Re: [Evolution-hackers] Bug #232

[Peter Colijn <pcolijn nit ca>]

On Wed, 2003-11-19 at 09:41, Randall Hand wrote:
> Well, it's doesn't handle *all* MAPI properties, but it does support
> for external programs to read them and do with them as they wish,
> using the same GUID+id syntax you mentioned.

Ok, cool.

> ytnef is mostly designed to take a tnef stream (winmail.dat), and
> strip out allthe attachments.  This includes all calendar entries,
> contact cards, and tasks, which are saved as vCal/vCard entries
> (plaintext).

So does ytnef get vCals/vCards as attachments out of the TNEF, or does
it build vCals/vCards from all the MAPI properties (eg. using "Common
Start" and "Common End" as DTSTART and DTEND)?

> But using the library, you can save these entries as anything you
> want, it's just a long list of queries for certain ID's.  The 2 main
> things that it does not support right now is Compressed RTF and
> Recurring calendar entries.  But I use it for all my tnef streams
> (procmail script to automatically process email) and haven't found a
> stream that broke it yet :)

WvTnef does support recurring calendar entries (and exceptions). MAPI ID
0x8216 is LOTS of fun, let me tell you =) It also has the fun property
that if don't repack that binary field *perfectly*, Outlook will crash
when you try to import the object into the calendar.

In any case, maybe if WvTnef isn't suitable to go in Evolution as it is,
when it's released some of the stuff from WvTnef could maybe be merged
into ytnef... Although I know very little about licensing issues. WvTnef
will be LGPL and I guess ytnef is GPL?

Have fun,

Peter