searx.engines.qwant Namespace Reference

Functions
	request (query, params)
	response (resp)
	parse_web_lite (resp)
	parse_web_api (resp)
	fetch_traits (EngineTraits engine_traits)

Variables
dict	about
list	categories = []
bool	paging = True
int	max_page = 5
	qwant_categ = None
bool	safesearch = True
list	qwant_news_locales
str	api_url = 'https://api.qwant.com/v3/search/'
str	web_lite_url = 'https://lite.qwant.com/'

Detailed Description

This engine uses the Qwant API (https://api.qwant.com/v3) to implement Qwant
-Web, -News, -Images and -Videos.  The API is undocumented but can be reverse
engineered by reading the network log of https://www.qwant.com/ queries.

For Qwant's *web-search* two alternatives are implemented:

- ``web``: uses the :py:obj:`api_url` which returns a JSON structure
- ``web-lite``: uses the :py:obj:`web_lite_url` which returns a HTML page


Configuration
=============

The engine has the following additional settings:

- :py:obj:`qwant_categ`

This implementation is used by different qwant engines in the :ref:`settings.yml
<settings engines>`:

.. code:: yaml

  - name: qwant
    qwant_categ: web-lite  # alternatively use 'web'
    ...
  - name: qwant news
    qwant_categ: news
    ...
  - name: qwant images
    qwant_categ: images
    ...
  - name: qwant videos
    qwant_categ: videos
    ...

Implementations
===============

Function Documentation

fetch_traits()

searx.engines.qwant.fetch_traits

(

EngineTraits

engine_traits

)

Definition at line 319 of file qwant.py.

def fetch_traits(engine_traits: EngineTraits):
 
    # pylint: disable=import-outside-toplevel
    from searx import network
    from searx.locales import region_tag
    from searx.utils import extr
 
    resp = network.get(about['website'])
    json_string = extr(resp.text, 'INITIAL_PROPS = ', '</script>')
 
    q_initial_props = loads(json_string)
    q_locales = q_initial_props.get('locales')
    eng_tag_list = set()
 
    for country, v in q_locales.items():
        for lang in v['langs']:
            _locale = "{lang}_{country}".format(lang=lang, country=country)
 
            if qwant_categ == 'news' and _locale.lower() not in qwant_news_locales:
                # qwant-news does not support all locales from qwant-web:
                continue
 
            eng_tag_list.add(_locale)
 
    for eng_tag in eng_tag_list:
        try:
            sxng_tag = region_tag(babel.Locale.parse(eng_tag, sep='_'))
        except babel.UnknownLocaleError:
            print("ERROR: can't determine babel locale of quant's locale %s" % eng_tag)
            continue
 
        conflict = engine_traits.regions.get(sxng_tag)
        if conflict:
            if conflict != eng_tag:
                print("CONFLICT: babel %s --> %s, %s" % (sxng_tag, conflict, eng_tag))
            continue
        engine_traits.regions[sxng_tag] = eng_tag

parse_web_api()

searx.engines.qwant.parse_web_api

(

resp

)

Parse results from Qwant's API

Definition at line 181 of file qwant.py.

def parse_web_api(resp):
    """Parse results from Qwant's API"""
    # pylint: disable=too-many-locals, too-many-branches, too-many-statements
 
    results = []
 
    # load JSON result
    search_results = loads(resp.text)
    data = search_results.get('data', {})
 
    # check for an API error
    if search_results.get('status') != 'success':
        error_code = data.get('error_code')
        if error_code == 24:
            raise SearxEngineTooManyRequestsException()
        if search_results.get("data", {}).get("error_data", {}).get("captchaUrl") is not None:
            raise SearxEngineCaptchaException()
        msg = ",".join(data.get('message', ['unknown']))
        raise SearxEngineAPIException(f"{msg} ({error_code})")
 
    # raise for other errors
    raise_for_httperror(resp)
 
    if qwant_categ == 'web':
        # The WEB query contains a list named 'mainline'.  This list can contain
        # different result types (e.g. mainline[0]['type'] returns type of the
        # result items in mainline[0]['items']
        mainline = data.get('result', {}).get('items', {}).get('mainline', {})
    else:
        # Queries on News, Images and Videos do not have a list named 'mainline'
        # in the response.  The result items are directly in the list
        # result['items'].
        mainline = data.get('result', {}).get('items', [])
        mainline = [
            {'type': qwant_categ, 'items': mainline},
        ]
 
    # return empty array if there are no results
    if not mainline:
        return []
 
    for row in mainline:
        mainline_type = row.get('type', 'web')
        if mainline_type != qwant_categ:
            continue
 
        if mainline_type == 'ads':
            # ignore adds
            continue
 
        mainline_items = row.get('items', [])
        for item in mainline_items:
 
            title = item.get('title', None)
            res_url = item.get('url', None)
 
            if mainline_type == 'web':
                content = item['desc']
                results.append(
                    {
                        'title': title,
                        'url': res_url,
                        'content': content,
                    }
                )
 
            elif mainline_type == 'news':
 
                pub_date = item['date']
                if pub_date is not None:
                    pub_date = datetime.fromtimestamp(pub_date)
                news_media = item.get('media', [])
                thumbnail = None
                if news_media:
                    thumbnail = news_media[0].get('pict', {}).get('url', None)
                results.append(
                    {
                        'title': title,
                        'url': res_url,
                        'publishedDate': pub_date,
                        'thumbnail': thumbnail,
                    }
                )
 
            elif mainline_type == 'images':
                thumbnail = item['thumbnail']
                img_src = item['media']
                results.append(
                    {
                        'title': title,
                        'url': res_url,
                        'template': 'images.html',
                        'thumbnail_src': thumbnail,
                        'img_src': img_src,
                        'resolution': f"{item['width']} x {item['height']}",
                        'img_format': item.get('thumb_type'),
                    }
                )
 
            elif mainline_type == 'videos':
                # some videos do not have a description: while qwant-video
                # returns an empty string, such video from a qwant-web query
                # miss the 'desc' key.
                d, s, c = item.get('desc'), item.get('source'), item.get('channel')
                content_parts = []
                if d:
                    content_parts.append(d)
                if s:
                    content_parts.append("%s: %s " % (gettext("Source"), s))
                if c:
                    content_parts.append("%s: %s " % (gettext("Channel"), c))
                content = ' // '.join(content_parts)
                length = item['duration']
                if length is not None:
                    length = timedelta(milliseconds=length)
                pub_date = item['date']
                if pub_date is not None:
                    pub_date = datetime.fromtimestamp(pub_date)
                thumbnail = item['thumbnail']
                # from some locations (DE and others?) the s2 link do
                # response a 'Please wait ..' but does not deliver the thumbnail
                thumbnail = thumbnail.replace('https://s2.qwant.com', 'https://s1.qwant.com', 1)
                results.append(
                    {
                        'title': title,
                        'url': res_url,
                        'content': content,
                        'iframe_src': get_embeded_stream_url(res_url),
                        'publishedDate': pub_date,
                        'thumbnail': thumbnail,
                        'template': 'videos.html',
                        'length': length,
                    }
                )
 
    return results
 
 

Referenced by response().

Here is the caller graph for this function:

parse_web_lite()

searx.engines.qwant.parse_web_lite

(

resp

)

Parse results from Qwant-Lite

Definition at line 160 of file qwant.py.

def parse_web_lite(resp):
    """Parse results from Qwant-Lite"""
 
    results = []
    dom = lxml.html.fromstring(resp.text)
 
    for item in eval_xpath_list(dom, '//section/article'):
        if eval_xpath(item, "./span[contains(@class, 'tooltip')]"):
            # ignore randomly interspersed advertising adds
            continue
        results.append(
            {
                'url': extract_text(eval_xpath(item, "./span[contains(@class, 'url partner')]")),
                'title': extract_text(eval_xpath(item, './h2/a')),
                'content': extract_text(eval_xpath(item, './p')),
            }
        )
 
    return results
 
 

Referenced by response().

Here is the caller graph for this function:

request()

searx.engines.qwant.request	(		query,
			params )

Qwant search request

Definition at line 109 of file qwant.py.

def request(query, params):
    """Qwant search request"""
 
    if not query:
        return None
 
    q_locale = traits.get_region(params["searxng_locale"], default='en_US')
 
    url = api_url + f'{qwant_categ}?'
    args = {'q': query}
    params['raise_for_httperror'] = False
 
    if qwant_categ == 'web-lite':
 
        url = web_lite_url + '?'
        args['locale'] = q_locale.lower()
        args['l'] = q_locale.split('_')[0]
        args['s'] = params['safesearch']
        args['p'] = params['pageno']
 
        params['raise_for_httperror'] = True
 
    elif qwant_categ == 'images':
 
        args['locale'] = q_locale
        args['safesearch'] = params['safesearch']
        args['count'] = 50
        args['tgp'] = 3
        args['offset'] = (params['pageno'] - 1) * args['count']
 
    else:  # web, news, videos
 
        args['locale'] = q_locale
        args['safesearch'] = params['safesearch']
        args['count'] = 10
        args['llm'] = 'false'
        args['tgp'] = 3
        args['offset'] = (params['pageno'] - 1) * args['count']
 
    params['url'] = url + urlencode(args)
 
    return params
 
 

response()

searx.engines.qwant.response

(

resp

)

Definition at line 153 of file qwant.py.

def response(resp):
 
    if qwant_categ == 'web-lite':
        return parse_web_lite(resp)
    return parse_web_api(resp)
 
 

References parse_web_api(), and parse_web_lite().

Here is the call graph for this function:

Variable Documentation

about

dict searx.engines.qwant.about

Initial value:

=  {
    "website": 'https://www.qwant.com/',
    "wikidata_id": 'Q14657870',
    "official_api_documentation": None,
    "use_official_api": True,
    "require_api_key": False,
    "results": 'JSON',
}

Definition at line 68 of file qwant.py.

api_url

str searx.engines.qwant.api_url = 'https://api.qwant.com/v3/search/'

Definition at line 102 of file qwant.py.

categories

list searx.engines.qwant.categories = []

Definition at line 78 of file qwant.py.

max_page

int searx.engines.qwant.max_page = 5

Definition at line 80 of file qwant.py.

paging

bool searx.engines.qwant.paging = True

Definition at line 79 of file qwant.py.

qwant_categ

searx.engines.qwant.qwant_categ = None

Definition at line 84 of file qwant.py.

qwant_news_locales

list searx.engines.qwant.qwant_news_locales

Initial value:

=  [
    'ca_ad', 'ca_es', 'ca_fr', 'co_fr', 'de_at', 'de_ch', 'de_de', 'en_au',
    'en_ca', 'en_gb', 'en_ie', 'en_my', 'en_nz', 'en_us', 'es_ad', 'es_ar',
    'es_cl', 'es_co', 'es_es', 'es_mx', 'es_pe', 'eu_es', 'eu_fr', 'fc_ca',
    'fr_ad', 'fr_be', 'fr_ca', 'fr_ch', 'fr_fr', 'it_ch', 'it_it', 'nl_be',
    'nl_nl', 'pt_ad', 'pt_pt',
]

Definition at line 91 of file qwant.py.

safesearch

bool searx.engines.qwant.safesearch = True

Definition at line 87 of file qwant.py.

web_lite_url

str searx.engines.qwant.web_lite_url = 'https://lite.qwant.com/'

Definition at line 105 of file qwant.py.