User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.3b) Gecko/20030104 Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.3b) Gecko/20030104 The Mach-O build is not sending getFocusEvent and loseFocusEvent events to plugins. As points of reference, those events are sent in each of the following: The last CFM build I looked at (late-december) Chimera (a recent build, like a few days ago) Safari These events are critical to our plugin (unfortunately, there are plenty of other blockers that affect us, too) Reproducible: Always Steps to Reproduce: 1. Go to page with plugin 2. Click in plugin Actual Results: mouseDown/mouseUp events are sent to plugin correctly Expected Results: A getFocusEvent should be sent as well
This remains a show-stopper for us. We haven't supported any build since 1.2.x (that is, any Mach-O build) with our Macintosh plugins. There were several reasons for that, but now that the new npruntime scripting enhancements are in place, this bug moves toward the top of our list. Any thoughts of progress?
Still no progress, still a show-stopper for us. I'm requesting that this be considered for blocking1.8a6, in hopes that someone will at least look at this. Please?
Works properly in the 1.7.5 release build, but not in today's top-of-tree 1.8a6 build. Any chance the fix could be moved over, whatever it was?
Jonathan, can you identify the change that fixed this on the 1.7 branch? That would probably go a long way towards getting it fixed on the trunk.
Yeah, I know. I looked for a likely candidate by searching the codebase for "focus" and looking at the commit messages, but I couldn't find one. Sorry. I was hoping that someone who had debuggable builds of 1.7.5 and 1.8a on their machine could find the culprit pretty quickly. If that someone was familiar with the code, they could put a breakpoint at the spot where the browser sends the focus event to the plugin in 1.7.5, and look at the backtrace. It should be easy after that, I hope -- something along that specific backtrace isn't happening the same way in the 1.8a codebase...
Jonathan, another approach would be to test nightly builds from the branch and see which day included the fix. Getting a one or two day window would probably make the chage easily identifyable in bonsai. <http://archive.mozilla.org/pub/mozilla/nightly/>
Every build I can find that *wasn't* from the mainline seems to send the event properly (I tested back to 2003-09-16-15-1.5). Every mainline build I can find (and I can't find that many, actually) fails to send the focus events at all. The last mainline build that did send the focus events was prior to the CFM->MachO migration in 12/ 2002 or so. I don't know enough about the history of the source trees to know if that tells you anything.
Johnny, Simon, Pink, Brian, do any of you have any idea what's going on here? I can't imagine how our release branches have this working but the trunk doesn't.
too late for non-critical fixes. blocking-