[enh] add engine - Docker Hub

Slightly modified merge of commit [1cb1d3ac] from searx [PR 2543]:

      This adds Docker Hub .. as a search engine .. the engine's favicon was
      downloaded from the Docker Hub website with wget and converted to a PNG
      with ImageMagick .. It supports the parsing of URLs, titles, content,
      published dates, and thumbnails of Docker images.

[1cb1d3ac] https://github.com/searx/searx/pull/2543/commits/1cb1d3ac
[PR 2543] https://github.com/searx/searx/pull/2543

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>

1 files changed, 65 insertions, 0 deletions

diff --git a/searx/engines/docker_hub.py b/searx/engines/docker_hub.py
new file mode 100644
index 000000000..d9d0f745b
--- /dev/null
+++ b/searx/engines/docker_hub.py
@@ -0,0 +1,65 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# lint: pylint
+# pylint: disable=missing-function-docstring
+"""Docker Hub (IT)
+
+"""
+
+from json import loads
+from urllib.parse import urlencode
+from dateutil import parser
+
+about = {
+     "website": 'https://hub.docker.com',
+     "wikidata_id": 'Q100769064',
+     "official_api_documentation": 'https://docs.docker.com/registry/spec/api/',
+     "use_official_api": True,
+     "require_api_key": False,
+     "results": 'JSON',
+ }
+
+categories = ['it']  # optional
+paging = True
+
+base_url = "https://hub.docker.com/"
+search_url = base_url + "api/content/v1/products/search?{query}&type=image&page_size=25"
+
+def request(query, params):
+
+    params['url'] = search_url.format(query=urlencode(dict(q=query, page=params["pageno"])))
+    params["headers"]["Search-Version"] = "v3"
+
+    return params
+
+def response(resp):
+    '''post-response callback
+    resp: requests response object
+    '''
+    results = []
+    body = loads(resp.text)
+
+    # Make sure `summaries` isn't `null`
+    search_res = body.get("summaries")
+    if search_res:
+        for item in search_res:
+            result = {}
+
+            # Make sure correct URL is set
+            filter_type = item.get("filter_type")
+            is_official = filter_type in ["store", "official"]
+
+            if is_official:
+                result["url"] = base_url + "_/" + item.get('slug', "")
+            else:
+                result["url"] = base_url + "r/" + item.get('slug', "")
+            result["title"] = item.get("name")
+            result["content"] = item.get("short_description")
+            result["publishedDate"] = parser.parse(
+                item.get("updated_at") or item.get("created_at")
+            )
+            result["thumbnail"] = (
+                item["logo_url"].get("large") or item["logo_url"].get("small")
+            )
+            results.append(result)
+
+    return results