diff --git a/cps/templates/detail.html b/cps/templates/detail.html
index 2e685d36f..3eff7483 100644
--- a/cps/templates/detail.html
+++ b/cps/templates/detail.html
@@ -41,22 +41,35 @@
 
             {% if g.user.kindle_mail and g.user.is_authenticated %}
             <a href="{{url_for('send_to_kindle', book_id=entry.id)}}" id="sendbtn" class="btn btn-primary" role="button"><span class="glyphicon glyphicon-send"></span> {{_('Send to Kindle')}}</a>
-            {% endif %}
-          {% if (g.user.role_download() and g.user.is_anonymous) or g.user.is_authenticated %}
-              <div class="btn-group" role="group">
-                <button id="read-in-browser" type="button" class="btn btn-primary dropdown-toggle" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">
-                  <span class="glyphicon glyphicon-eye-open"></span> {{_('Read in browser')}}
-                  <span class="caret"></span>
-                </button>
-                <ul class="dropdown-menu" aria-labelledby="read-in-browser">
-                  {% for format in entry.data %}
-                    {%if format.format|lower == 'epub' or format.format|lower == 'txt' or format.format|lower == 'pdf' or ( format.format|lower == 'cbr' and rarsupport ) or format.format|lower == 'cbt' or format.format|lower == 'cbz' %}
-                      <li><a target="_blank" href="{{ url_for('read_book', book_id=entry.id, book_format=format.format|lower) }}">{{format.format}}</a></li>
-                    {% endif %}
-                  {%endfor%}
-                </ul>
-              </div>
           {% endif %}
+          {% if (g.user.role_download() and g.user.is_anonymous) or g.user.is_authenticated %}
+            {% with formats = entry.data|select('canread')|list %}
+              {% if formats|length == 1 %}
+                {% set format = formats|first %}
+                <div class="btn-group" role="group">
+                  <a target="_blank" class="btn btn-primary" href="{{ url_for('read_book', book_id=entry.id, book_format=format.format|lower) }}">{{format.format}}
+                    <span class="glyphicon glyphicon-eye-open"></span> {{_('Read in browser')}}
+                  </a>
+                </div>
+              {% elif formats|length > 1 %}
+                <div class="btn-group" role="group">
+                  <button id="read-in-browser" type="button" class="btn btn-primary dropdown-toggle" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">
+                    <span class="glyphicon glyphicon-eye-open"></span> {{_('Read in browser')}}
+                    <span class="caret"></span>
+                  </button>
+                  <ul class="dropdown-menu" aria-labelledby="read-in-browser">
+                    {% for format in formats %}
+                      {%if format.format|lower|canread %}
+                        <li><a target="_blank" href="{{ url_for('read_book', book_id=entry.id, book_format=format.format|lower) }}">{{format.format}}</a></li>
+                      {% endif %}
+                    {%endfor%}
+                  </ul>
+                </div>
+              {%else%}
+                {{formats|length}}
+              {%endif%}
+            {%endwith%}
+          {%endif%}
         </div>
       </div>
       <h2>{{entry.title}}</h2>
diff --git a/cps/web.py b/cps/web.py
index ba57e651..8d00ad53 100755
--- a/cps/web.py
+++ b/cps/web.py
@@ -116,6 +116,7 @@ gdrive_watch_callback_token = 'target=calibreweb-watch_files'
 global_task = None
 
 ALLOWED_EXTENSIONS = set(['txt', 'pdf', 'epub', 'mobi', 'azw', 'azw3', 'cbr', 'cbz', 'cbt', 'djvu', 'prc', 'doc', 'docx', 'fb2'])
+READER_EXTENSIONS = set(['txt', 'pdf', 'epub', 'zip', 'cbz', 'tar', 'cbt'] + (['rar','cbr'] if rar_support else []))
 
 def md5(fname):
     hash_md5 = hashlib.md5()
@@ -473,6 +474,14 @@ def yesno(value, yes, no):
     return yes if value else no
 
 
+@app.template_filter('canread')
+@app.template_test('canread')
+def canread(ext):
+    if type(ext) == db.Data:
+        ext = ext.format
+    return ext.lower() in READER_EXTENSIONS
+
+
 def admin_required(f):
     """
     Checks if current_user.role == 1
@@ -937,10 +946,8 @@ def get_comic_book(book_id, book_format, page):
                         rarfile.UNRAR_TOOL = config.config_rarfile_location
                         try:
                             rf = rarfile.RarFile(cbr_file)
-                            rarNames = sort(rf.namelist())
-                            b64 = codecs.encode(rf.read(rarNames[page]), 'base64').decode()
-                            extractedfile="data:image/png;base64," + b64
-                            fileData={"name": rarNames[page],"page":page, "last":rarNames.__len__()-1, "content": extractedfile}
+                            names = sort(rf.namelist())
+                            extract = lambda page: rf.read(names[page])
                         except:
                             # rarfile not valid
                             app.logger.error('Unrar binary not found, or unable to decompress file ' + cbr_file)
@@ -949,25 +956,27 @@ def get_comic_book(book_id, book_format, page):
                         app.logger.info('Unrar is not supported please install python rarfile extension')
                         # no support means return nothing
                         return "", 204
-                if book_format in ("cbz", "zip"):
+                elif book_format in ("cbz", "zip"):
                     zf = zipfile.ZipFile(cbr_file)
-                    zipNames=sort(zf.namelist())
-                    if sys.version_info.major >= 3:
-                        b64 = codecs.encode(zf.read(zipNames[page]), 'base64').decode()
-                    else:
-                        b64 = zf.read(zipNames[page]).encode('base64')
-                    extractedfile="data:image/png;base64," + b64
-                    fileData={"name": zipNames[page],"page":page, "last":zipNames.__len__()-1, "content": extractedfile}
-
-                if book_format in ("cbt", "tar"):
+                    names=sort(zf.namelist())
+                    extract = lambda page: zf.read(names[page])
+                elif book_format in ("cbt", "tar"):
                     tf = tarfile.TarFile(cbr_file)
-                    tarNames=sort(tf.getnames())
-                    if sys.version_info.major >= 3:
-                        b64 = codecs.encode(tf.extractfile(tarNames[page]).read(), 'base64').decode()
-                    else:
-                        b64 = (tf.extractfile(tarNames[page]).read()).encode('base64')
-                    extractedfile="data:image/png;base64," + bs
-                    fileData={"name": tarNames[page],"page":page, "last":tarNames.__len__()-1, "content": extractedfile}
+                    names=sort(tf.getnames())
+                    extract = lambda page: tf.extractfile(names[page]).read()
+                else:
+                    app.logger.error('unsupported comic format')
+                    return "", 204
+
+                if sys.version_info.major >= 3:
+                    b64 = codecs.encode(extract(page), 'base64').decode()
+                else:
+                    b64 = extract(page).encode('base64')
+                ext = names[page].rpartition('.')[-1]
+                if ext not in ('png', 'gif', 'jpg', 'jpeg'):
+                    ext = 'png'
+                extractedfile="data:image/" + ext + ";base64," + b64
+                fileData={"name": names[page], "page":page, "last":len(names)-1, "content": extractedfile}
                 return make_response(json.dumps(fileData))
         return "", 204
 
@@ -3264,4 +3273,3 @@ def start_gevent():
         app.logger.info('Unable to listen on \'\', trying on IPv4 only...')
         gevent_server = WSGIServer(('0.0.0.0', ub.config.config_port), app)
         gevent_server.serve_forever()
-