on master help sign in
Merge pull request #182 from dalf/enh-proxy

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

commit285f991cd078db7df5622e3109984f1f3c651988 authorAdam Tauber <asciimoo@gmail.com> dateSun 18 Jan 2015, 10:39 (12 years ago) parenta865e661d51512
11 files changed+39−15
searx/templates/courgette/result_templates/code.html+2−2 searx/templates/courgette/result_templates/videos.html+2−2 searx/templates/default/infobox.html+1−1 searx/templates/default/result_templates/code.html+1−1 searx/templates/default/result_templates/default.html+1−1 searx/templates/default/result_templates/map.html+1−1 searx/templates/default/result_templates/videos.html+1−1 searx/templates/oscar/infobox.html+1−1 searx/templates/oscar/result_templates/videos.html+1−1 searx/utils.py+8 searx/webapp.py+20−4
searx/templates/courgette/result_templates/code.html+2−2
@@ -1,9 +1,9 @@
11<div class="result {{ result.class }}">
22 <h3 class="result_title">{% if result['favicon'] %}<img width="14" height="14" class="favicon" src="static/{{theme}}/img/icon_{{result['favicon']}}.ico" alt="{{result['favicon']}}" />{% endif %}<a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
33 {% if result.publishedDate %}<span class="published_date">{{ result.publishedDate }}</span>{% endif %}
4− <p class="content">{% if result.img_src %}<img src="{{ result.img_src }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
4+ <p class="content">{% if result.img_src %}<img src="{{ image_proxify(result.img_src) }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
55 {% if result.repository %}<p class="content"><a href="{{ result.repository|safe }}">{{ result.repository }}</a></p>{% endif %}
66 {{ result.codelines|code_highlighter(result.code_language)|safe }}
77
88 <p class="url">{{ result.pretty_url }}</p>
9−</div>
9+</div>
searx/templates/courgette/result_templates/videos.html+2−2
@@ -5,6 +5,6 @@
55
66 <h3 class="result_title"><a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
77 {% if result.publishedDate %}<span class="published_date">{{ result.publishedDate }}</span><br />{% endif %}
8− <a href="{{ result.url }}"><img width="400" src="{{ result.thumbnail }}" title="{{ result.title|striptags }}" alt="{{ result.title|striptags }}"/></a>
8+ <a href="{{ result.url }}"><img width="400" src="{{ image_proxify(result.thumbnail) }}" title="{{ result.title|striptags }}" alt="{{ result.title|striptags }}"/></a>
99 <p class="url">{{ result.pretty_url }}</p>
10−</div>
10+</div>
searx/templates/default/infobox.html+1−1
@@ -1,6 +1,6 @@
11<div class="infobox">
22 <h2>{{ infobox.infobox }}</h2>
3− {% if infobox.img_src %}<img src="{{ infobox.img_src }}" title="{{ infobox.infobox|striptags }}" alt="{{ infobox.infobox|striptags }}" />{% endif %}
3+ {% if infobox.img_src %}<img src="{{ image_proxify(infobox.img_src) }}" title="{{ infobox.infobox|striptags }}" alt="{{ infobox.infobox|striptags }}" />{% endif %}
44 <p>{{ infobox.entity }}</p>
55 <p>{{ infobox.content | safe }}</p>
66 {% if infobox.attributes %}
searx/templates/default/result_templates/code.html+1−1
@@ -2,7 +2,7 @@
22 <h3 class="result_title"> {% if result['favicon'] %}<img width="14" height="14" class="favicon" src="static/{{theme}}/img/icon_{{result['favicon']}}.ico" alt="{{result['favicon']}}" />{% endif %}<a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
33 <p class="url">{{ result.pretty_url }} <a class="cache_link" href="https://web.archive.org/web/{{ result.url }}">cached</a></p>
44 {% if result.publishedDate %}<p class="published_date">{{ result.publishedDate }}</p>{% endif %}
5− <p class="content">{% if result.img_src %}<img src="{{ result.img_src }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
5+ <p class="content">{% if result.img_src %}<img src="{{ image_proxify(result.img_src) }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
66 {% if result.repository %}<p class="result-content"><a href="{{ result.repository|safe }}">{{ result.repository }}</a></p>{% endif %}
77
88 {{ result.codelines|code_highlighter(result.code_language)|safe }}
searx/templates/default/result_templates/default.html+1−1
@@ -2,5 +2,5 @@
22 <h3 class="result_title">{% if "icon_"~result.engine~".ico" in favicons %}<img width="14" height="14" class="favicon" src="{{ url_for('static', filename='img/icons/icon_'+result.engine+'.ico') }}" alt="{{result.engine}}" />{% endif %}<a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
33 <p class="url">{{ result.pretty_url }} <a class="cache_link" href="https://web.archive.org/web/{{ result.url }}">cached</a>
44 {% if result.publishedDate %}<span class="published_date">{{ result.publishedDate }}</span>{% endif %}</p>
5− <p class="content">{% if result.img_src %}<img src="{{ result.img_src }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
5+ <p class="content">{% if result.img_src %}<img src="{{ image_proxify(result.img_src) }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
66</div>
searx/templates/default/result_templates/map.html+1−1
@@ -8,6 +8,6 @@
88 <h3 class="result_title"><a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
99 <p class="url">{{ result.pretty_url }} <a class="cache_link" href="https://web.archive.org/web/{{ result.url }}">cached</a>
1010 {% if result.publishedDate %}<span class="published_date">{{ result.publishedDate }}</span>{% endif %}</p>
11− <p class="content">{% if result.img_src %}<img src="{{ result.img_src }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
11+ <p class="content">{% if result.img_src %}<img src="{{ image_proxify(result.img_src) }}" class="image" />{% endif %}{% if result.content %}{{ result.content|safe }}<br class="last"/>{% endif %}</p>
1212 </div>
1313</div>
searx/templates/default/result_templates/videos.html+1−1
@@ -1,6 +1,6 @@
11<div class="result">
22 <h3 class="result_title">{% if "icon_"~result.engine~".ico" in favicons %}<img width="14" height="14" class="favicon" src="{{ url_for('static', filename='img/icons/icon_'+result.engine+'.ico') }}" alt="{{result.engine}}" />{% endif %}<a href="{{ result.url }}">{{ result.title|safe }}</a></h3>
33 {% if result.publishedDate %}<span class="published_date">{{ result.publishedDate }}</span><br />{% endif %}
4− <a href="{{ result.url }}"><img class="thumbnail" src="{{ result.thumbnail }}" title="{{ result.title|striptags }}" alt="{{ result.title|striptags }}"/></a>
4+ <a href="{{ result.url }}"><img class="thumbnail" src="{{ image_proxify(result.thumbnail) }}" title="{{ result.title|striptags }}" alt="{{ result.title|striptags }}"/></a>
55 <p class="url">{{ result.url }}</p>
66</div>
searx/templates/oscar/infobox.html+1−1
@@ -3,7 +3,7 @@
33 <h4 class="panel-title">{{ infobox.infobox }}</h4>
44 </div>
55 <div class="panel-body">
6− {% if infobox.img_src %}<img class="img-responsive center-block infobox_part" src="{{ infobox.img_src }}" alt="{{ infobox.infobox }}" />{% endif %}
6+ {% if infobox.img_src %}<img class="img-responsive center-block infobox_part" src="{{ image_proxify(infobox.img_src) }}" alt="{{ infobox.infobox }}" />{% endif %}
77 {% if infobox.content %}<p class="infobox_part">{{ infobox.content }}</p>{% endif %}
88
99 {% if infobox.attributes %}
searx/templates/oscar/result_templates/videos.html+1−1
@@ -15,7 +15,7 @@
1515
1616<div class="container-fluid">
1717 <div class="row">
18− <a href="{{ result.url }}"><img class="thumbnail col-xs-6 col-sm-4 col-md-4 result-content" src="{{ result.thumbnail|safe }}" alt="{{ result.title|striptags }} {{ result.engine }}" /></a>
18+ <a href="{{ result.url }}"><img class="thumbnail col-xs-6 col-sm-4 col-md-4 result-content" src="{{ image_proxify(result.thumbnail) }}" alt="{{ result.title|striptags }} {{ result.engine }}" /></a>
1919 {% if result.content %}<p class="col-xs-12 col-sm-8 col-md-8 result-content">{{ result.content|safe }}</p>{% endif %}
2020 </div>
2121</div>
searx/utils.py+8
@@ -206,3 +206,11 @@ def format_date_by_locale(date_string, locale_string):
206206 except:
207207 logger.warning('cannot set original locale: {0}'.format(orig_locale))
208208 return formatted_date
209+
210+
211+def dict_subset(d, properties):
212+ result = {}
213+ for k in properties:
214+ if k in d:
215+ result[k] = d[k]
216+ return result
searx/webapp.py+20−4
@@ -25,6 +25,7 @@ if __name__ == '__main__':
2525import json
2626import cStringIO
2727import os
28+import hashlib
2829
2930from datetime import datetime, timedelta
3031from requests import get as http_get
@@ -41,7 +42,7 @@ from searx.engines import (
4142)
4243from searx.utils import (
4344 UnicodeWriter, highlight_content, html_to_text, get_themes,
44− get_static_files, get_result_templates, gen_useragent
45+ get_static_files, get_result_templates, gen_useragent, dict_subset
4546)
4647from searx.version import VERSION_STRING
4748from searx.languages import language_codes
@@ -216,8 +217,10 @@ def image_proxify(url):
216217 if not settings['server'].get('image_proxy') and not request.cookies.get('image_proxy'):
217218 return url
218219
220+ h = hashlib.sha256(url + settings['server']['secret_key']).hexdigest()
221+
219222 return '{0}?{1}'.format(url_for('image_proxy'),
220− urlencode(dict(url=url)))
223+ urlencode(dict(url=url, h=h)))
221224
222225
223226def render(template_name, override_theme=None, **kwargs):
@@ -562,10 +565,21 @@ def image_proxy():
562565 if not url:
563566 return '', 400
564567
568+ h = hashlib.sha256(url + settings['server']['secret_key']).hexdigest()
569+
570+ if h != request.args.get('h'):
571+ return '', 400
572+
573+ headers = dict_subset(request.headers, {'If-Modified-Since', 'If-None-Match'})
574+ headers['User-Agent'] = gen_useragent()
575+
565576 resp = http_get(url,
566577 stream=True,
567578 timeout=settings['server'].get('request_timeout', 2),
568− headers={'User-Agent': gen_useragent()})
579+ headers=headers)
580+
581+ if resp.status_code == 304:
582+ return '', resp.status_code
569583
570584 if resp.status_code != 200:
571585 logger.debug('image-proxy: wrong response code: {0}'.format(resp.status_code))
@@ -586,7 +600,9 @@ def image_proxy():
586600 return '', 502 # Bad gateway - file is too big (>5M)
587601 img += chunk
588602
589− return Response(img, mimetype=resp.headers['content-type'])
603+ headers = dict_subset(resp.headers, {'Content-Length', 'Length', 'Date', 'Last-Modified', 'Expires', 'Etag'})
604+
605+ return Response(img, mimetype=resp.headers['content-type'], headers=headers)
590606
591607
592608@app.route('/stats', methods=['GET'])