I just took a quick look at the java.awt.EventQueue class where the dispatch thread is managed, and I could see no obvious place you could hook in your own calling code, which you'd need to wrap it with a (binding ...) form.
If you really want to hack the system, look at the code in core.clj for binding. It looks like you could force a call into Var.pushThreadBinding() in the dispatch thread. However, doing that is risky and is definately subverting the system. Don't complain if it blows up in your face. I would suggest you either implment it as a global Atom, or perhaps go down the the Java level and use a ThreadLocal. On Wed, Feb 11, 2009 at 10:30 PM, Chas Emerick <cemer...@snowtide.com>wrote: > > def doesn't create a thread-local binding, it creates a var within a > particular namespace -- the binding form is what established the > thread-local binding. So, when used on the AWT/Swing thread, > > (binding [*my_environment* (new_environment)] > ...) > > will execute the body of the binding form with the *my_environment* > var bound to the value of (new_environment). > > - Chas > > On Feb 11, 2009, at 8:40 PM, CuppoJava wrote: > > > > > Hi, > > I just wrote a library that creates an un-initialized global variable, > > and requires the user to create his own thread-local binding in order > > to run it. > > > > eg. > > (def *my_environment*) > > ... > > > > and the user would use it like so: > > (binding [*my_environment* (new_environment)] > > (start_engine_thread)) > > > > And this works wonderfully for my custom engine. > > > > But now I wrote a engine using Swing. And I don't have control over > > when the Swing thread is started. So there's no way for me to create a > > thread-local binding to use my library. > > > > eg. I can't do this: > > (binding [*my_environment* (new_environment)] > > (start_swing_event_thread) > > (.setVisible (new_engine))) > > > > So is there anyway for me to push a thread-local binding, when the > > thread has already been started? > > > > Or is there a clean workaround for this problem? Instead of having to > > resort to passing *my_environment* around through function parameters. > > > > Thanks a lot > > -Patrick > > > > > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~----------~----~----~----~------~----~------~--~---
