From cd7849ffc82dc319e25ed80cae03dc021e5c7263 Mon Sep 17 00:00:00 2001
From: Denis Wernert <deniswernert@gmail.com>
Date: Tue, 2 Oct 2018 15:08:43 +0200
Subject: Adds the Unsplash image engine

---
 searx/engines/unsplash.py | 39 +++++++++++++++++++++++++++++++++++++++
 searx/settings.yml        |  5 +++++
 2 files changed, 44 insertions(+)
 create mode 100644 searx/engines/unsplash.py

(limited to 'searx')

diff --git a/searx/engines/unsplash.py b/searx/engines/unsplash.py
new file mode 100644
index 000000000..03db97788
--- /dev/null
+++ b/searx/engines/unsplash.py
@@ -0,0 +1,39 @@
+"""
+ Unsplash
+
+ @website     https://unsplash.com
+ @provide-api yes (https://unsplash.com/developers)
+
+ @using-api   no
+ @results     JSON (using search portal's infiniscroll API)
+ @stable      no (JSON format could change any time)
+ @parse       url, title, img_src, thumbnail_src
+"""
+
+from searx.url_utils import urlencode
+from json import loads
+
+url = 'https://unsplash.com/'
+search_url = url + 'napi/search/photos?'
+categories = ['images']
+page_size = 20
+paging = True
+
+
+def request(query, params):
+    params['url'] = search_url + urlencode({'query': query, 'page': params['pageno'], 'per_page': page_size})
+    return params
+
+
+def response(resp):
+    results = []
+    json_data = loads(resp.text)
+
+    for result in json_data['results']:
+        results.append({'template': 'images.html',
+                        'url': result['links']['html'],
+                        'thumbnail_src': result['urls']['thumb'],
+                        'img_src': result['urls']['full'],
+                        'title': result['description'],
+                        'content': ''})
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
index 6e1b5fb74..b48fada8e 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -642,6 +642,11 @@ engines:
 #    content_xpath : //*[@class="meaning"]
 #    shortcut : ud
 
+  - name : unsplash
+    engine : unsplash
+    disabled: True
+    shortcut : us
+
   - name : yahoo
     engine : yahoo
     shortcut : yh
-- 
cgit v1.2.3


From 72d063d27d9277b12efe2ec16fe1c95ef651c2a2 Mon Sep 17 00:00:00 2001
From: Denis Wernert <deniswernert@gmail.com>
Date: Mon, 8 Oct 2018 14:01:35 +0200
Subject: Uses the raw url for the image result, rather than the full size
 result.

---
 searx/engines/unsplash.py | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'searx')

diff --git a/searx/engines/unsplash.py b/searx/engines/unsplash.py
index 03db97788..04a943297 100644
--- a/searx/engines/unsplash.py
+++ b/searx/engines/unsplash.py
@@ -29,11 +29,12 @@ def response(resp):
     results = []
     json_data = loads(resp.text)
 
-    for result in json_data['results']:
-        results.append({'template': 'images.html',
-                        'url': result['links']['html'],
-                        'thumbnail_src': result['urls']['thumb'],
-                        'img_src': result['urls']['full'],
-                        'title': result['description'],
-                        'content': ''})
+    if 'results' in json_data:
+        for result in json_data['results']:
+            results.append({'template': 'images.html',
+                            'url': result['links']['html'],
+                            'thumbnail_src': result['urls']['thumb'],
+                            'img_src': result['urls']['raw'],
+                            'title': result['description'],
+                            'content': ''})
     return results
-- 
cgit v1.2.3


From b9ada93b3ade2b4268bdc898e2c67b156b4dba92 Mon Sep 17 00:00:00 2001
From: Denis Wernert <deniswernert@gmail.com>
Date: Mon, 8 Oct 2018 14:56:20 +0200
Subject: Removes what looks like tracking parameters

---
 searx/engines/unsplash.py | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

(limited to 'searx')

diff --git a/searx/engines/unsplash.py b/searx/engines/unsplash.py
index 04a943297..2e8d6fdfc 100644
--- a/searx/engines/unsplash.py
+++ b/searx/engines/unsplash.py
@@ -10,7 +10,7 @@
  @parse       url, title, img_src, thumbnail_src
 """
 
-from searx.url_utils import urlencode
+from searx.url_utils import urlencode, urlparse, urlunparse, parse_qsl
 from json import loads
 
 url = 'https://unsplash.com/'
@@ -20,6 +20,18 @@ page_size = 20
 paging = True
 
 
+def clean_url(url):
+    parsed = urlparse(url)
+    query = [(k, v) for (k, v) in parse_qsl(parsed.query) if k not in ['ixid', 's']]
+
+    return urlunparse((parsed.scheme,
+                       parsed.netloc,
+                       parsed.path,
+                       parsed.params,
+                       urlencode(query),
+                       parsed.fragment))
+
+
 def request(query, params):
     params['url'] = search_url + urlencode({'query': query, 'page': params['pageno'], 'per_page': page_size})
     return params
@@ -32,9 +44,9 @@ def response(resp):
     if 'results' in json_data:
         for result in json_data['results']:
             results.append({'template': 'images.html',
-                            'url': result['links']['html'],
-                            'thumbnail_src': result['urls']['thumb'],
-                            'img_src': result['urls']['raw'],
+                            'url': clean_url(result['links']['html']),
+                            'thumbnail_src': clean_url(result['urls']['thumb']),
+                            'img_src': clean_url(result['urls']['raw']),
                             'title': result['description'],
                             'content': ''})
     return results
-- 
cgit v1.2.3