[Evolution] EDS 3.16, CalDAV and FastMail (Allows header disagreement)

[Chris Schooley <chris schooley name>]

Hello. Long-time user, first time poster,

I've been follwing the threads on "read-only CalDAV issues" in 3.16
becaue I'm running 3.16.5-1 and I use FastMail's Cal/CardDAV services. I
contacted them about the issue yesterday and referenced the following
threads from this mailing list:

https://mail.gnome.org/archives/evolution-list/2015-May/msg00104.html
https://mail.gnome.org/archives/evolution-list/2015-May/msg00096.html

To summarize the above links: evolution-data-server's caldav backend is
looking for a 'PUT' and 'DELETE' verb in the Allows header to determine
if it can write to a CalDAV collection. FastMail do not include the
'PUT'. Feeling that this was a provider configuration issue I contacted
FastMail and asked them to look into the issue.

Here was their response (their words not mine):

> We had one other Evolution user report the same issue. We checked with our engineers and it appears what 
> Evolution is doing is >wrong. This is what they mentioned:
>
> ---
> The Evolution developers are wrong - you can't PUT to the collection, only to resources within the 
> collection.
>
> It can check if it can PUT any particular resource (and it will get "Allow: PUT" for any resource) but we 
> don't support PUT on >collections because it doens't make sense to PUT to a collection.  Evolution should be 
> using the ACL extension or just trying a >write if it wants to know if it's allowed to write.
> ---

As a troubleshooting step/workaround I removed the check for PUT from
evolution-data-server/src/evolution-data-server-3.16.5/calendar/backends/caldav/e-cal-backend-caldav.c
at line 1248 and I was able to execute all CRUD operations on
Cal/CardDAV data witout issue. I am comfortable (not thrilled) managing
this as part of my system's "ports" process, but hopeful there is a
better solution for users that are not in a position to do this.

With FastMail saing they are right and the recent comments on this list
suggesting that they are not I'm kind of at a loss for a way ahead. Sure
I could submit a patch/bug/feature against evolution-data-server;s
CalDAV backend to optionally disable the PUT check or implement one of
FastMail's other suggestions, but that won't get much traction if the
general consensus is that this amounts to maintaining some kind of
"FastMail Quirks Mode". I can be a squeaky wheel of a FastMail customer,
but it sounds like the Great and Powerful Oz has spoken on that front.

Any suggestions on how to proceed?

Thanks,
-- 
Chris Schooley