"Danny Epstein" <[EMAIL PROTECTED]> wrote in message news:84666@palm-dev-forum... > > MP> [...] I could swear that the docs said that using ExgConnect is > MP> _required_ when sending multiple objects. I guess you're providing the > MP> additional clarification that the term "object" doesn't include > databases.
> Here's a quote from the docs I referenced: > DOC> The ExgConnect call is optional. Some exchange libraries, such as > DOC> the IR Library, support the sending of multiple objects but do not > DOC> support ExgConnect. [...] I remember this now. I think I was thrown by the (wrong) thought that ExgConnect couldn't know I was doing IR, since that was the first API call I used. The content of the exchange socket told it this, though. > MP> Can the docs be modified to > MP> make this clear? Without knowledge of OS internals, one would never > MP> think that the collection was being modified in this case. > I'm not sure how to document this. [...] Perhaps the existing warning in > DmGetNextDatabaseByTypeCreator could be extended to mention that > beaming or sending may create a temporary database. That would've got my attention. > [...] Another alternative would be to add > a note to ExgPut indicating that some exchange libraries may create a > database to buffer the object. This could be useful for other reasons as > well. [...] This would help as a backup. The question in my mind was "How come DmGetNextDatabaseByTypeCreator isn't working?", so a note in the documentation for it about ExgPut interfering with its correct operation would've answered my question right away. > MP> how does an application that has told the > MP> system that it is the launcher prevent itself from being stopped upon > MP> receiving a beamed application? > What do you mean by "told the system that it is the launcher"? And what do > you mean by "being stopped"? I think I can guess... These refer to setting the system preference for the app to be launched when tapping the "Applications" button, and getting an appStopEvent after an app has been beamed to the device because the Exchange Manager wants to display it in the built-in launcher. I've decided for the time being not to ignore the stop event in this case . With Palm hardware I think it can be done reliably by a process of elimination (not an app launch my own app did, nor a hard key, find, calc or the cradle button), but other devices may provide some other way to generate a stop event. > [more good info on registering for data deleted for space] Thanks! Mark Peters -- For information on using the Palm Developer Forums, or to unsubscribe, please see http://www.palmos.com/dev/support/forums/
