[developers] Trying to operate on background stream : [incr tsdb()] affected as well

[Ann Copestake]

for the LKB use at least, it might be worth trying `finish-output' instead of 
`force-output'.  The difference, in theory, is that finish-output waits till 
the output has happened before returning, while force-output is intended to 
give things a push but to return immediately.  Despite this, from the Allegro 
documentation, it looks as though they are implemented the same way, but then 
it's interesting that you haven't seen any problems with the things that call 
finish-output.  I'm afraid I don't have time to look at this right now, though.

Ann

crysmann at dfki.de said:
> crysmann at dfki.de said:
>    >> I've narrowed down the problem now: if I comment out (force-output t)
> in >> read-tdl-type-files-aux and check-generator-environment the error >>
> disappears.  >>      >
> good!  the reason those `force-output's are there is just to give the user 
> feedback about what's going on while it's going on (as opposed to some time 
> later).  Commenting them out is not going to cause any big problems.
> Although 
> of course if Andreas can find an explanation/fix, that would be better! >
>    It seems that [incr tsdb()] is also affected by Andreas's patch.
> (force-output *tsdb-io*) in create-cache and flush-cache acts up now.

> Behaviour is exactly analogous to LKB, i.e., with Emacs it is bad,  without
> (=tty), it is fine. Uncommenting solves the issue. To test: Invoke
> Trees|Annotate from the menu...