> We have test cases for this issue:
>
>
> http://saros-build.imp.fu-berlin.de/gerrit/#/c/210/3/de.fu_berlin.inf.dpp/test/stf/de/fu_berlin/inf/dpp/stf/test/partialsharing/ShareFilesToProjectsWithDifferentEncodingTest.java
>
> Fix / Hack is already present in the code base but not activated.
>
> See:
>
> http://saros-build.imp.fu-berlin.de/gerrit/#/c/200/8/de.fu_berlin.inf.dpp/src/de/fu_berlin/inf/dpp/ui/util/CollaborationUtils.java
>
> After some testing I think it is save to execute the block.
>
> Affected editors are closed during project synchronization (well you can at
> least reopen them manually ... but I do not think anyone will do this)
>

Hi Stefan!

Well, that's exactly what we decided not to do :)
And the comment above the block describes it well I think, it is unknown
what sharing .settings
in a partially shared project might cause (at least to us)

My favorite idea so far is to encode the encoding (no pun intended) in the
hashCode of the
checkSum, have it reverse engineered (or just read) on the client side, and
have it handled
properly (maybe even exactly like in the comment).

If the peer has an older version of Saros, and this inconsistency occurs,
it might be fair to
display an InfoMessage to the user, telling him how to resolve (changing
the encoding)
or  updating his version. If it's a normal inconsistency, then just not
encode checkSums
for outdated clients.

That way the fix is at least located in the inconsistency/recovery code and
is less
side-effect-ish.


Cheers,
Arsenij
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Dpp-devel mailing list
Dpp-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dpp-devel

Reply via email to