +81,7 @@ engine_shortcuts = {} """ -def load_engine(engine_data): +def load_engine(engine_data: dict) -> Optional[Engine]: """Load engine from ``engine_data``. :param dict engine_data: Attributes from YAML ``settings:engines/<engine>`` @@ -157,7 +157,7 @@ def set_loggers(engine, engine_name): module.logger = logger.getChild(module_engine_name) -def update_engine_attributes(engine, engine_data): +def update_engine_attributes(engine: Engine, engine_data): # set engine attributes from engine_data for param_name, param_value in engine_data.items(): if param_name == 'categories': @@ -175,7 +175,7 @@ def update_engine_attributes(engine, engine_data): setattr(engine, arg_name, copy.deepcopy(arg_value)) -def set_language_attributes(engine): +def set_language_attributes(engine: Engine): # assign supported languages from json file if engine.name in ENGINES_LANGUAGES: engine.supported_languages = ENGINES_LANGUAGES[engine.name] @@ -248,7 +248,7 @@ def is_missing_required_attributes(engine): return missing -def is_engine_active(engine): +def is_engine_active(engine: Engine): # check if engine is inactive if engine.inactive is True: return False @@ -260,7 +260,7 @@ def is_engine_active(engine): return True -def register_engine(engine): +def register_engine(engine: Engine): if engine.name in engines: logger.error('Engine config error: ambigious name: {0}'.format(engine.name)) sys.exit(1) diff --git a/searx/engines/microsoft_academic.py b/searx/engines/microsoft_academic.py deleted file mode 100644 index a869daf2f..000000000 --- a/searx/engines/microsoft_academic.py +++ /dev/null @@ -1,77 +0,0 @@ -# SPDX-License-Identifier: AGPL-3.0-or-later -""" - Microsoft Academic (Science) -""" - -from json import dumps, loads -from searx.utils import html_to_text - -# about -about = { - "website": 'https://academic.microsoft.com', - "wikidata_id": 'Q28136779', - "official_api_documentation": 'http://ma-graph.org/', - "use_official_api": False, - "require_api_key": False, - "results": 'JSON', -} - -categories = ['images'] -paging = True -search_url = 'https://academic.microsoft.com/api/search' -_paper_url = 'https://academic.microsoft.com/paper/{id}/reference' - - -def request(query, params): - params['url'] = search_url - params['method'] = 'POST' - params['headers']['content-type'] = 'application/json; charset=utf-8' - params['data'] = dumps( - { - 'query': query, - 'queryExpression': '', - 'filters': [], - 'orderBy': 0, - 'skip': (params['pageno'] - 1) * 10, - 'sortAscending': True, - 'take': 10, - 'includeCitationContexts': False, - 'profileId': '', - } - ) - - return params - - -def response(resp): - results = [] - response_data = loads(resp.text) - if not response_data: - return results - - for result in response_data.get('pr', {}): - if 'dn' not in result['paper']: - continue - - title = result['paper']['dn'] - content = _get_content(result['paper']) - url = _paper_url.format(id=result['paper']['id']) - results.append( - { - 'url': url, - 'title': html_to_text(title), - 'content': html_to_text(content), - } - ) - - return results - - -def _get_content(result): - if 'd' in result: - content = result['d'] - if len(content) > 300: - return content[:300] + '...' - return content - - return '' |