IRC Log Viewer » #firebreath » 2012-12-13

IRC Nick Time (GMT-7) Message
taxilian 00:12 HA! That was the #1 issue that needed to be fixed for 1.7; I think we're about at RC status now
reichi 00:12 oh
at this time?
taxilian 00:12 yep
reichi 00:12 good morning ;)
taxilian 00:12 good morning
reichi 00:12 any news on the new logo?
taxilian 00:12 well, didn't really get any feedback about it other than a tiny bit here
and I got sidetracked and haven't thought more about it =]
niftylettuce_ 00:12 anyone here doing a kickstarter/indiegogo project or know someone that is?
cafe 04:12 Hj All
my plugin run normal on Firefox, chrome
but IE crash
please help me
didu 07:12 Hello everyone, did someone have knowledge about m_host->getDOMDocument() and can help me :) ?
nirvdrum 07:12 didu: What do you need help with?
didu 08:12 I have a error from boost shared_ptr when I do => m_host->getDOMDocument()->getElementById("lastname");
and if there is no "lastname" on the current webpage
Did I have to make a check if there is a "lastname" before I call getElementById ?
didu 08:12 What kind of exception we have to catch with Firebreath :P ?
taxilian 09:12 didu: it will throw an exception if you try to getElementById and there isn't one
you just need to catch the exception
I'm thinking that should be changed to not throw an exception but just return undefined
actually it can't
so there you go
though maybe it should return NULL...
anyway, just wrap it in a try … catch and you'll be good
———— so good morning all. I have a question. I'm about ready to release FireBreath-1.7.0rc1. Does anyone know of any issues that need to be resolved before I do this?
jshanab 09:12 good morning. I can't think of anything. BTW how do I get the browserhost pointer? is there a static getter for it? or do I need to pass it around so my class can use it. (i need the ScheduleOnMainThread
didu 09:12 Thanks you taxilian but I have a popup crash on firefox about shared_ptr != 0 when I do =>
taxilian 09:12 jshanab: no static getter; it's specific to the plugin instance
didu: that assertion failure means that you have a NULL pointer
it's either m_host or the return value of getDOMDocument
find out which
didu 09:12 taxilian: <== I have the same issue
I think it is a shared pointer use on getDOMDocument because I have only this on my code
and if the currentweb page have the id "lastname", there is no assertion failure
taxilian 09:12 didu:
you can't actually check if a shared_ptr == NULL
it doesn't
but you can check if (doc) ...
hmm. Actually there could potentially be a bug in getElementById, now that I think about it..
didu 09:12 I try your code but that change nothing
taxilian 09:12 well, probably time to start debugging
I can't track down everything for you, I'm afraid. it might be in FB code
so feel free to dig down in
add some log statements, step through if you can attach a debugger, find where it is
it's not that complicated of a calling chain
didu 09:12 haha Ok :) I will try
taxilian 09:12 it seems most likely to me that the issue is in your code, but you could also try reproducing it with a simple example in FBTestPlugin; if you can do that you can submit a jira ticket and I'll look at it
I just need sample code that reliably reproduces it
didu 09:12 OK I will do it :) Hum otherwise I want to make use getElementsByName
taxilian 09:12 whatever you like
didu 09:12 But I get a exception where the message is just "getElementsByName"
taxilian 09:12 okay, I'm eating breakfast and going downstairs to my office. be back in a bit
didu 10:12 taxilian I have debug the plugin and Firebreath
to find where it crash
taxilian 10:12 and?
didu 10:12 (I just relaunch the debug for getting the exact line :P)
On the file Node.h
line 95: FB::variant tmp = m_element->Invoke(name, args); <= tmp = 0x13766950 FB::variant_detail::lessthan<FB::variant_detail::null>::impl(const boost::any &, const boost::any &)
I think it is good because he found nothing on the webpage right ?
but line 96: return tmp.convert_cast<T>(); <= that return something empty
jshanab 10:12 What is the generally acceptable or best way to make plugin installer for OSX
didu 10:12 so when he try to create on Element.cpp line 87: return Element::create(api); there is a crash
FB::JSObjectPtr &api is totally empty so when he try to api->getHost() on static ElementPtr create(const FB::JSObjectPtr &api) { return api->getHost()->_createElement(api); } line 61
there is a assert
taxilian 11:12 if anyone is interested, I just posted a nice long rant on the google group =]
didu 11:12 It is cool to read such message from admin of a project :) I hope you will find the help you need !
taxilian 11:12 honestly I don't really expect to
but it feels good to vent now and again
and hopefully at least people will start stepping up a bit to fix their own problems more
didu 11:12 like my problem :P ?
taxilian 11:12 yours is a fairly mild case; there does come a point where you'll just have to track it down yourself, though =]
didu 11:12 Yes because the crash is deep on Firebreath and I don't know if everything before the crash work normally
taxilian 11:12 you're on linux, right?
didu 11:12 Windows :( (I prefer Linux :P)
taxilian 11:12 ahh
well that's easy then
attach a debugger
then you can see the stack trace of where ti's actualy crashing
and it'll save you a lot of time
jshanab 11:12 I have been debugging recently on the big3, apple,windows, and linux. I found kdevelop worked really nice on my linux boxes (even posted a bit on about it)
didu 11:12 I did it already :) That's why I give so many detail :P
taxilian 11:12 ahh
sorry, I missed something you wrote
so that is interesting
let me think about that
didu 11:12 I can follow the callstack but I don't know if the call are normal or not because I don't know it enough :(
taxilian 11:12 yeah, that one is actually a bug
but I can fix it pretty easily
give me a minute
basically if the API is null it shouldn't be creating an ElementPtr to wrap it, it hsould be returning a NULL ElementPtr
didu 11:12 I just put a if ? :O
taxilian 11:12 yeah
I'll make it give you a NULL elementptr and you just do if (retval) { … }
and that will only be true if an element was returned
didu 11:12 Ok I try to fix it like you said
taxilian 11:12 didu: pull that, it fixes it so it gives you the NULL pointer
didu 11:12 Oh Ok ! thanks you :P !
taxilian 11:12 kylehuff: don't suppose you're up to compiling a changelist again are you? =]
lucent_js 12:12 Hi taxilian
didu 12:12 taxilian: The patch work thanks you a lot :) !!!
jshanab 12:12 So....I just pulled and something changed in (log4cplus) cmake? I am getting an error about get_filename_component) not having the correct number of arguments
lucent_js 12:12 Hi, taxilian..Thanks for helping yesterday. I have another newbie question to which I am unable to find an answer. I am instanting a object in my plugin and returning a pointer to it to Javascript. In another API I am accepting the same pointer as input. Now how do I cast it back to my class type ? unable to figure out how to do it usinf FB::variant
didu 12:12 taxilian: do you know if it is possible to do: because I try to make the same as getElementsByTagName but it throw an exception with "getElementsByName" because the invoke failed.
(I normally do the same as getElementsByTagName function)
jshanab 12:12 Cmake error solved. I agreed to a small OSX update a few days ago. Terminal I prep in was older than that. Get a fresh terminal and prep ROTFL
jshanab 12:12 Cmake error still there :-( The error is there the first time i run prep on an freshly deleted build directory. It does not show up second time
taxilian 12:12 lucent_js: you can't, really
because the JSObjectPtr you get back in isn't the same as what was sent out
jshanab 12:12 There is a warning about pthreads. but my main problem is new errors in the build. lockfile.cxx no usable file locking and undeclared identifier OPEN_FLAGS This is all in log4cplus updates? I think
taxilian 12:12 all you can do is have some kind of global map that you can find it in by an id or something
didu: you'd have to try it
jshanab: odd; it's working for me with the new log4cplus
jshanab 12:12 Maybe it is the warning about pthred. The error metions pthread locking
#IF ! DEFINED (log$cplus_use_posix_locking...
lucent_js 12:12 aah..thanks a lot..that's what I have implemented while waiting for your reply :) . anyways it's good to know there's no other option and I am doing the right thing
jshanab_ 12:12 Is there a way to check the status or valu of pre-processor defines in Xcode? THere is a rather complicated ifdef block to determine the defines needed in lockfile.cxx
taxilian 13:12 probably
(sorry, on the phone)
jshanab 14:12 After a pull today, i cannot build my plugin. I am on mac osx using 10.7 and have traced it as far as a problem with the new log4cplus in which LOG4CPLUS_USE_POSIX_LOCKING is not set. This is becasue the 4 possible optins are also not set. This may be because log4cplus does not have LOG4CPLUS_HAVE_fcntl_H is not set. That is as far as I can trace it and I am kinda stuck.
Anyone building todays pull on mac in xcode?
taxilian 14:12 I built it yesterday
but I'll try it again
can you build FBTestPlugin?
jshanab 14:12 ok, will try that
taxilian 14:12 I want to know that is fixed before I release 1.7, so keep me posted
jshanab_ 14:12 here is my prep command
../common/FireBreath/ FBTestPlugin build/FBTestPlugin -D CMAKE_OSX_SYSROOT=/Applications/ -D CMAKE_OSX_DEPLOYMENT_TARGET=10.7 -D CMAKE_OSX_ARCHITECTURES=x86_64
jshanab 14:12 same errors. I am doing a deeper clean and re-trying
There must be something wrong or changed on my machine. I get those errors and then a boatload more before clang aborts on me
jshanab 15:12 What compiler do people see detected and used in xcode after preping the FBTestPlugin? I am seeing Apple LLVM compiler 4.1
jshanab_ 15:12 FBTestPlugin builds if I comment out add_firebreath_library(log4cplus) in PluginConfig.cmake
taxilian 15:12 hmm. okay. give me a few
jshanab_ 15:12 NP. Just keeping you posted. I will do the same in my project and figure out the dmg and we can attack that when you have more time.
taxilian 15:12 I have more time right now
let's figure this out
jshanab_ 15:12 ok
taxilian 15:12 I'll try building FBTestPlugin with the parameters you just specified
what version of cmake and xcode?
jshanab_ 15:12 cmake 2.8.10, xcode 4.5.2
taxilian 15:12 I'm on
probably doesn't matter
jshanab_ 15:12 oh let me check the build number
taxilian 15:12 [email protected]:~/code/FireBreath $ ./ examples/FBTestPlugin buildex -D CMAKE_OSX_SYSROOT=/Applications/ -D CMAKE_OSX_DEPLOYMENT_TARGET=10.7 -D CMAKE_OSX_ARCHITECTURES=x86_64
jshanab_ 15:12
taxilian 15:12 hmm. well, we'll see if I can repro, if not that's the first thing I'll try
jshanab_ 15:12 I noticed that xcode now complains (sdk not found) on the far right edge of that path. but recognizes the sdk elsewhere. I changed it to 10.8 and 10.7 but same issue.. I am down to just those 2 issues i put in email
taxilian 15:12 so mine succeeds
you're using built-in boost, all submodules are updated, and you're on the latest master?
jshanab_ 15:12 built in, yes to all submodule updates and yes on latest master. (let me know if you want to drive)
taxilian 15:12 probably a good place to start
let's do that
jshanab_ 15:12 joinme 163-980-346
(office has peopple in it so we will stay off voice chat for now)
taxilian 15:12 okay. I'm in
just for a sanity check, we'll try building it where I normally do
now isn't that interesting?
jshanab_ 16:12 Looks like clone != pull
taxilian 16:12 which means somehow you still have changes in your other clone
let's go look
jshanab_ 16:12 oh shucks our boost traits fix
taxilian 16:12 actually I don't think this one was needed, was it?
this isn't one of the ones we made while I was working with you
at any rate, let's try turning it off
Step 1 is checking to see if that fixes it; step 2 is seeing if your stuff still builds after if it does
I can't think of any reason that should fix it, TBH
jshanab_ 16:12 TBH?
taxilian 16:12 To Be Honest
jshanab_ 16:12 LOL I am behind on my TLA's
taxilian 16:12 as opposed to TBATJWITTKY (To be a total jerk who is trying to kill you)
ya gotta realize that my acronymns came more from morse code and ham radio than from "popular" texting
okay, I'm done
it's fixed
let me think how to fix that properly...
jshanab_ 16:12 WOW!
taxilian 16:12 I would still see if you can get by with reverting those two files and still build
I am now checking to see if I can build using the 10.5 sdk on mac os 10.8 and 4.5.2
jshanab_ 16:12 You popped the stash so the files are revereted? I am gonna prep and build
taxilian 16:12 popped the stash so the files that I reverted are no longer reverted
stash it again to try with them reverted
and I'll remove that file from .gitignore in the trunk
that was entirely my fault, btw; it's a file that I configured in a bad place
over a year ago
fixed it when I updated (almost accidentally) but since the old file was still there, it picked up the wrong one
jshanab_ 16:12 Is there any documentation on the new dmg creator code
since this is in log4cplus, I thought I had better point it out, is this on my machine only? line 56 in tchar.h is dsfsdafdsafds
taxilian 16:12 pretty much… no.
jshanab_: yeah, someone else just pointed it out on the mailing list =] that was not intended to be checked in, it was to make sure that branch of code was never executed
jshanab_: you back in business then?
jshanab 16:12 well...I have linker errors on log4cplus and am checking to see if i am abusing.
taxilian 16:12 I'm working on building w/ the 10.5 sdk; it's requiring adding type_traits in several places :-/
jshanab_ 16:12 oooh
taxilian 16:12 oh the bright side, it should help with your issue
that'll need to go in the FAQ, though
I'm glad you told me about that; it has saved me a ton of time
jshanab_: pull that, it may help
jshanab_ 16:12 here is one for you
taxilian 16:12 you don't see it on FBTestPlugin I assume?
jshanab_ 17:12 I want to get a packaged plugin for the mac, even with a few features disabled, I need to get back onto the Android app.
Not in FBTestPlugin. That is mine.
jshanab 17:12 pull says my .gitignore change will be overwritten. stash and go?
link_libraries 17:12 hey guys I'm having trouble linking a custom framework on OSX can anyone help me out?
jshanab_ 17:12 There is CMAKE for adding a framework to an xcode project. What is your particular question? Ask away
link_libraries 17:12 so first I added the line include_directories(/Library/Frameworks) to projectDef.cmake
then I added find_library(NAME_FRAMEWORK Name)
and finally I added the variable to target_link_libraries
but whenever I tried to include the framework in one of the plugin files in xcode
it says that the framework does not exist
jshanab_ 17:12 I have had to use MESSAGE(${INCLUDE_DIR}) all over in my cmakes to see what it finds. Most the time it is and "oh shit it is blank" moment
here is my opengl section in the same file
find_library(OPENGL_FRAMEWORK OpenGL)
find_library(QUARTZ_CORE_FRAMEWORK QuartzCore)
# add library dependencies here; leave ${PLUGIN_INTERNAL_DEPS} there unless you know what you're doing!
link_libraries 17:12 yeah I added the previous right in that section
jshanab_ 17:12 Maybe frameworks are added without need for include mods. framework specifing both include and code path?
link_libraries 17:12 what exactly do you mean?
jshanab_ 17:12 I realize when I look at mine I do not have and include and and library. I just have the framework. Taxilian can help more on frameworks, I am still a bit new to osx
taxilian 17:12 gah, this no longer has been notifying me when someone tries to talk to me...
jshanab_ 17:12 First order of buisness chastise me for not using paste :-)
taxilian 17:12 link_libraries: you shouldn't need to add /Library/Frameworks as an include
link_libraries: also look at the Using Libraries page on the wiki
link_libraries 17:12 that's what i figured but for some reason the framework wasn't showing up
I did
are things different for custom frameworks
because mine isn't in /System/Library/Frameworks
jshanab_ 17:12 Uh. Isn't the point of a framework, the automtic switching of directory based on sdk choice? Which would mean it must be "installed" in a known location
taxilian 17:12 nah, there are other reasons for a framework
mostly that it's all self-contained
easy to distribute
jshanab_ 17:12 ah. My mac education continues.
Is there anything else besides add_firebreath_library(log4cplus) needed? my plugin is having linker issues with log4cplus
taxilian 17:12 are you doing anything in any of your projects that would cause any of the log4cplus includes to be included
jshanab_ 17:12 includes in the cpp or do you mean in the cmake
taxilian 17:12 cpp
jshanab_ 17:12 I use the logger starting with something like "logger_ = Logger::getInstance(LOG4CPLUS_TEXT("CLivePlayerMac"));" in most ctors
taxilian 17:12 oh
yeah, that'll do it
wow. yep, that'll definitely do it
hang on =]
jshanab_ 17:12 I have been doing that for a year
taxilian 17:12 add "add_definitions (-DUNICODE -D_UNICODE -UMBCS -U_MBCS)" to your CMakeLists.txt file
should fix it
I didn't realize you were using log4cplus more than just with regular firebreath logging :-P
might have to have you teach me how to use it properly one of these days
jshanab_ 17:12 I do have the -DUNICODE in my windows one
lucent_js 17:12 Hi taxilian, If you remember the typed array issue I had yesterday. Would you be able to comment on what people have done in the past to address it. ? Is there a trick or hack ? What is the best possible way to do it if I don't have to retrieve any data from it but to only copy back and forth in internally allocated buffer. Can there be a JS closure solution like you proposed yesterday for copying to another host buffer ?
taxilian 17:12 actually generally people just don't use typed arrays with plugins
the sad fact is that there is no really performant way to transfer large blocks of data to or from a plugin
string is the closest thing
for binary data I usually end up using base64
lucent_js 17:12 okk.
base64 ?
taxilian 17:12 google it
lucent_js 17:12 ok
jshanab_ 17:12 taxilian. That did not fix it but I can see that it is looking for getInstance(string) and not wstring
uh, tchar.h looks wrong to me
oh, nevermind that is fine.
taxilian 17:12 jshanab_: Yeah, that's the issue
jshanab_ 17:12 It drives me nuts that xcode cannot show you the value of the defines.
taxilian 17:12 gotta figure out why ti's doing that
jshanab_ 17:12 On windows you can't chose both MCBS and UNICODE. Maybe they are mutually exclusive
taxilian 17:12 … huh. I copied that line from the log4cplus cmake
I didn't write it
jshanab_ 17:12 Is that what the -U was? undefine
taxilian 17:12 and it's working elsewhere
maybe so
wasn't really payign attention :-P
jshanab_ 17:12 I have never seen that. me neither
Low blood sugar, it is dinner time in this timezone
taxilian 17:12 hehe
it's 6pm in this timezone, and most people consider that dinnertime
jshanab_ 18:12 I have my add_firebreath_library for bosot above log4cplus and notice that it finds log4cplus stuff during the boost add_library call. I don't know if cmake is really that order dependent though
I keep thinking you are 2 hours off from me
grr. this new log4cplus is being a pain
taxilian 18:12 you're central time?
jshanab_ 18:12 yup, Chicago
taxilian 18:12 eh, I know where that is, but I don't *really* know where that is :-P I don't do geography
jshanab_ 18:12 Topright corner of illinois or bottom left corner of lake michigan ;-) about 2800 miles from my house in California
taxilian 18:12 hehe. yeah, wherever that is :-P My point is just that I can't think of a place and think "oh, that's in this time zone". I know which time zones CA and NY are in, but tha'ts mostly just 'cause I've worked for companies in those places =]
jshanab_ 18:12 Yeah. timezones, DST, etc is always an interesting topic. Hits me in every job when I have to explain why every server must be UTC don't convert until the last second. (server in chicago, customer in california and camera in brazil. What time is 8am friday video)?
I have that line defineing UNICODE in one form or another in at least 10 locations :-)
taxilian 18:12 lol
jshanab_ 18:12 But I cannot find any undef. However when I look in the project I see the MCBS flags but NOT the UNICODE flags in the Other CFlags and othe r C++FLags in xcode
They are Undefines, oh god, did I spell unicode wrong?
taxilian 18:12 lol. well, not that I've seen....
but I make no guarantees =]
I'm finally really starting to feel good about the 1.7 release
jshanab_ 18:12 I see Debug, UNICODE etc in zero_checks GCC_PREPROCESSOR but Not in other projects cflags.
taxilian 18:12 huh
jshanab_ 18:12 If I go to FireBreath project or my plugin target I cannot see the UNICODE in the Other C Flags or other C++ Flags section, they are default. As I move thru the targets, i see all the undefines and some defines but never _DUNICODE
taxilian 18:12 interesting
I don't know :-/
jshanab_ 18:12 almost like clang 4.1 does not allow it
jshanab_ 18:12 looking closely at my usage. I have directories and static libs I have added to the project. They have a target_link_libraries( mylib log4cplus)
In the past this pulled in the log4cplus. It is these static libs that are having the problem.
taxilian 18:12 ooh
jshanab_ 18:12 "/Users/jshanab/archiver/src/FireBreathPlugin/build/MVSArchivePlayer/fblibs/log4cplus/src/Debug/liblog4cplus.a" is in the -L line so I still think it is a UNICODE thing
when I bulit log4clus for a desktop ap, it actually had different names for the unicode libs.
taxilian 18:12 this used to have that
I disabled it
because it caused problems with the way I was including it
and I was too lazy to fix it the correct way
jshanab_ 18:12 With the ability to re-prep, I don't think we need seperate names. I am just looking for what changed that is breaking my build
Mac nm does not demangle?
gotta pipe thru c++filt, lovely
nm confirms log4cplus is indeed built for unicode
taxilian 19:12 right
jshanab_ 19:12 I think I see what is going on. I now need to add_definitions into all my cmakeLists.txt that are in sub folders I include with add_directory. Let me test
taxilian 19:12 would think that the parent would be enough :-/
maybe we need to add the define to the buildsettings.cmake in firebreath
to avoid this
and now to The Great Remaining Task
actually, could you try that?
try adding those to buildsettings.cmake
jshanab_ 19:12 Unless project dependency order changed, I don;t see why it changed from this morning
taxilian 19:12 because before there was a #define UNICODE or something similar in defines.hxx
and it's not there anymore
jshanab_ 19:12 ok
taxilian 19:12 and I will begin on The Great Remaining Task (TGRT for short)
jshanab_ 19:12 where is buildsettings.cmake?
taxilian 19:12 namely, the Building Of The Changelog
it's in cmake/
jshanab_ 19:12 Changelog - jeffs skin crawls
taxilian 19:12 I hates them
but they are pretty important for a project like this
jshanab_ 19:12 I love having them, I hate makeing them
taxilian 19:12 realize I have to go through an *entire year* of commits
jshanab_ 19:12 Ther are a bunch of .cmake files but no buildsettings. could it be buildconfig.cmake?
taxilian 19:12 yeah, that's it
never assume I've remembered a name correctly =]
particularly not that one; dunno why, but I can never remember the correct name for that file
jshanab_ 19:12 I see some UNICODE, but it is in the win32 section. I think all linux/apple are unicode, but we actually need it specifically stated for some libraries to detect and use it
taxilian 19:12 you shoudl be able to add it to the cflags
make sure it's the correct defines
you'll need both -DUNICODE and -D_UNICODE
I bet one of them isn't there
jshanab_ 19:12 Neither for UNIX UNICODE and _UNICODE is only in win32
but they should use CMAKE_CXXFLAGS hummm
addative I see
Build succeeded!
taxilian 19:12 nice
okay; do you know how to do a pull request?
jshanab_ 19:12 Let me clear out the cruft I added and make sure that is all we need. Does it need to be selectable?
taxilian 19:12 what do you mean?
jshanab_ 19:12 nope. I was wondering about that before
taxilian 19:12 wanna learn?
jshanab_ 19:12 Well it is posible to create a unix install that is not unicode. right?
taxilian 19:12 I don't think so, no
jshanab_ 19:12 most definitly wanna learn
taxilian 19:12 FireBreath expects (and has always expected) log4cplus to be unicode
it's not that hard
1) fork FireBreath on github
2) Put your changes in a commit
3) Push them to github
4) go to github and create a pull request
One trick to know is that it's always easiest if you keep all your changes related to each other in a given pull request, and if you put more on the branch with the pull request it'll add that to the request
so it's often a good idea to create a new branch for each change and put just that change on it
in this case that doesnt' really matter since I'll be accepting it almost immediately
jshanab_ 19:12 when I fork on github I then have to pull it make changes and test. Normally we start with that. or is my existing pull that first step. the "forking" is not entirely clear. is it just a branch
taxilian. I sent you the whole file via email. I am having trouble looking at the screen any longer. Gotta eat and get home. I WILL learn and practice the git pull in the future
Thanks for all the help today.
taxilian 19:12 that works
take care
jshanab_ 19:12 I realize that as I will be constanly building my plugin on linux,mac,and windows. Good test bed for updates.
taxilian 19:12 yep =]
jshanab_ 19:12 Man this is cool. Just gotta figure out packaging in the morning
taxilian 19:12 hehe. cool. you can look at hte fbgen template for an example
jshanab 19:12 Oh, thnaks for turning off PCH
taxilian 19:12 lol
niftylettuce_ 20:12 taxilian_away: you should launch kickstarter campaign to build open source alternative to crossrider and kango
then use for stickers/goodies :D
taxilian 21:12 lol
except that I don't personally really care about extensions =] and even if I had funding I wouldn't really be interested in leaving my current job to pursue something like that =]
taxilian 21:12 I have just completed and am now uploading the new FireBreath Getting Started tutorial for Mac
niftylettuce_ 21:12 lol
make firebreathe tees on kickstarter