From b70772717500394a44f494e8ee612bbba2297054 Mon Sep 17 00:00:00 2001
From: zhouji <calf.zhou@gmail.com>
Date: Mon, 14 Oct 2013 17:18:57 +0800
Subject: [PATCH 1/3] a simple solution for #1117

---
 pelican/contents.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pelican/contents.py b/pelican/contents.py
index dbc33716..62406904 100644
--- a/pelican/contents.py
+++ b/pelican/contents.py
@@ -190,7 +190,7 @@ class Content(object):
         instrasite_link_regex = self.settings['INTRASITE_LINK_REGEX']
         regex = r"""
             (?P<markup><\s*[^\>]*  # match tag with src and href attr
-                (?:href|src)\s*=)
+                (?:href|src|poster)\s*=)
 
             (?P<quote>["\'])      # require value to be quoted
             (?P<path>{0}(?P<value>.*?))  # the url value

From d85ef66698f441b82b7d7f21c88f7ef2a396fee1 Mon Sep 17 00:00:00 2001
From: zhouji <calf.zhou@gmail.com>
Date: Wed, 16 Oct 2013 16:43:34 +0800
Subject: [PATCH 2/3] Fix #1117 and add all possible url-value HTML attributes.

---
 pelican/contents.py | 60 ++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 56 insertions(+), 4 deletions(-)
 mode change 100644 => 100755 pelican/contents.py

diff --git a/pelican/contents.py b/pelican/contents.py
old mode 100644
new mode 100755
index 62406904..f89b08ff
--- a/pelican/contents.py
+++ b/pelican/contents.py
@@ -125,6 +125,52 @@ class Content(object):
         if 'summary' in metadata:
             self._summary = metadata['summary']
 
+        # prepare the list of HTML tag attributes which have a URL value.
+        # refer: http://stackoverflow.com/questions/2725156/complete-list-of-html-tag-attributes-which-have-a-url-value
+        self._url_attributes = {  # each item in this set is a tuple composed by tag_name, attr_name
+            # HTML4 tags
+            ('a', 'href'),
+            ('applet', 'codebase'),
+            ('area', 'href'),
+            ('base', 'href'),
+            ('blockquote', 'cite'),
+            ('body', 'background'),
+            ('del', 'cite'),
+            ('form', 'action'),
+            ('frame', 'longdesc'),
+            ('frame', 'src'),
+            ('head', 'profile'),
+            ('iframe', 'longdesc'),
+            ('iframe', 'src'),
+            ('img', 'longdesc'),
+            ('img', 'src'),
+            ('img', 'usemap'),
+            ('input', 'src'),
+            ('input', 'usemap'),
+            ('ins', 'cite'),
+            ('link', 'href'),
+            ('object', 'classid'),
+            ('object', 'codebase'),
+            ('object', 'data'),
+            ('object', 'usemap'),
+            ('q', 'cite'),
+            ('script', 'src'),
+
+            # HTML5 tags
+            ('audio', 'src'),
+            ('button', 'formaction'),
+            ('command', 'icon'),
+            ('embed', 'src'),
+            ('html', 'manifest'),
+            ('input', 'formaction'),
+            ('source', 'src'),
+            ('video', 'poster'),
+            ('video', 'src'),
+        }
+        """:type: set of (tuple of (string, string)"""
+        attribute_names = set(pair[1] for pair in self._url_attributes)
+        self._url_attr_pattern = '|'.join(attribute_names)
+
         signals.content_object_init.send(self)
 
     def __str__(self):
@@ -189,20 +235,26 @@ class Content(object):
 
         instrasite_link_regex = self.settings['INTRASITE_LINK_REGEX']
         regex = r"""
-            (?P<markup><\s*[^\>]*  # match tag with src and href attr
-                (?:href|src|poster)\s*=)
-
+            (?P<markup><\s*(?P<tag>[^\s\>]+)[^\>]*  # match tag with all url-value attributes
+            (?P<attr>{1})\s*=)
             (?P<quote>["\'])      # require value to be quoted
             (?P<path>{0}(?P<value>.*?))  # the url value
-            \2""".format(instrasite_link_regex)
+            \4""".format(instrasite_link_regex, self._url_attr_pattern)
         hrefs = re.compile(regex, re.X)
 
         def replacer(m):
+            print(m.group(0))
             what = m.group('what')
             value = urlparse(m.group('value'))
             path = value.path
             origin = m.group('path')
 
+            # verify HTML tag and attribute pair to avoid miss-replacing
+            tag = m.group('tag')
+            attr = m.group('attr')
+            if attr != 'href' and attr != 'src' and (tag, attr) not in self._url_attributes:
+                return m.group(0)
+
             # XXX Put this in a different location.
             if what == 'filename':
                 if path.startswith('/'):

From fa150ad51daedefcf7d6c3a75a5c2c156e002695 Mon Sep 17 00:00:00 2001
From: zhouji <calf.zhou@gmail.com>
Date: Wed, 16 Oct 2013 16:45:31 +0800
Subject: [PATCH 3/3] Restore file mode of content.py, which is miss-updated in
 last commit.

---
 pelican/contents.py | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 mode change 100755 => 100644 pelican/contents.py

diff --git a/pelican/contents.py b/pelican/contents.py
old mode 100755
new mode 100644