zaclys/searxng
1
0
Fork 1
mirror of https://github.com/searxng/searxng synced 2024-01-01 19:24:07 +01:00
Code Issues Projects Releases Packages Wiki Activity

feat(google_cs): Add more metadata

Browse source
This commit is contained in:
Manatsawin Hanmongkolchai 2022-10-09 19:51:45 +07:00
parent 99cc43b9b7
commit d898cfcb38
1 changed files with 29 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

39
searx/engines/google_cs.py
View file

@ -37,8 +37,10 @@
""" """
from urllib.parse import urlencode from urllib.parse import urlencode
from dateutil.parser import isoparse
from searx.engines.google import get_lang_info from searx.engines.google import get_lang_info
from searx.exceptions import SearxEngineAPIException, SearxEngineTooManyRequestsException from searx.exceptions import SearxEngineTooManyRequestsException
from searx.network import raise_for_httperror from searx.network import raise_for_httperror
about = { about = {
@ -215,20 +217,37 @@ def response(resp):
return metadata + [_convert_result(search, search_type) for search in result.get('items', [])] return metadata + [_convert_result(search, search_type) for search in result.get('items', [])]
def _convert_result(search, search_type=''): def _convert_result(result, search_type=''):
"""Convert `result JSON <https://developers.google.com/custom-search/v1/reference/rest/v1/Search#Result>`_ """Convert `result JSON <https://developers.google.com/custom-search/v1/reference/rest/v1/Search#Result>`_
to Searx result""" to Searx result"""
out = { out = {
"url": search['link'], "url": result['link'],
"title": search['title'], "title": result['title'],
"content": search.get('snippet', ''), "content": result.get('snippet', ''),
} }
if search_type == 'image' and 'image' in search: try:
dt = result['pagemap']['metatags'][0]['date']
parsed_dt = isoparse(dt)
out['publishedDate'] = parsed_dt
except (KeyError, IndexError):
pass
try:
out['author'] = result['pagemap']['metatags'][0]['author']
except (KeyError, IndexError):
pass
try:
out['img_src'] = result['pagemap']['cse_thumbnail'][0]['src']
except (KeyError, IndexError):
pass
if search_type == 'image' and 'image' in result:
out['template'] = 'images.html' out['template'] = 'images.html'
out['img_src'] = search['link'] out['img_src'] = result['link']
out['thumbnail_src'] = search['image']['thumbnailLink'] out['thumbnail_src'] = result['image']['thumbnailLink']
out['img_format'] = f"{search['image']['width']} x {search['image']['height']} {search['fileFormat']}" out['img_format'] = f"{result['image']['width']} x {result['image']['height']} {result['fileFormat']}"
out['url'] = search['image']['contextLink'] out['url'] = result['image']['contextLink']
return out return out