{"id":878,"date":"2009-08-18T22:42:53","date_gmt":"2009-08-18T22:42:53","guid":{"rendered":"http:\/\/dalelane.co.uk\/blog\/?p=878"},"modified":"2009-08-21T16:36:01","modified_gmt":"2009-08-21T16:36:01","slug":"stripping-out-mime-headers","status":"publish","type":"post","link":"https:\/\/dalelane.co.uk\/blog\/?p=878","title":{"rendered":"Stripping out MIME headers"},"content":{"rendered":"

A couple of years ago<\/a>, I wrote a small plugin for Outlook Mobile<\/a> (the version of Outlook that you get on Windows Mobile phones) to help me triage my emails.<\/p>\n

The idea was to make it quicker to process my email<\/a> from my phone, by adding a couple of context menu items to emails that let you create a new To Do item in Outlook Tasks, or a new diary entry in Outlook Calendar. <\/p>\n

So if someone sends you an email asking you to do something, with one tap you can create a new To Do list item, pre-filling it with information from the email. <\/p>\n

And if someone sends you an email about an event or meeting you need to go to, with one tap you can create a new Calendar item, prefilling it with information from the email. <\/p>\n


It worked okay, but there was one problem.<\/p>\n

The API for getting the body out of the email<\/a> gives you a stream to read from. <\/p>\n

And what you (can) read from the stream is the MIME<\/a> text of the email. <\/p>\n

Which meant that while I did get the body of the email in my new Task or Calendar items, I also got a ton of MIME header junk that I didn’t really want. <\/p>\n

A lot of it.
\n
I’ve just been ignoring it. But it is annoying to have to scroll so far down to get to the actual email body – particularly on phones with smaller screens. <\/p>\n

Tonight I decided to have a go at fixing this.<\/p>\n

I’ve written what is possibly the world’s dumbest MIME-parser \ud83d\ude42 <\/p>\n

It basically scans through the stream, assuming that if it sees a ‘:<\/code>‘ character before any whitespace or newline, then the current line is a MIME-header. <\/p>\n

And assumes that the first line where this is not true is the start of the email body. <\/p>\n

The slightly fiddly bit was handling where some MIME header values are continued on to the following line.
\n
A quick test on the emails in my inbox seems to suggest that it basically works. <\/p>\n

Doesn’t this look much more useful?<\/p>\n

However, it’s a fairly quick hack, so I’m guessing that it’ll probably falls over if it encounters some sequence of MIME header lines that I’ve not considered. <\/p>\n


If you’d like to give it a try, the updated version is here<\/a>.<\/p>\n

Update (19-Aug-2009):<\/strong> I updated the code the following evening to:<\/p>\n