Actually, when Mozilla 1.0 officially ships, I will send you a new version of the widget DLL that logs all focus calls.
Then you can send the log back to me and we will see what is going on. Something should happen within the next couple days. In addition, I'll be more than happy to send you a version of the DLL without the new focus fixes so you can continue to use Mozilla. Mike Derek J Decker wrote: > In message <[EMAIL PROTECTED]> - Michael Kaply <[EMAIL PROTECTED]> > writes: > :> > :>I think that would be overkill. It will be part of the installer if it ever gets > :>done. > :> > :>Anyone who is running OS/2 with FP1 deserves what they get. > :> > > Well, it's FixPack 1 plus individual corrections as required. > > Which is why I generally haven't had problems with previous releases > of Mozilla, or Netscape/2 4.61, or Java 1.1.8, or Java 1.3.0, or Visual Age > For Java, or many other things that claim to require Fixpack X. This is the > first and only time I've had a problem of this magnitude - and it's the first > and only case of text input areas not accepting focus in any application. > > I prefer to install individual fixes - it's a much more controllable > evolution, > and easier to back out of if it causes problems. I've been burned by FixPacks > before, and at this point, any monolithic install that causes system changes > scares the hell out of me. > > Still, if you were interested in fixing the focus problem that others have > seen as well, an analysis of why this box is different from boxes that do > not exhibit the aberrant behaviour could provide a serious debugger with > enough clues to identify and fix the problem for everyone, as I'm the only > one who has responded who has a completely repeatable scenario. > > That's why I responded in the first place - to see if I could provide > information that could assist you in locating, reproducing, and eventually > fixing an annoying UI problem that affects many end users - and will likely > affect many more after a full release. Your posts on the subject seemed > increasingly anxious to find a repeatable scenario. > > Now, if there's an individual correction that could be applied here as an > experiment to see if the problem gets fixed or goes away ("Here, try this > DLL, and tell me what happens"), I'd be happy to cooperate. If you'd like to > know dates/versions/other info about specific DLLs here to see if you can > set up a box there that replicates the problem reliably, I'll be happy to > answer. These are things that could generate clues as to the nature of the > problem, leading to a fix. A lot of debugging comes down to asking "What's > different between these two scenarios?". > > The other way to fix a bug these days is to tell all the people who are > experiencing it to Go Away And Leave You Alone. I can do that too, but that > won't help anyone else. > > It's your call, > > -Derek > > -- > +--------------------------------------------------------------------------+ > | Derek J Decker [EMAIL PROTECTED] Decker Automation | > | | > | http://DeckerAutomation.myiglou.com | > +--------------------------------------------------------------------------+
