Merge branch 'master' into bbmaster

util/html.py

@@ -36,6 +36,7 @@ validStyleParts = 'border padding'.split()
 
 escCommPattern = re.compile(r'<\!--\[if .*?\!\[endif\]-->', re.DOTALL)
 
+sentencePattern = re.compile(r'[:.\?\!]')
 
 def sanitize(value, validTags=validTags, validAttrs=validAttrs,
                     validStyles=validStyles, stripEscapedComments=True):
@@ -104,3 +105,14 @@ def stripAll(value):
     text = u''.join(data).replace(u'\n', u'').replace(u' ', u' ')
     return text
 
+
+def extractFirstPart(value):
+    soup = BeautifulSoup(value)
+    for tag in soup.findAll(True):
+        if tag.name in ('p',):
+            part = tag.renderContents()
+            break
+    else:
+        text = stripAll(value)
+        part = sentencePattern.split(text)[0]
+    return ('<p>%s</p>' % part).decode('utf8')

util/html.txt

@@ -10,6 +10,9 @@ Tweaking HTML text
   ... </p>
   ... </html>"""
 
+Sanitize HTML
+-------------
+
   >>> sanitize(input, validAttrs=['style'])
   u'\n<p style="font-weight: bold">\n<a><b>Text</b>, and more</a>\n</p>\n'
 
@@ -36,3 +39,14 @@ It is also possible to strip all HTML tags from the input string.
   >>> from cybertools.util.html import stripAll
   >>> stripAll(input)
   u'Text, and more'
+
+Extract first part of an HTML text
+----------------------------------
+
+  >>> from cybertools.util.html import extractFirstPart
+
+  >>> extractFirstPart(input)
+  u'<p>\n<a href="blubb"><b>Text</b>, and more</a>\n</p>'
+
+  >>> extractFirstPart(input2)
+  u'<p>text</p>'