[Mailman-Users] odd archive behavior

[Mark Sapiro]

Mark Sapiro wrote:

>Con Wieland wrote:
>
>>I don't understand why the body of this message was removed in the  
>>archives. (I have removed the picture at the bottom to meet the lists  
>>size requirement)
>>
>>http://maillists.uci.edu/mailman/public/visgrads/2008-March/000174.html
>
>
>The short answer is Mailman 2.1.4
>
>>From the above archive entry
>
>Skipped content of type multipart/alternative-------------- next part
>--------------
>A non-text attachment was scrubbed...
>
>
>So the 'body' was multipart/alternative. There have been many changes
>to the scrubber since 2.1.4, and I think that currently, assuming the
>multipart/alternative part had text/plain and text/html subparts, the
>text/plain part would be the body in the archive, and the text/html
>part would have been stored, but the 2.1.4 scrubber didn't do that
>well with 'not so simple' message structures.


To be more specific, I suspect the original message had a structure
similar to the following (except the jpeg image had content-type
application/octet-stream)

multipart/related
    multipart/alternative
        text/plain
        text/html
    image/jpeg

I tested this with the current scrubber and with the 2.1.4 scrubber.
The 2.1.4 scrubber produces the result you see in your archive and the
current scrubber produces the expected result with the text/plain body
in the message and the text/html and image/jpeg parts scrubbed.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan