searx

commit: b6d27aca590c034710544a982bca2b71abb878c5
parent a865e6672fa2a289937ff378e8893c19f71c5e0c
Author: dalf <alex@al-f.net>
Date:   Sat, 17 Jan 2015 21:54:40 +0100

[enh] image-proxy : handle ETag and date related headers, add hash to URL

Diffstat:
M
searx/utils.py
8
++++++++
M
searx/webapp.py
24
++++++++++++++++++++----

2 files changed, 28 insertions(+), 4 deletions(-)
diff --git a/searx/utils.py b/searx/utils.py
@@ -206,3 +206,11 @@ def format_date_by_locale(date_string, locale_string):
     except:
         logger.warning('cannot set original locale: {0}'.format(orig_locale))
     return formatted_date
+
+
+def dict_subset(d, properties):
+    result = {}
+    for k in properties:
+        if k in d:
+            result[k] = d[k]
+    return result
diff --git a/searx/webapp.py b/searx/webapp.py
@@ -25,6 +25,7 @@ if __name__ == '__main__':
 import json
 import cStringIO
 import os
+import hashlib
 
 from datetime import datetime, timedelta
 from requests import get as http_get
@@ -41,7 +42,7 @@ from searx.engines import (
 )
 from searx.utils import (
     UnicodeWriter, highlight_content, html_to_text, get_themes,
-    get_static_files, get_result_templates, gen_useragent
+    get_static_files, get_result_templates, gen_useragent, dict_subset
 )
 from searx.version import VERSION_STRING
 from searx.languages import language_codes
@@ -213,11 +214,13 @@ def image_proxify(url):
     if url.startswith('//'):
         url = 'https:' + url
 
+    h = hashlib.sha256(url + settings['server']['secret_key']).hexdigest()
+
     if not settings['server'].get('image_proxy') and not request.cookies.get('image_proxy'):
         return url
 
     return '{0}?{1}'.format(url_for('image_proxy'),
-                            urlencode(dict(url=url)))
+                            urlencode(dict(url=url, h=h)))
 
 
 def render(template_name, override_theme=None, **kwargs):
@@ -562,10 +565,21 @@ def image_proxy():
     if not url:
         return '', 400
 
+    h = hashlib.sha256(url + settings['server']['secret_key']).hexdigest()
+
+    if h != request.args.get('h'):
+        return '', 400
+
+    headers = dict_subset(request.headers, {'If-Modified-Since', 'If-None-Match'})
+    headers['User-Agent'] = gen_useragent()
+
     resp = http_get(url,
                     stream=True,
                     timeout=settings['server'].get('request_timeout', 2),
-                    headers={'User-Agent': gen_useragent()})
+                    headers=headers)
+
+    if resp.status_code == 304:
+        return '', resp.status_code
 
     if resp.status_code != 200:
         logger.debug('image-proxy: wrong response code: {0}'.format(resp.status_code))
@@ -586,7 +600,9 @@ def image_proxy():
             return '', 502  # Bad gateway - file is too big (>5M)
         img += chunk
 
-    return Response(img, mimetype=resp.headers['content-type'])
+    headers = dict_subset(resp.headers, {'Content-Length', 'Length', 'Date', 'Last-Modified', 'Expires', 'Etag'})
+
+    return Response(img, mimetype=resp.headers['content-type'], headers=headers)
 
 
 @app.route('/stats', methods=['GET'])