IRC Log Viewer » #firebreath » 2011-10-11

IRC Nick Time (GMT-7) Message
byt 02:10 anyone here?
linearray 02:10 yes
linearray 03:10 taxilian:
taxilian 09:10 linearray: hehe; it's funny to watch the press and the companies battle for image
linearray 09:10 true
linearray 09:10 when building in VS2010 something is dragging in both msvcprtd.lib (== /MDd) and also MSVCRTD.lib (== /MD) even though MTd linking is specified for all targets.
this is what the error looks like
any idea what could cause this?
taxilian 09:10 is this a newer version of boost than we are using?
linearray 09:10 yes
I tried building it in several ways
taxilian 09:10 it's possible it's related to that; usually when I see this error it's because you're linking to a library that was compiled in a different mode
linearray 09:10 when/if you use system boost, do you pass anything specific to the build script?
taxilian 09:10 I don't, but on windows system boost uses autolinking
I think
maybe we disabled that, actually
I forget
Jamil 10:10 Hi all
taxilian 10:10 hello
linearray 10:10 hi
Jamil 10:10 I am moving from 1.3 to 1.6
seems there is a change in reading params
taxilian 10:10 *cringe* that'll be fun =]
better to stay up to date
Jamil 10:10 any doc
taxilian 10:10 shouldn't be terrible, though
sure, it's all in version history
all the breaking changes
the main difference is that you no longer declare the params you want to support, though
so m_supportedParams is gone (or whatever it was called)
Jamil 10:10 replace by what
yes it was m_supportedParams
no need
taxilian 10:10 yep, no replacement
just gone
no longer needed
Jamil 10:10 and
boost::optional<std::string> scTimeStamp = getPlugin()->getParam("timestamp");
taxilian 10:10 sure, you can do that if you want
I think
!find getParam
FireBreathBot 10:10 Found 1 possible matches. Displaying 1
/^boost::optional<std::string> PluginCore::getParam(const std::string& key) {$/ (f) found in src/PluginCore/PluginCore.cpp:
Jamil 10:10 how get scTimeStamp as string
taxilian 10:10 yeah, looks like it
I'd check if (scTimeStamp) first
Jamil 10:10 no convert needed?
taxilian 10:10 that's the purpose of boost::optional and that call
Jamil 10:10 great
taxilian 10:10 it's a boost::optional<std::string>, so whatever is inside, if there is something inside, will be a string
params can only be two types; a string or a function object
Jamil 10:10 another question
taxilian 10:10 and only a function object because firebreath will auotmatically look up anything referenced by "on<something>"
Jamil 10:10 openssl I was using it
any changes
to link to both openssl libs
taxilian 10:10 you're using the add_firebreath_library version or your own?
Jamil 10:10 add_firebreath_library
taxilian 10:10 nothing significant should have changed, then
Jamil 10:10 and openssl folder from the repository
taxilian 10:10 as far as how to use it
I think I updated the versions in my repo, is all
Jamil 10:10 i will check after finishing the compile error
last question
taxilian 10:10 replaced with shared_from_this()
Jamil 10:10 changed to shared_from_this()
taxilian 10:10 it's all on the version history page
!wiki version
FireBreathBot 10:10 8 results found. Note: Results limited to 8
"Development Versions":
"Version History":
"Old Versions":
"FireBreath 1.3.0 RC1 released!":
"Using Libraries":
"FireBreath 1.3.1 Released!":
"FireBreath 1.3.0 Released!":
Jamil 10:10 thx for now
taxilian 10:10 good luck
sabotaged|wk 10:10 so the status of drawing with firebreath on linux is still.. flaky?
taxilian 10:10 last I heard it was working fine
what is the problem?
some of you may find this useful: (tips on customizing visual studio to remove bloat)
hnan 10:10 hi, I tried to build my first plugin, everything works well except that the generated project is extrememly big, 450M and it takes a long time to build, I found that it mostly because the precomiled headers are extremely big, is this normal?
taxilian 10:10 pretty much, yes. We could add an option to disable precompiled headers, but it would make it take much longer to build
sabotaged|wk 10:10 i haven't even tried it yet. i was just poking around the fb website to see if video on linux was supported. and i saw mentioned on the wiki that it is, but flaky
taxilian 10:10 the good news is that anything that is not used gets optimized out, so your resulting binary in release mode is much smaller
sabotaged|wk: more specific might be to say that it is, but I don't know how it works
hnan 10:10 the result binary is fine
but it's really painful to wati for the building
taxilian 10:10 there are several who are using it
hnan: you should see it without precompiled headers
hnan 10:10 haha, but, Is there a way to reduce the headers, since I believe most of them are not used
taxilian 10:10 nope, most of them are used
otherwise we wouldn't include them
that would be silly
there could hypothetically be a few boost headers that get included by other boost headers that we do use that aren't used, but I doubt there are many, because boost is pretty good about that
unfortunately, boost is also very *very* heavy on headers
kylehuff 10:10 is the slow building a result of your setup/process? I build on 3 different platforms (some with 2 different architectures) and most of them are in visualized environments on a crappy laptop with 2gb of ram and doing a rebuild of a specific target doesn't take that long.
hnan 10:10 emm, there are really a lot of boost headers
what's that mean by setup/process?
I followed
taxilian 10:10 he's saying that the initial build takes a long time, but then when you're developing you usually only need to rebuild your main project
the rest doesn't have to be rebuilt each time
but if you're rebuilding the entire solution each time that may explain part of your issue
hnan 10:10 ah, yes, I right click on my plugin project and build, looks like it 'Rebuild All Started...'
taxilian 10:10 did you click build or rebuild?
hnan 10:10 rebuild
taxilian 10:10 don't click rebuild
click build
problem solved ;-)
(well, severity decreased, anyway)
kylehuff 10:10 word. doing a rebuild every time would be excruciating... not to mention, entirely pointless... =c )
hnan 10:10 em, I am doing build now, still building...:(
taxilian 10:10 once you hit rebuild, you have to rebuild
hnan 10:10 I clicked cancel and then build
taxilian 10:10 right
but you already told it to rebuild
so it flagged everything as needing to be rebuilt
hnan 10:10 ah, right, great!!!
now it builds immediately
learned something new, thanks all
Sajty 12:10 +1 for dropping ATL :)
taxilian 12:10 hehe. lots of people are
however, so far none of them have the budget to pay me to do it or the time/knowledge to do it themselves =]
Sajty 12:10 yeah, its easy to say, but its hard to do...
taxilian 13:10 dougma: wow; so the problem is that I'm an idiot. I totally messed up the scoping for the peruser thing, which is why it hasn't been working.
I feel really really stupid
I should have caught that in 5 minutes, not 5 hours
FireBreathBot 13:10 Commit f9afa1c on master by Richard Bateman: "Fixed tab issue"
Commit 672636c on master by Richard: "FIREBREATH-116: fixed per-user install from vs express or 20..."
Commit f9afa1c on firebreath-1.6 by Richard Bateman: "Fixed tab issue"
Commit 672636c on firebreath-1.6 by Richard: "FIREBREATH-116: fixed per-user install from vs express or 20..."
taxilian 13:10 okay, I need some hlep
who is around?
awesome. can you go to and see what things on there seem important to you, first of all?
I'm getitng ready to release 1.6rc2
FireBreathBot 13:10 JIRA issue issue resolved by richard "This is completely my fault; this was the result of a bug I introduced when I added a fix to work..."
linearray 13:10 phew
taxilian 13:10 hmm. FIREBREATH-123 should be really easy to fix; do you think you could throw that together?
FireBreathBot 13:10 FIREBREATH-123: Summary: Support for synchronous POST in SimpleStreamHelper
linearray 13:10 all three bugs seem important
FireBreathBot 13:10 FIREBREATH-123: Assigned To: richard
FIREBREATH-123: Priority: Minor, Status: Open,
taxilian 13:10 all you should need to do is copy the syncGet method and make a syncPost method out of it
linearray 13:10 I'll have a look
FireBreathBot 13:10 JIRA issue issue resolved by richard "I don't think this will be possible; I think a new http abstraction will need to be created to ad..."
taxilian 13:10 thanks.
kylehuff, do you have any time?
linearray 13:10 windows is an abomination
I'm going back to xcode...
taxilian 13:10 lol
kylehuff 13:10 taxilian: back, sorry - was out and about.
taxilian 13:10 for that one that should be just fine
no problem; you have a linux environment easily useable, right?
can you test the latest in master and let me know if it works?
kylehuff 13:10 yes, I do. if you just need a successful build against master on linux, I can do that. (did it yesterday, but will do it again today to be sure.)
taxilian 13:10 that's mainly all I need
though if you could run the test for FBTestPlugin that'd be good too
kylehuff 13:10 sure, not a problem. as soon as my project is done building I'll build FBTest
taxilian 13:10 awesome, thanks
if you have time to put together a quick list of hcanges since rc1 that'd be really helpful also
but if not I can do it; it shouldn't be too hard
I'm trying to pull in one more patch from someone
(in case you were wondering why I'm not doing these things myself)
kylehuff 13:10 pfft, yeah... I was wondering why we are doing all the work around here... not even sure why you hand around anymore. this place practically runs itself...
taxilian 13:10 ;-)
you know, you're right… good luck, all!
kylehuff 13:10 lol
taxilian 13:10 if I were really a jerk, I'd leave and not come back for a few days, just to make you stew :-P
kylehuff 13:10 heh, I'd just blame it on linearray not jumping high enough when you said jump... =c p
taxilian 13:10 lol
kylehuff 13:10 anyway, I am preparing for a meeting that I have to attend that starts in 15 minutes, so I might not be able to work on the release notes until after my meeting
taxilian 13:10 no prob
linearray 14:10 is _NO_MAIN supposed to be defined?
taxilian 14:10 not usually, no
linearray 14:10 hm can't compile FBTestPlugin
taxilian 14:10 on mac?
linearray 14:10 xcode 4.2, neither with llvm nor gcc
taxilian 14:10 yeah, I'm seeing a build error somehwere as well
let me look at it
I need to go get lunch, though :-/
linearray 14:10 I was trying to write a test case for SyncPost
do you want untested code, or get tested code tomorrow? :)
taxilian 14:10 hehe. can you put it in a fork and I'll pull it and test it?
linearray 14:10 sure
or actually I can test it in a pet plugin
taxilian 14:10 9c23e342bb8f5f
FireBreathBot 14:10 9c23e34 by Richard Bateman: Updated log4cplus to work w/ 10.4 SDK on mac
taxilian 14:10 hmm. that "Fix" there may have broken your build
you did update to the latest, right?
linearray 14:10 yep
taxilian 14:10 hmm. did you try deleting the build dir / starting with a fresh build dir?
linearray 14:10 yep
taxilian 14:10 this is odd, it builds fine on my machine
I'll have to look at this more
linearray 14:10 I'll add the gcc output
right... now it works
gotta love elusive errors
still broken with LLVM though
ok, I added the (occasional) gcc output to the file
kylehuff 15:10 I forgot to report back, building my pluing on linux was without incident, same for FBTestPlugin. All tests green (except the ones that should fail)
taxilian 15:10 kylehuff: thanks
FireBreathBot 15:10 Commit 0b35146 on firebreath-1.6 by Richard Bateman: "FIREBREATH-130: add Navigate method to BrowserHost"
Commit 0b35146 on master by Richard Bateman: "FIREBREATH-130: add Navigate method to BrowserHost"
taxilian 15:10 linearray: which platform are you on? 10.6? 10.7?
linearray 15:10 10.6
FireBreathBot 15:10 JIRA issue issue resolved by richard "Thanks for the patch; I had to make some tweaks on the IE side. Not sure why you were creating a ..."
taxilian 15:10 hmm. I wonder why I can't repro it. what is the gcc output?
do you have it handy?
linearray 15:10 on the bottom of the page
I forked FB some time ago on github... anything I need to do to stay current?
taxilian 15:10 pull from the real one now and again
linearray 15:10 how? :)
I know how to do it locally, but not on github
taxilian 15:10 you clone, do it locally, and then push to your fork
it isn't automatic
linearray 15:10 i see
taxilian 15:10 git remote add fbcore git://
remotes are your friend =]
linearray 16:10 !pull
git pull requests
FireBreathBot 16:10 1 open pull request:
linearray: Added SynchronousPost(.).
linearray 16:10 this is still untested, because my test case always died on me this is what I used, but it never joins the thread and just hangs
manoj 16:10 hey ...
linearray 16:10 yo
taxilian 16:10 hello
manoj 16:10 i'm trying to download something within firebreath, using the wiki instructions...
the download works
but interestingly, my evt->getProgress() is always 0
i mean, inside the onDataArrived event handler
linearray 16:10 taxilian: if you can live with that until tomorrow I'll be on my way
manoj 16:10 onStreamDataArrived, sorry
taxilian 16:10 it could be that the server just isn't telling the browser how big the file is
linearray: did you push it somewhere?
linearray 16:10 00:05 < FireBreathBot> 1 open pull request:
taxilian 16:10 oh, missed that
I'll play with it
linearray 16:10 ok, cu tomorrow
taxilian 16:10 c u. I appreciate your help
manoj 16:10 hmm... why* does this happen? why would'nt the server tell the browser?
taxilian 16:10 ask the server
I don't know
manoj 16:10 :)
taxilian 16:10 some servers don't give a content-size header
manoj 16:10 oh...
taxilian 16:10 node.js servers don't, for example
manoj 16:10 so, i'm just trying to download an executable
taxilian 16:10 and in that case there is no way to know how big it is
manoj 16:10 i have a direct link to http://blah/something.exe say
is downloading a file like that done using another method...?
taxilian 16:10 which method are you using?
did you implement your own browserstreams handler, or are you using SimpleStreamHelper, or..?
manoj 16:10 i have a class that overrides DefaultBrowserStreamHandler
and the onStream* methods
so i guess, using browser streams
taxilian 16:10 to be honest, I don't think I've ever even checked that particular value, so I don't know for sure if it works
I recommend looking through to see where it gets set
and figure out why it isn't getting set
manoj 16:10 alrighty...
taxilian 16:10 sorry I can't help better
manoj 16:10 okay, i'll do that...
no thanks for the tip
fyi, when i downloaded the png file mentioned in the wiki document (, i could see 2 values for the progress, 0 and 99
and then, the download completed
taxilian 16:10 sounds like it works, then
manoj 16:10 anyways, i'll dig around.. thanks again :)
taxilian 16:10 good luck
that's quite the original nick
fdsfa 16:10 Is there a best practice for writing/reading a registry key with firebreath? Also anyone have suggestions for an alternative to reg keys on MAC
taxilian 16:10 I would definitely suggest that you not delete things you shouldn't and make sure you access the registry responsibly
other than that, no, not really
on the Mac you can use the preferences plists
fdsfa 16:10 will do... Just wanted to make sure there wasnt something already out there
taxilian 16:10 someone started a configuration manager intended to eventually be used in firebreath
but never finsihed it
would be neat
fdsfa 16:10 will keep that in mind
Sudeep 16:10 hi
I am trying to use objective-c code in fire breath
can any one help me in right direction
taxilian 16:10 yes. go look at
FireBreathBot 16:10 Found 1 matching file(s) in the master branch. First 1 are:
Sudeep 16:10 ok.. thanks
FireBreathBot 18:10 Commit 36d16b9 on master by Max Amanshauser: "FIREBREATH-123: Added SynchronousPost()"
Commit 04dd1f5 on master by Richard Bateman: "Changed syncget test to use syncpost to validate feature"
Commit 36d16b9 on firebreath-1.6 by Max Amanshauser: "FIREBREATH-123: Added SynchronousPost()"
Commit 04dd1f5 on firebreath-1.6 by Richard Bateman: "Changed syncget test to use syncpost to validate feature"
JIRA issue issue updated by richard
JIRA issue issue resolved by richard "Fix is verified; thanks, Max!"
taxilian 18:10 if everyone could please test the master branch, I plan to release firebreath-1.6rc2 in the morning, barring any issues anyone finds before then.
taxilian 21:10 nirvdrum: are you on the latest master branch?
nirvdrum 21:10 I am now.