>> Javier Delgadillo wrote: >> Another way to do it is to not register a UI callback in psm-glue. If >> you look at the function that handles the error cases for UI in PSM >> (http://lxr.mozilla.org/mozilla/source/security/psm/server/sslconn.c#1383), >> you'll notice the first check is to see if it thinks the calling app >> does UI. If it doesn't think the app does UI, then no UI is thrown and >> the connection fails silently. >> >> So, if you look at the function at >> (http://lxr.mozilla.org/mozilla/source/extensions/psm-glue/src/nsPSMUICallba cks.cpp#222), >> and you make it not call CMT_SetUIHandlerCallback, then my hunch is the >> connection will fail silently just like you want. >> >> I'm trying to spend all of my time on PSM 2.0, so if someone else come >> up with a patch for this, that would be great. >> >Doug Turner wrote: >Yes! I remember this now. Lets add a pref around this so that it can be disabled in non UI >clients or clients that do not want PSM's HTML dialogs. Javi thanks for reminding us about >this. > >I think that we can code the pref thing without interrupting you. However, we need an >approval since, I guess, you are the module owner now :-) >Rick, can you try this suggestion out? I commented out the call to CMT_SetUIHandlerCallback and allowed the function InitPSMUICallbacks to return PR_Success. This avoided me hanging, waiting for further OnStartRequestEvents, and the destructor for my consumer was called. However, the desired behavior would be for the OnStopRequest method for my consumer to be called. This behavior is exhibited when the function InitPSMUICallbacks returns PR_Failure.
