mirror of
				https://github.com/searxng/searxng
				synced 2024-01-01 19:24:07 +01:00 
			
		
		
		
	[mod] searx.engines.__init__: refactoring
This commit is contained in:
		
							parent
							
								
									e3f4a77311
								
							
						
					
					
						commit
						70a9208972
					
				
					 1 changed files with 108 additions and 104 deletions
				
			
		| 
						 | 
					@ -1,59 +1,46 @@
 | 
				
			||||||
 | 
					# SPDX-License-Identifier: AGPL-3.0-or-later
 | 
				
			||||||
'''
 | 
					# lint: pylint
 | 
				
			||||||
searx is free software: you can redistribute it and/or modify
 | 
					# pylint: disable=missing-function-docstring
 | 
				
			||||||
it under the terms of the GNU Affero General Public License as published by
 | 
					"""
 | 
				
			||||||
the Free Software Foundation, either version 3 of the License, or
 | 
					Engine loader:
 | 
				
			||||||
(at your option) any later version.
 | 
					call load_engines(settings['engines])
 | 
				
			||||||
 | 
					to initialiaze categories, engines, engine_shortcuts
 | 
				
			||||||
searx is distributed in the hope that it will be useful,
 | 
					"""
 | 
				
			||||||
but WITHOUT ANY WARRANTY; without even the implied warranty of
 | 
					 | 
				
			||||||
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 | 
					 | 
				
			||||||
GNU Affero General Public License for more details.
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
You should have received a copy of the GNU Affero General Public License
 | 
					 | 
				
			||||||
along with searx. If not, see < http://www.gnu.org/licenses/ >.
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
(C) 2013- by Adam Tauber, <asciimoo@gmail.com>
 | 
					 | 
				
			||||||
'''
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
import sys
 | 
					import sys
 | 
				
			||||||
import threading
 | 
					import copy
 | 
				
			||||||
 | 
					
 | 
				
			||||||
from os.path import realpath, dirname
 | 
					from os.path import realpath, dirname
 | 
				
			||||||
from babel.localedata import locale_identifiers
 | 
					from babel.localedata import locale_identifiers
 | 
				
			||||||
from urllib.parse import urlparse
 | 
					from searx import logger, settings
 | 
				
			||||||
from operator import itemgetter
 | 
					 | 
				
			||||||
from searx import settings
 | 
					 | 
				
			||||||
from searx import logger
 | 
					 | 
				
			||||||
from searx.data import ENGINES_LANGUAGES
 | 
					from searx.data import ENGINES_LANGUAGES
 | 
				
			||||||
from searx.exceptions import SearxEngineResponseException
 | 
					from searx.network import get
 | 
				
			||||||
from searx.network import get, initialize as initialize_network, set_context_network_name
 | 
					from searx.utils import load_module, match_language, gen_useragent
 | 
				
			||||||
from searx.utils import load_module, match_language, get_engine_from_settings, gen_useragent
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
logger = logger.getChild('engines')
 | 
					logger = logger.getChild('engines')
 | 
				
			||||||
 | 
					ENGINE_DIR = dirname(realpath(__file__))
 | 
				
			||||||
engine_dir = dirname(realpath(__file__))
 | 
					BABEL_LANGS = [lang_parts[0] + '-' + lang_parts[-1] if len(lang_parts) > 1 else lang_parts[0]
 | 
				
			||||||
 | 
					 | 
				
			||||||
engines = {}
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
categories = {'general': []}
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
babel_langs = [lang_parts[0] + '-' + lang_parts[-1] if len(lang_parts) > 1 else lang_parts[0]
 | 
					 | 
				
			||||||
               for lang_parts in (lang_code.split('_') for lang_code in locale_identifiers())]
 | 
					               for lang_parts in (lang_code.split('_') for lang_code in locale_identifiers())]
 | 
				
			||||||
 | 
					ENGINE_DEFAULT_ARGS = {
 | 
				
			||||||
 | 
					    "engine_type": "online",
 | 
				
			||||||
 | 
					    "inactive": False,
 | 
				
			||||||
 | 
					    "disabled": False,
 | 
				
			||||||
 | 
					    "timeout": settings["outgoing"]["request_timeout"],
 | 
				
			||||||
 | 
					    "shortcut": "-",
 | 
				
			||||||
 | 
					    "categories": ["general"],
 | 
				
			||||||
 | 
					    "supported_languages": [],
 | 
				
			||||||
 | 
					    "language_aliases": {},
 | 
				
			||||||
 | 
					    "paging": False,
 | 
				
			||||||
 | 
					    "safesearch": False,
 | 
				
			||||||
 | 
					    "time_range_support": False,
 | 
				
			||||||
 | 
					    "enable_http": False,
 | 
				
			||||||
 | 
					    "display_error_messages": True,
 | 
				
			||||||
 | 
					    "tokens": [],
 | 
				
			||||||
 | 
					}
 | 
				
			||||||
 | 
					categories = {'general': []}
 | 
				
			||||||
 | 
					engines = {}
 | 
				
			||||||
engine_shortcuts = {}
 | 
					engine_shortcuts = {}
 | 
				
			||||||
engine_default_args = {'paging': False,
 | 
					 | 
				
			||||||
                       'categories': ['general'],
 | 
					 | 
				
			||||||
                       'supported_languages': [],
 | 
					 | 
				
			||||||
                       'safesearch': False,
 | 
					 | 
				
			||||||
                       'timeout': settings['outgoing']['request_timeout'],
 | 
					 | 
				
			||||||
                       'shortcut': '-',
 | 
					 | 
				
			||||||
                       'disabled': False,
 | 
					 | 
				
			||||||
                       'enable_http': False,
 | 
					 | 
				
			||||||
                       'time_range_support': False,
 | 
					 | 
				
			||||||
                       'engine_type': 'online',
 | 
					 | 
				
			||||||
                       'display_error_messages': True,
 | 
					 | 
				
			||||||
                       'tokens': []}
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
def load_engine(engine_data):
 | 
					def load_engine(engine_data):
 | 
				
			||||||
| 
						 | 
					@ -67,64 +54,59 @@ def load_engine(engine_data):
 | 
				
			||||||
        engine_name = engine_name.lower()
 | 
					        engine_name = engine_name.lower()
 | 
				
			||||||
        engine_data['name'] = engine_name
 | 
					        engine_data['name'] = engine_name
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    # load_module
 | 
				
			||||||
    engine_module = engine_data['engine']
 | 
					    engine_module = engine_data['engine']
 | 
				
			||||||
 | 
					 | 
				
			||||||
    try:
 | 
					    try:
 | 
				
			||||||
        engine = load_module(engine_module + '.py', engine_dir)
 | 
					        engine = load_module(engine_module + '.py', ENGINE_DIR)
 | 
				
			||||||
    except (SyntaxError, KeyboardInterrupt, SystemExit, SystemError, ImportError, RuntimeError):
 | 
					    except (SyntaxError, KeyboardInterrupt, SystemExit, SystemError, ImportError, RuntimeError):
 | 
				
			||||||
        logger.exception('Fatal exception in engine "{}"'.format(engine_module))
 | 
					        logger.exception('Fatal exception in engine "{}"'.format(engine_module))
 | 
				
			||||||
        sys.exit(1)
 | 
					        sys.exit(1)
 | 
				
			||||||
    except:
 | 
					    except BaseException:
 | 
				
			||||||
        logger.exception('Cannot load engine "{}"'.format(engine_module))
 | 
					        logger.exception('Cannot load engine "{}"'.format(engine_module))
 | 
				
			||||||
        return None
 | 
					        return None
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    update_engine_attributes(engine, engine_data)
 | 
				
			||||||
 | 
					    set_language_attributes(engine)
 | 
				
			||||||
 | 
					    update_attributes_for_tor(engine)
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    if is_missing_required_attributes(engine):
 | 
				
			||||||
 | 
					        sys.exit(1)
 | 
				
			||||||
 | 
					    if not is_engine_active(engine):
 | 
				
			||||||
 | 
					        return None
 | 
				
			||||||
 | 
					    return engine
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					def update_engine_attributes(engine, engine_data):
 | 
				
			||||||
 | 
					    # set engine attributes from engine_data
 | 
				
			||||||
    for param_name, param_value in engine_data.items():
 | 
					    for param_name, param_value in engine_data.items():
 | 
				
			||||||
        if param_name == 'engine':
 | 
					        if param_name == 'categories':
 | 
				
			||||||
            pass
 | 
					            if isinstance(param_value, str):
 | 
				
			||||||
        elif param_name == 'categories':
 | 
					                param_value = list(map(str.strip, param_value.split(',')))
 | 
				
			||||||
            if param_value == 'none':
 | 
					            engine.categories = param_value
 | 
				
			||||||
                engine.categories = []
 | 
					        elif param_name != 'engine':
 | 
				
			||||||
            else:
 | 
					 | 
				
			||||||
                engine.categories = list(map(str.strip, param_value.split(',')))
 | 
					 | 
				
			||||||
        else:
 | 
					 | 
				
			||||||
            setattr(engine, param_name, param_value)
 | 
					            setattr(engine, param_name, param_value)
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    for arg_name, arg_value in engine_default_args.items():
 | 
					    # set default attributes
 | 
				
			||||||
 | 
					    for arg_name, arg_value in ENGINE_DEFAULT_ARGS.items():
 | 
				
			||||||
        if not hasattr(engine, arg_name):
 | 
					        if not hasattr(engine, arg_name):
 | 
				
			||||||
            setattr(engine, arg_name, arg_value)
 | 
					            setattr(engine, arg_name, copy.deepcopy(arg_value))
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    # checking required variables
 | 
					 | 
				
			||||||
    for engine_attr in dir(engine):
 | 
					 | 
				
			||||||
        if engine_attr.startswith('_'):
 | 
					 | 
				
			||||||
            continue
 | 
					 | 
				
			||||||
        if engine_attr == 'inactive' and getattr(engine, engine_attr) is True:
 | 
					 | 
				
			||||||
            return None
 | 
					 | 
				
			||||||
        if getattr(engine, engine_attr) is None:
 | 
					 | 
				
			||||||
            logger.error('Missing engine config attribute: "{0}.{1}"'
 | 
					 | 
				
			||||||
                         .format(engine.name, engine_attr))
 | 
					 | 
				
			||||||
            sys.exit(1)
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					def set_language_attributes(engine):
 | 
				
			||||||
 | 
					    # pylint: disable=protected-access
 | 
				
			||||||
    # assign supported languages from json file
 | 
					    # assign supported languages from json file
 | 
				
			||||||
    if engine_data['name'] in ENGINES_LANGUAGES:
 | 
					    if engine.name in ENGINES_LANGUAGES:
 | 
				
			||||||
        setattr(engine, 'supported_languages', ENGINES_LANGUAGES[engine_data['name']])
 | 
					        engine.supported_languages = ENGINES_LANGUAGES[engine.name]
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    # find custom aliases for non standard language codes
 | 
					    # find custom aliases for non standard language codes
 | 
				
			||||||
    if hasattr(engine, 'supported_languages'):
 | 
					    for engine_lang in engine.supported_languages:
 | 
				
			||||||
        if hasattr(engine, 'language_aliases'):
 | 
					        iso_lang = match_language(engine_lang, BABEL_LANGS, fallback=None)
 | 
				
			||||||
            language_aliases = getattr(engine, 'language_aliases')
 | 
					 | 
				
			||||||
        else:
 | 
					 | 
				
			||||||
            language_aliases = {}
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
        for engine_lang in getattr(engine, 'supported_languages'):
 | 
					 | 
				
			||||||
            iso_lang = match_language(engine_lang, babel_langs, fallback=None)
 | 
					 | 
				
			||||||
        if iso_lang and iso_lang != engine_lang and not engine_lang.startswith(iso_lang) and \
 | 
					        if iso_lang and iso_lang != engine_lang and not engine_lang.startswith(iso_lang) and \
 | 
				
			||||||
               iso_lang not in getattr(engine, 'supported_languages'):
 | 
					            iso_lang not in engine.supported_languages:
 | 
				
			||||||
                language_aliases[iso_lang] = engine_lang
 | 
					            engine.language_aliases[iso_lang] = engine_lang
 | 
				
			||||||
 | 
					 | 
				
			||||||
        setattr(engine, 'language_aliases', language_aliases)
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
    # language_support
 | 
					    # language_support
 | 
				
			||||||
    setattr(engine, 'language_support', len(getattr(engine, 'supported_languages', [])) > 0)
 | 
					    engine.language_support = len(engine.supported_languages) > 0
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    # assign language fetching method if auxiliary method exists
 | 
					    # assign language fetching method if auxiliary method exists
 | 
				
			||||||
    if hasattr(engine, '_fetch_supported_languages'):
 | 
					    if hasattr(engine, '_fetch_supported_languages'):
 | 
				
			||||||
| 
						 | 
					@ -132,38 +114,60 @@ def load_engine(engine_data):
 | 
				
			||||||
            'User-Agent': gen_useragent(),
 | 
					            'User-Agent': gen_useragent(),
 | 
				
			||||||
            'Accept-Language': 'ja-JP,ja;q=0.8,en-US;q=0.5,en;q=0.3',  # bing needs a non-English language
 | 
					            'Accept-Language': 'ja-JP,ja;q=0.8,en-US;q=0.5,en;q=0.3',  # bing needs a non-English language
 | 
				
			||||||
        }
 | 
					        }
 | 
				
			||||||
        setattr(engine, 'fetch_supported_languages',
 | 
					        engine.fetch_supported_languages =\
 | 
				
			||||||
                lambda: engine._fetch_supported_languages(get(engine.supported_languages_url, headers=headers)))
 | 
					            lambda: engine._fetch_supported_languages(get(engine.supported_languages_url, headers=headers))
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    # tor related settings
 | 
					
 | 
				
			||||||
    if settings['outgoing'].get('using_tor_proxy'):
 | 
					def update_attributes_for_tor(engine):
 | 
				
			||||||
        # use onion url if using tor.
 | 
					    if settings['outgoing'].get('using_tor_proxy') and hasattr(engine, 'onion_url'):
 | 
				
			||||||
        if hasattr(engine, 'onion_url'):
 | 
					 | 
				
			||||||
        engine.search_url = engine.onion_url + getattr(engine, 'search_path', '')
 | 
					        engine.search_url = engine.onion_url + getattr(engine, 'search_path', '')
 | 
				
			||||||
    elif 'onions' in engine.categories:
 | 
					        engine.timeout += settings['outgoing'].get('extra_proxy_timeout', 0)
 | 
				
			||||||
        # exclude onion engines if not using tor.
 | 
					 | 
				
			||||||
        return None
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
    engine.timeout += settings['outgoing']['extra_proxy_timeout']
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
    for category_name in engine.categories:
 | 
					def is_missing_required_attributes(engine):
 | 
				
			||||||
        categories.setdefault(category_name, []).append(engine)
 | 
					    """an attribute is required when its name doesn't start with '_'.
 | 
				
			||||||
 | 
					    Required attributes must not be None
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
 | 
					    missing = False
 | 
				
			||||||
 | 
					    for engine_attr in dir(engine):
 | 
				
			||||||
 | 
					        if not engine_attr.startswith('_') and getattr(engine, engine_attr) is None:
 | 
				
			||||||
 | 
					            logger.error('Missing engine config attribute: "{0}.{1}"'
 | 
				
			||||||
 | 
					                         .format(engine.name, engine_attr))
 | 
				
			||||||
 | 
					            missing = True
 | 
				
			||||||
 | 
					    return missing
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					def is_engine_active(engine):
 | 
				
			||||||
 | 
					    # check if engine is inactive
 | 
				
			||||||
 | 
					    if engine.inactive is True:
 | 
				
			||||||
 | 
					        return False
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    # exclude onion engines if not using tor
 | 
				
			||||||
 | 
					    if 'onions' in engine.categories and not settings['outgoing'].get('using_tor_proxy'):
 | 
				
			||||||
 | 
					        return False
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    return True
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					def register_engine(engine):
 | 
				
			||||||
 | 
					    engines[engine.name] = engine
 | 
				
			||||||
    if engine.shortcut in engine_shortcuts:
 | 
					    if engine.shortcut in engine_shortcuts:
 | 
				
			||||||
        logger.error('Engine config error: ambigious shortcut: {0}'.format(engine.shortcut))
 | 
					        logger.error('Engine config error: ambigious shortcut: {0}'.format(engine.shortcut))
 | 
				
			||||||
        sys.exit(1)
 | 
					        sys.exit(1)
 | 
				
			||||||
 | 
					 | 
				
			||||||
    engine_shortcuts[engine.shortcut] = engine.name
 | 
					    engine_shortcuts[engine.shortcut] = engine.name
 | 
				
			||||||
 | 
					    for category_name in engine.categories:
 | 
				
			||||||
    return engine
 | 
					        categories.setdefault(category_name, []).append(engine)
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
def load_engines(engine_list):
 | 
					def load_engines(engine_list):
 | 
				
			||||||
    global engines, engine_shortcuts
 | 
					    """Use case: engine_list = settings['engines']
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
    engines.clear()
 | 
					    engines.clear()
 | 
				
			||||||
    engine_shortcuts.clear()
 | 
					    engine_shortcuts.clear()
 | 
				
			||||||
 | 
					    categories.clear()
 | 
				
			||||||
 | 
					    categories['general'] = []
 | 
				
			||||||
    for engine_data in engine_list:
 | 
					    for engine_data in engine_list:
 | 
				
			||||||
        engine = load_engine(engine_data)
 | 
					        engine = load_engine(engine_data)
 | 
				
			||||||
        if engine is not None:
 | 
					        if engine:
 | 
				
			||||||
            engines[engine.name] = engine
 | 
					            register_engine(engine)
 | 
				
			||||||
    return engines
 | 
					    return engines
 | 
				
			||||||
| 
						 | 
					
 | 
				
			||||||
		Loading…
	
	Add table
		
		Reference in a new issue