Regular expression altered by GUI

[Panagiotis Issaris]

When entering any regular expression containing a "<" in the "List of non-member addresses whose postings should be automatically accepted." field, it is replaced by a "&lt;". I am not 100% sure this is a bug, as it might convert it back to "<" before passing it to Python's re module.

For example:
^.*<konijn.com$
becomes:
^.*&lt;konijn.com$

I tried escaping although I figured it wasn't needed:
^.*\<konijn.com$ ---> ^.*\&lt;konijn.com$(?<!...)
^.*\\<konijn.com$ ---> ^.*\\&lt;konijn.com$

The reason I was using a "<" in a regular expression was for using "(?<!...)" as shown on http://docs.python.org/library/re.html

[Mark Sapiro (msapiro)]

What Mailman version is this? What browser?

With recent Mailman, the "<" is sent back to the browser as "&lt;", but the browser should render that as "<".

If you enter "<", that is what is stored in the list attribute. It is only converted to "&lt;" in the HTML sent back to the browser. Now if the browser is broken and displays the "&lt;" as "&lt;" and not as "<", and you "Submit Your Changes" a second time with "&lt;" in the text box, that could store the wrong data.