FileTransfer would previously not change the status, even though an
exception has been set, leading users to believe that the transfer is
still ongoing, when it is not.
Some methods of OutgoingFileTransfer take already the
NegotiationProgress callback as argument, some do not. The API is not
ideal, and adding a setter for the callback at least makes it possible
to always have the callback.
The FillableForm API is the most convenient way of filling out a
form. Currently only forms with a FORM_TYPE can be filled, due to a
restriction in FillableForm. This makes filling out untyped forms very
cumbersome.
This commit removes the required FORM_TYPE when using FilledForm (and
therefor FillableForm). Some subclasses of FilledForm already check
that the correct FORM_TYPE is set using FilledForm#ensureFormType().
This also resulted in a refactoring of the Providers and parsing
Exceptions. NumberFormatException and ParseException can now be thrown
directly, the wrapping in a SmackParsingException is down at a higher
layer, i.e. in AbstractProvider.
This fixes an inconsistent representation of a 'no error' representation, that
causes a file transfer in which not enough bytes were transferred to _not_ be
marked as having a problem.
This prevents a race condition of enter() with the presence listern by
waiting until all presences have been processed.
Reported-by: Guus der Kinderen <guus@goodbytes.nl>
To increase readability, and with that maintainability, we rename
'presence' to 'reflectedSelfPresence' in MultiUserChat.enter(), to
make it clear what kind of presence this variable holds.
Also mark the variable as final.
We previously only looked in the registry for 'submit' type forms. But
also 'result' type forms may be send without field type
annotations. Same is true, but less likely, for 'form' type forms.
InBandBytestreamManager followed an unusual pattern: Within the
connectionTermianted() callback, it would remove itself from the
'managers' map. This allowed for multiple instances of an
InBandBytestreamManager to exist for the same connection, causing all
kinds of issues.
This fixes the issue by changing InBandBytestreamManager to use the
Smack-idiomatic pattern used by managers.
We also do no longer reset the listeners if the connection is
termianted, as listeners (and handlers) typically persist until they
are explicitly removed by the user.
As positive side-effect, the number of indeterministic unit-tests,
caused by using Thread.sleep(), is reduced. The executor service in
InitiationListener was also removed, because the IQ handler is already
called asynchronously to the connections main loop.
Thanks to Anno van Vliet for reporting this.
