IRC Log Viewer » #firebreath » 2011-10-26

IRC Nick Time (GMT-7) Message
FireBreathBot 01:10 JIRA issue issue created by kobymeir
agd 04:10 Hi everyone, does anyone know if it's possible to get separate up/down events for caps lock on cocoa?
i imagine it's by design but for every caps lock press you only get one event rather than the usual 2 events (keydown/keyup)
dougma 06:10 agd: that'd be a strange design.
is that safari? tried other browsers?
agd 06:10 it's all browsers
i guess it is a strange design - we're building a plugin for game devs and we want to provide the current state of all keys
dougma 06:10 you want a way to query the current state... tracking it will be too flakey
agd 06:10 but it seems like the npapi cocoa event spec follows the usual cocoa one - i.e. only one event per caps lock (or num lock) "toggle"
dougma 06:10 i see
agd 06:10 tacking it is flakey but we want to pass on the input events so devs can use them in their games
seems like it's not possible in this case though...
dougma 06:10 well you can pass along what you get...
agd 06:10 yeah
gargle 07:10 hi all
just a question, someone use firebreath plugin in prism?
FireBreathBot 08:10 JIRA issue issue commented by jshanab "Looking in the CMakeList.txt file the 2 lines that process thsi file were moved from outside the ..."
FireBreathBot 08:10 JIRA issue issue created by tippex
taxilian 08:10 why do people pollute a perfectly good wiki by trying to put ads on it? it is infuriating
people who do that (or write bots to do that) should be drawn and quartered
linearray 08:10 really? i didn't get an edit message recently
taxilian 08:10 strange
ID id
I did
and had to go delete the pages
3rd or 4th time in the last few weeks
will need to turn on captcha… hopefully it's a bot
linearray 08:10 last edit message I got was oct 21
taxilian 08:10 strange; you should have at least gotten some about your own edit on the 23rd
don't restore it please =]
linearray 08:10 I never get messages about my own edits
do you?
taxilian 08:10 hmm. thought I did, but maybe not. I do at least get a nightly "here is what changed today" email
that includes my own stuff
linearray 08:10 hm, interesting
taxilian 08:10 well, I've turned on captcha for now
if it's bugging anyone let me know and I'll add you to a group that is excluded from them
linearray 08:10 but I know, spam is super annoying... I turned on two different captchas for my site now
recaptcha alone is not working unfortunately
FireBreathBot 09:10 JIRA issue issue resolved by richard "Please try things before you report them as not working; this works fine in XP, and commenting ou..."
FireBreathBot 09:10 JIRA issue issue reopened by richard "From what you have told me on the wiki, it sounds like this issue is specific to vs2005; I'll try..."
JSHANAB 10:10 Good morning all.
taxilian 10:10 g'morning
JSHANAB 10:10 taxilian. Did you see my comment on FIREBREATH-137. I didn't dare change it without asking. locally I copied those two lines so the were also outside the if to get our builds going. Those lines were moved inside 5 days ago for the mac fix ???
FireBreathBot 10:10 FIREBREATH-137: Summary: defines.hxx in <Firebreath>\src\libs\log4cplus\include\log4cplus\config is not generated
FIREBREATH-137: Assigned To: richard
FIREBREATH-137: Priority: Trivial, Status: Open,
taxilian 10:10 yeah, I did just see that… they need to be inside for the mac fix, but you could add another else condition to do them on windows as well
Sudi 11:10 Good morning guys !!
taxilian 11:10 morning
Sudi 11:10 taxilian.. can I use .m file in fb project .. i mean I can create a mm file.. so if I add .m file and call it from .mm will it compile and work ???
taxilian 11:10 Sudi: you can use a .m file if you want, you just can't use C++ code in it, only C
why does it matter .m or .mm?
Sudi 11:10 nope.. I try prevoiusly to add .m file with objective-c code and it was giving errors .. some file was compiling well in webkit project so I was confused
taxilian 11:10 Sudi: you can use a .m file if you want, you just can't use C++ code in it, only C
I promise
Sudi 11:10 let me try it again :)
linearray 11:10 taxilian: what do you make of the talk in the forums about IE using several processes for one plugin?
taxilian 12:10 it doesn't
Sudi 12:10 wow.. it compiled well.. I guess should work now
taxilian 12:10 it uses several processes for different tabs
linearray 12:10 alright :)
taxilian 12:10 but only one for a plugin
welcome to the world of COM black magic
linearray 12:10 isn't that quite standard?
staylor 12:10 any reason with m_host->htmlLog would not work in IE9 but work fine in Firefox/Chrome?
taxilian 12:10 in all browsers a single plugin uses a single process
staylor: it may be that IE9's javascript console works differently; I don't recommend using htmlLog in production, so we haven't done much testing
staylor 12:10 well it's off in production, but I like being able to use it in debug builds
taxilian 12:10 I can understand that. I don't know why it isn't working, however
I haven't tried it in IE9
EL45 15:10 Hi all. Im getting ready to start porting my Firebreath plugin from Windows to Mac (and unfortunately I have never touched a mac). Currently I am trying to build the examples, but am running into some problems. For both BasicMediaPlayer and FBTestPlugin I receive the following error: 'Command /Developer/usr/bin/g++-4.2 failed with exit code 1'
Does anyone have suggestions where I should start debugging?
linearray 15:10 what exactly did you do?
EL45 15:10 Pulled FB from git, installed boost submodule, ran examples, from within buildex\ directory ran xcodebuild
linearray 15:10 OS X and Xcode version?
EL45 15:10 OSX 10.6.6 xcode: 3.2.5
linearray 15:10 ok, I don't have much experience with xcodebuild, but try opening the FireBreath.xcodeproj file in XCode and build from the GUI
it might give you a better error message as well
EL45 15:10 I did that and received the same result. That is actually where I saw the error messages.
I couldn't find why it failed with xcodebuild.
linearray 15:10 can you just paste the whole build output?
FireBreathBot 15:10 When you need to share code, logs, or anything else longer than a couple of lines, use a pastebin.,, and are all good options
EL45 15:10
I dont know how to output from the xcode GUI
linearray 15:10 hmmm
i686-apple-darwin10-g++-4.2.1: Programm: No such file or directory
i686-apple-darwin10-g++-4.2.1: plugin/firebreath/gen_templates/ExportList_plugin.txt: No such file or directory
this doesn't look good
it appears like the spaces cause problems
can you rename the directory to something without spaces and see if that fixed it?
EL45 15:10 My firebreath folder is located in a folder with a space. Im and going to try and rebuild
kylehuff 15:10 if you move the dir, you will probably need to rerun the prep script to recreate the build environment...
EL45 16:10 phew... just finished re-building. That was it. Thank you very much linearray and kylehuff!
linearray 16:10 you are welcome. can you post a ticket at about this bug, so we keep track of it?
kylehuff 16:10 cool. also, for your testing, you probably don't want to compile for every arch supported; you might save yourself some development/compile time if you specify a specific architecture. (currently it looks like it is trying to compile for x86_64 and i386, possibly more - but I could be reading the output wrong)
linearray 16:10 kylehuff: I *think* when you select x86_64 in xcode it will build for i386 as well...
unless you check 'build for active architecture only'
kylehuff 16:10 ah, okay. I seem to remember needing to specify to only do i386 on mine when using xcodebuild, but I mostly don't know what I am doing in OSX...
EL45 16:10 Me either Kyle :p
FireBreathBot 16:10 JIRA issue issue created by el45
kylehuff 16:10 well, if you end up running into architecture problems while building, here is my cli line for my project: "firebreath/ . build -DCMAKE_OSX_ARCHITECTURES=i386 -DCMAKE_BUILD_TYPE=MinSizeRel"
EL45 16:10 great, thank you.
FireBreathBot 16:10 JIRA issue issue commented by richard "you could perhaps provide a few more details? Such as, for example, exactly what happens when the..."
taxilian 16:10 EL45: seems like we have to fix that every few months or so :-/ if you can give me some logs that would help a lot
linearray 16:10 he did :)
taxilian 16:10 ahh, hadn't read the backlog. logs in chat aren't nearly as useful as logs attached to an issue =]
… the exportlist is having trouble? that's weird; I know we've had this working since the last time exportlist changed
FireBreathBot 16:10 JIRA issue issue commented by el45 "Received error message when using the xcode GUI:
taxilian 16:10 EL45: thanks. I'll have to look at that one
Sudi 16:10 taxilian.. if I change project settings to Standard (32/64-bit Universal) .. it should work in both 32 and 64 bit rigt ???
taxilian 16:10 if you make it 32 bit, then it will work in both 32 and 64 bit
if you make it universal then it will work in both but in 64 bit it will actually run in 64 bit mode
instead of using the 32 bit version of the plugin
Sudi 16:10 ok.. just want to check with you .. beside that I dont need to change anything to make it work ???
taxilian 16:10 most likely not
Sudi 16:10 ok.. thanks
taxilian 16:10 wow… in confluence 4.0 the RSS feeds work correctly finally for changes
works in ie8 as well
woops, wrong window :-P
dougma 20:10 say taxilian: the delayed invoke delegate stuff... what problem was that solving again? i wonder if it just hid the re-entrancy problems of the fireAsyncEvent. might try removing it now to see what happens.
taxilian 20:10 no, it solved issues with firefox4 and later
dougma 20:10 but what issues? hangs and crashes?
taxilian 20:10 the plugin would be "ready" to go before the scripting interface was useable
dougma 20:10 ah!
taxilian 20:10 so the onload event didn't work right
dougma 20:10 i see...
taxilian 20:10 then with the deferred event it ended up solving a lot of other issues as well, such as hangs and crashes when using an alert in an event handler
I am kinda surprised that we didn't run into your reentrance bug earlier, though; it certainly makes sense that there is a bug there
git pull requests
FireBreathBot 20:10 1 open pull request:
dougma: fix reentrancy problems around m_eventMap/m_evtIfaces
dougma 20:10 yeah.. it happens when we're really working the browser thread hard. :)
taxilian 20:10 to make sure I'm reading this correctly, you basically are just copying the event lists at the beginning of the function?
dougma 20:10 that's the idea
taxilian 20:10 seems reasonable
dougma 20:10 in my look at those types it looked to be safe
taxilian 20:10 can I ask a slightly annoying favor? =]
dougma 20:10 yes
(one sec afk)
taxilian 20:10 could you create a jira issue for this and rewrite the commit message to have the issue # on it?
then if you'd like you can push it into master yourself
dougma 20:10 ok... without a reproducible case i was reluctant. :)
taxilian 20:10 it's a valid problem; it is certainly something that could happen
windows has an annoying habbit of reentering your code in random places and ways
dougma 20:10 message pumps will do that do you. :)
taxilian 21:10 hehe
dougma 21:10 but here i think it's the javascript engine not unreasonably deciding to dispatch some random window.timeout code