zaclys/searxng
1
0
Fork 1
mirror of https://github.com/searxng/searxng synced 2024-01-01 18:24:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

[fix] wikidata info box images

Browse Source 
Wikidata info box images are now loaded from uploads.wikimedia.org instead of commons.wikimedia.org to prevent redirects

Co-authored-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Léon Tiekötter 2022-02-06 17:50:00 +01:00
parent 10e6881ca8
commit 560a14e77b
No known key found for this signature in database
GPG Key ID: 02869F7817D9EC08
1 changed files with 32 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
searx/engines/wikidata.py
View File

@ -18,6 +18,7 @@ from searx.engines.wikipedia import ( # pylint: disable=unused-import
_fetch_supported_languages,
supported_languages_url,
)
from hashlib import md5
# about
about = {
@ -185,6 +186,36 @@ def response(resp):
return results
_IMG_SRC_DEFAULT_URL_PREFIX = "https://commons.wikimedia.org/wiki/Special:FilePath/"
_IMG_SRC_NEW_URL_PREFIX = "https://upload.wikimedia.org/wikipedia/commons/"
def get_thumbnail(img_src):
"""Get Thumbnail image from wikimedia commons
Images from commons.wikimedia.org are (HTTP) redirected to
upload.wikimedia.org. The redirected URL can be calculated by this
function.
- https://stackoverflow.com/a/33691240
"""
logger.debug('get_thumbnail(): %s', img_src)
if not img_src is None and _IMG_SRC_DEFAULT_URL_PREFIX in img_src.split()[0]:
img_src_name = (
img_src.replace(_IMG_SRC_DEFAULT_URL_PREFIX, "")
.split("?", 1)[0]
.replace("%20", "_")
.replace("%28", "(")
.replace("%29", ")")
)
img_src_name_md5 = md5(img_src_name.encode("utf-8")).hexdigest()
img_src = _IMG_SRC_NEW_URL_PREFIX + img_src_name_md5[0] + "/" + img_src_name_md5[0:2] + "/" + img_src_name
logger.debug('get_thumbnail() redirected: %s', img_src)
return img_src
def get_results(attribute_result, attributes, language):
results = []
infobox_title = attribute_result.get('itemLabel')
@ -221,7 +252,7 @@ def get_results(attribute_result, attributes, language):
# replace the current image only the priority is lower
# (the infobox contain only one image).
if attribute.priority < img_src_priority:
img_src = value
img_src = get_thumbnail(value)
img_src_priority = attribute.priority
elif attribute_type == WDGeoAttribute:
# geocoordinate link