Merge branch 'master' into simple_httpclient

bdarnell · Oct 8, 2010 · a80333a · a80333a
2 parents 48d698e + 160c961
commit a80333a
Show file tree

Hide file tree

Showing 5 changed files with 100 additions and 21 deletions.
diff --git a/demos/chat/templates/message.html b/demos/chat/templates/message.html
@@ -1 +1,2 @@
-<div class="message" id="m{{ message["id"] }}"><b>{{ escape(message["from"]) }}: </b>{{ escape(message["body"]) }}</div>
+{% import tornado.escape %}
+<div class="message" id="m{{ message["id"] }}"><b>{{ escape(message["from"]) }}: </b>{{ tornado.escape.linkify(message["body"]) }}</div>
diff --git a/tornado/escape.py b/tornado/escape.py
@@ -49,7 +49,7 @@ def _json_decode(s):
 
 def xhtml_escape(value):
     """Escapes a string so it is valid within XML or XHTML."""
-    return utf8(xml.sax.saxutils.escape(value, {'"': "&quot;"}))
+    return xml.sax.saxutils.escape(value, {'"': "&quot;"})
 
 
 def xhtml_unescape(value):
@@ -95,6 +95,90 @@ def utf8(value):
     return value
 
 
+# Regex from http://daringfireball.net/2010/07/improved_regex_for_matching_urls
+# Modified to capture protocol and to avoid HTML character entities other than &amp;
+_URL_RE = re.compile(ur"""(?i)\b((?:([a-z][\w-]+):(?:(/{1,3})|[a-z0-9%])|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>&]+|&amp;|\(([^\s()<>&]+|(\([^\s()<>&]+\)))*\))+(?:\(([^\s()<>&]+|(\([^\s()<>&]+\)))*\)|[^\s`!()\[\]{};:'".,<>?\xab\xbb\u201c\u201d\u2018\u2019&]))""")
+
+
+def linkify(text, shorten=False, extra_params="",
+            require_protocol=False, permitted_protocols=["http", "https"]):
+    """Converts plain text into HTML with links.
+
+    For example: linkify("Hello http://tornadoweb.org!") would return
+    Hello <a href="http://tornadoweb.org">http://tornadoweb.org</a>!
+
+    Parameters:
+    shorten: Long urls will be shortened for display.
+    extra_params: Extra text to include in the link tag,
+        e.g. linkify(text, extra_params='rel="nofollow" class="external"')
+    require_protocol: Only linkify urls which include a protocol. If this is
+        False, urls such as www.facebook.com will also be linkified.
+    permitted_protocols: List (or set) of protocols which should be linkified,
+        e.g. linkify(text, permitted_protocols=["http", "ftp", "mailto"]).
+        It is very unsafe to include protocols such as "javascript".
+    """
+    if extra_params:
+        extra_params = " " + extra_params.strip()
+
+    def make_link(m):
+        url = m.group(1)
+        proto = m.group(2)
+        if require_protocol and not proto:
+            return url  # not protocol, no linkify
+
+        if proto and proto not in permitted_protocols:
+            return url  # bad protocol, no linkify
+
+        href = m.group(1)
+        if not proto:
+            href = "http://" + href   # no proto specified, use http
+
+        params = extra_params
+
+        # clip long urls. max_len is just an approximation
+        max_len = 30
+        if shorten and len(url) > max_len:
+            before_clip = url
+            if proto:
+                proto_len = len(proto) + 1 + len(m.group(3) or "")  # +1 for :
+            else:
+                proto_len = 0
+
+            parts = url[proto_len:].split("/")
+            if len(parts) > 1:
+                # Grab the whole host part plus the first bit of the path
+                # The path is usually not that interesting once shortened
+                # (no more slug, etc), so it really just provides a little
+                # extra indication of shortening.
+                url = url[:proto_len] + parts[0] + "/" + \
+                        parts[1][:8].split('?')[0].split('.')[0]
+
+            if len(url) > max_len * 1.5:  # still too long
+                url = url[:max_len]
+
+            if url != before_clip:
+                amp = url.rfind('&')
+                # avoid splitting html char entities
+                if amp > max_len - 5:
+                    url = url[:amp]
+                url += "..."
+
+                if len(url) >= len(before_clip):
+                    url = before_clip
+                else:
+                    # full url is visible on mouse-over (for those who don't
+                    # have a status bar, such as Safari by default)
+                    params += ' title="%s"' % href
+
+        return u'<a href="%s"%s>%s</a>' % (href, params, url)
+
+    # First HTML-escape so that our strings are all safe.
+    # The regex is modified to avoid character entites other than &amp; so
+    # that we won't pick up &quot;, etc.
+    text = _unicode(xhtml_escape(text))
+    return _URL_RE.sub(make_link, text)
+
+
 def _unicode(value):
     if isinstance(value, str):
         return value.decode("utf-8")

diff --git a/tornado/iostream.py b/tornado/iostream.py
@@ -21,6 +21,7 @@
 import socket
 
 from tornado import ioloop
+from tornado import stack_context
 
 try:
     import ssl # Python 2.6+
@@ -79,14 +80,15 @@ def __init__(self, socket, io_loop=None, max_buffer_size=104857600,
         self._write_callback = None
         self._close_callback = None
         self._state = self.io_loop.ERROR
-        self.io_loop.add_handler(
+        with stack_context.NullContext():
-            self.socket.fileno(), self._handle_events, self._state)
+            self.io_loop.add_handler(
+                self.socket.fileno(), self._handle_events, self._state)
 
     def read_until(self, delimiter, callback):
         """Call callback when we read the given delimiter."""
         assert not self._read_callback, "Already reading"
         self._read_delimiter = delimiter
-        self._read_callback = callback
+        self._read_callback = stack_context.wrap(callback)
         while True:
             # See if we've already got the data from a previous read
             if self._read_from_buffer():
@@ -103,7 +105,7 @@ def read_bytes(self, num_bytes, callback):
             callback("")
             return
         self._read_bytes = num_bytes
-        self._read_callback = callback
+        self._read_callback = stack_context.wrap(callback)
         while True:
             if self._read_from_buffer():
                 return
@@ -123,11 +125,11 @@ def write(self, data, callback=None):
         self._check_closed()
         self._write_buffer += data
         self._add_io_state(self.io_loop.WRITE)
-        self._write_callback = callback
+        self._write_callback = stack_context.wrap(callback)
 
     def set_close_callback(self, callback):
         """Call the given callback when the stream is closed."""
-        self._close_callback = callback
+        self._close_callback = stack_context.wrap(callback)
 
     def close(self):
         """Close this stream."""
@@ -177,6 +179,8 @@ def _run_callback(self, callback, *args, **kwargs):
         try:
             callback(*args, **kwargs)
         except:
+            logging.error("Uncaught exception, closing connection.",
+                          exc_info=True)
             # Close the socket on an uncaught exception from a user callback
             # (It would eventually get closed when the socket object is
             # gc'd, but we don't want to rely on gc happening before we

diff --git a/tornado/stack_context.py b/tornado/stack_context.py
@@ -101,6 +101,8 @@ def wrap(fn):
     different execution context (either in a different thread or
     asynchronously in the same thread).
     '''
+    if fn is None:
+      return None
     # functools.wraps doesn't appear to work on functools.partial objects
     #@functools.wraps(fn)
     def wrapped(callback, contexts, *args, **kwargs):

diff --git a/website/templates/index.html b/website/templates/index.html
@@ -45,18 +45,6 @@ <h2>Hello, world</h2>
   <p>See the <a href="/documentation">Tornado documentation</a> for a detailed walkthrough of the framework.</p>
 
   <h2>Discussion and support</h2>
-  <p>You can discuss Tornado and report bugs on <a href="http://groups.google.com/group/python-tornado">the Tornado developer mailing list</a>.
+  <p>You can discuss Tornado and report bugs on <a href="http://groups.google.com/group/python-tornado">the Tornado developer mailing list</a>.  Links to additional resources can be found on the <a href="http://github.com/facebook/tornado/wiki/Links">Tornado wiki</a>.
-
-  <h2>Links and resources</h2>
-  <ul>
-    <li><a href="http://tornado.poweredsites.org">tornado.poweredsites.org</a> - listing of sites using Tornado</li>
-    <li><a href="http://github.com/szabotshark/t3">T3</a> - Tornado ported to Python 3</li>
-    <li><a href="http://github.com/fiorix/cyclone">Cyclone</a> - Tornado ported to the Twisted event loop</li>
-    <li><a href="http://github.com/bdarnell/tornado_tracing">Tornado Tracing</a> - Performance tracing library for Tornado</li>
-  </ul>
-
-  <h2>Updates</h2>
-  <p>Follow us on <a href="http://www.facebook.com/pages/Tornado-Web-Server/144144048921">Facebook</a>, <a href="http://twitter.com/tornadoweb">Twitter</a>, or <a href="http://friendfeed.com/tornado-web">FriendFeed</a> to get updates and announcements:</p>
-  <div style="margin-top:1em"><a href="http://www.facebook.com/pages/Tornado-Web-Server/144144048921" style="margin-right:10px"><img src="/static/facebook.png" style="width:64px;height:64px" alt="Facebook"/></a><a href="http://twitter.com/tornadoweb" style="margin-right:10px"><img src="/static/twitter.png" style="width:64px;height:64px" alt="Twitter"/></a><a href="http://friendfeed.com/tornado-web" style="margin-right:10px"><img src="/static/friendfeed.png" style="width:64px;height:64px" alt="Facebook"/></a></div>
 
 {% end %}