Ponysearch/searx/engines/json_engine.py

163 lines
4.2 KiB
Python
Raw Permalink Normal View History

# SPDX-License-Identifier: AGPL-3.0-or-later
"""The JSON engine is a *generic* engine with which it is possible to configure
engines in the settings.
.. todo::
- The JSON engine needs documentation!!
- The parameters of the JSON engine should be adapted to those of the XPath
engine.
"""
from collections.abc import Iterable
2016-11-30 18:43:03 +01:00
from json import loads
from urllib.parse import urlencode
from searx.utils import to_string, html_to_text
2016-11-30 18:43:03 +01:00
2013-11-19 15:49:52 +01:00
2014-01-20 02:31:20 +01:00
search_url = None
url_query = None
2023-10-20 11:26:28 +02:00
url_prefix = ""
2013-11-19 15:49:52 +01:00
content_query = None
2014-01-20 02:31:20 +01:00
title_query = None
content_html_to_text = False
title_html_to_text = False
2016-11-30 18:43:03 +01:00
paging = False
2016-08-12 12:15:27 +02:00
suggestion_query = ''
results_query = ''
2013-11-19 15:49:52 +01:00
cookies = {}
headers = {}
'''Some engines might offer different result based on cookies or headers.
Possible use-case: To set safesearch cookie or header to moderate.'''
# parameters for engines with paging support
#
# number of results on each page
# (only needed if the site requires not a page number, but an offset)
page_size = 1
# number of the first page (usually 0 or 1)
first_page_num = 1
2014-01-20 02:31:20 +01:00
2013-11-19 15:49:52 +01:00
def iterate(iterable):
if isinstance(iterable, dict):
items = iterable.items()
2013-11-19 15:49:52 +01:00
else:
items = enumerate(iterable)
for index, value in items:
2013-11-19 15:49:52 +01:00
yield str(index), value
2014-01-20 02:31:20 +01:00
2013-11-19 15:49:52 +01:00
def is_iterable(obj):
if isinstance(obj, str):
2014-01-20 02:31:20 +01:00
return False
2013-11-19 15:49:52 +01:00
return isinstance(obj, Iterable)
2014-01-20 02:31:20 +01:00
def parse(query): # pylint: disable=redefined-outer-name
q = [] # pylint: disable=invalid-name
2013-11-19 15:49:52 +01:00
for part in query.split('/'):
if part == '':
continue
q.append(part)
2013-11-19 15:49:52 +01:00
return q
2014-01-20 02:31:20 +01:00
def do_query(data, q): # pylint: disable=invalid-name
2013-11-19 15:49:52 +01:00
ret = []
2014-02-11 13:13:51 +01:00
if not q:
2013-11-19 15:49:52 +01:00
return ret
qkey = q[0]
2014-01-20 02:31:20 +01:00
for key, value in iterate(data):
2013-11-19 15:49:52 +01:00
if len(q) == 1:
if key == qkey:
ret.append(value)
elif is_iterable(value):
ret.extend(do_query(value, q))
else:
if not is_iterable(value):
continue
if key == qkey:
ret.extend(do_query(value, q[1:]))
else:
ret.extend(do_query(value, q))
return ret
2014-01-20 02:31:20 +01:00
2013-11-19 15:49:52 +01:00
def query(data, query_string):
q = parse(query_string)
return do_query(data, q)
2014-01-20 02:31:20 +01:00
def request(query, params): # pylint: disable=redefined-outer-name
2013-11-19 15:49:52 +01:00
query = urlencode({'q': query})[2:]
fp = {'query': query} # pylint: disable=invalid-name
if paging and search_url.find('{pageno}') >= 0:
fp['pageno'] = (params['pageno'] - 1) * page_size + first_page_num
params['cookies'].update(cookies)
params['headers'].update(headers)
params['url'] = search_url.format(**fp)
2013-11-19 15:49:52 +01:00
params['query'] = query
2013-11-19 15:49:52 +01:00
return params
def identity(arg):
return arg
2013-11-19 15:49:52 +01:00
def response(resp):
results = []
json = loads(resp.text)
title_filter = html_to_text if title_html_to_text else identity
content_filter = html_to_text if content_html_to_text else identity
2016-08-12 12:15:27 +02:00
if results_query:
rs = query(json, results_query) # pylint: disable=invalid-name
if not rs:
return results
for result in rs[0]:
try:
url = query(result, url_query)[0]
title = query(result, title_query)[0]
except: # pylint: disable=bare-except
continue
try:
content = query(result, content_query)[0]
except: # pylint: disable=bare-except
content = ""
results.append(
{
2023-10-20 11:26:28 +02:00
'url': url_prefix + to_string(url),
'title': title_filter(to_string(title)),
'content': content_filter(to_string(content)),
}
)
2016-08-12 12:15:27 +02:00
else:
for url, title, content in zip(query(json, url_query), query(json, title_query), query(json, content_query)):
results.append(
{
2023-10-20 11:26:28 +02:00
'url': url_prefix + to_string(url),
'title': title_filter(to_string(title)),
'content': content_filter(to_string(content)),
}
)
2016-08-12 12:15:27 +02:00
if not suggestion_query:
return results
for suggestion in query(json, suggestion_query):
results.append({'suggestion': suggestion})
2013-11-19 15:49:52 +01:00
return results