IRC Log Viewer » #firebreath » 2011-01-07

IRC Nick Time (GMT-7) Message
NoAntzWk 04:01 Hi nitrogenycs!
nitrogenycs 04:01 Hi NoAntzWk
NoAntzWk 04:01 Were you able to check the issue 125? Do you happens to you too?
I think that download streams are broken with FB v1.3.2 and above.
nitrogenycs 04:01 I've read through the issue now
At the moment I can't fix it as I have to work on other things
You can go ahead and fix this if you want, I'll answer questions you might have
The other solution is to wait till I get back to this which might take some time
NoAntzWk 04:01 Ok. I just wanted to verify that happened to someone else.
nitrogenycs 04:01 It seems strange that IE and FF are broken. The implementation is quite different. Also, I haven't tested seekable (readRange) streams in a while as I don't need them anymore
NoAntzWk 04:01 There are some changes affecting streams in FB v1.3.2, I suspect that these changes has introduced the bug but by now, I haven't time to find the bug. I will should update to 1.4 alpha version and I will should test it again.
NoAntzWk 05:01 I do not use readRange(). Yes is strange that all browsers are broken. So, I suspect the problem is not in the browsers but the bug has introduced in high-level changes from version 1.3.2.
jtojanen 06:01 Taxilian, add Detach() to CComBSTRs in JSAPI_IDispatchEx.h line 428 and 429
scJohn 07:01 taxilian: i am trying to use express with the latest git pull.. i seem to have forgotten how to use the prep scripts.
CMake Error at CMakeLists.txt:91 (MESSAGE):
any ideas?
i must have just called prep2010.cmd incorrectly
iaincollins 07:01 Hmm it shouldn't need any arguments
(unless things have changed)
scJohn 07:01 i had to specify projects and build..
FireBreath\prep2010.cmd projects build
iaincollins 07:01 Oh I see, your projects dir is outside of the firebreath dir
(I should be doing that...)
scJohn 07:01 ah
it rebuilt the build dir from scratch.. compiling now
iaincollins 07:01 I find I need that with major changes (e.g. if I change CSLID)
I can't remember the full list of reasons why that might be needed, but have seen it on a wiki or blog page somewhere
would be nice if the prep scripts could catch that
scJohn 08:01 any idea why this file is not found "Win/win_common.h"?
iaincollins 08:01 hmm no, was the project directory created using scripts from an older version of Firebreath maybe?
scJohn 08:01 yes
i had to copy a few files from the current firebreath to the project
iaincollins 08:01 I found the change notes very good from 1.2.x to 1.3.x but that's the only migration I've done so far
(I think the steps were covered on the Wiki)
I am not sure if there are notes for breaking changes in 1.4 yet
(well, I upgraded a couple of times before that, but there were no breaking changes at the time)
taxilian 09:01 scJohn: There are a few breaking changes with recent builds; one is that Win/win_common.h is now just "win_common.h"
scJohn 09:01 ok, just let me know and i will test
taxilian 09:01 it's just a few include paths changed is all
try that, let me know the next one you have problems with
I posted them to the mailing list
scJohn 09:01 ok, i had to copy win_common and resources.h
i figured it was because i was using a newer version of firebreath with an old project
but i did run the preop2010 command.
taxilian 09:01 don't copy those
use them in the right place
scJohn 09:01 understand.. just needed it to compile
taxilian 09:01 "win_common.h" with no "Win/" and "global/resources.h"
I *did* post those instructions… it isn't my fault… :-P ;-)
scJohn 09:01 i will read the docs
taxilian 09:01 it's on the mailing list
not posted to the docs yet, I'm afriad
scJohn 09:01 i was actually looking on the website just now
taxilian 09:01!msg/firebreath-dev/BHhtdwzq860/v5PoM2kX4qcJ
taxilian 09:01 if anyone in here has not filled out the survey, could you do so?
taxilian 10:01 jtojanen: this is the weirdest back and forth IRC conversation I've ever had, but thanks for the tip. it'll be in source control when you get back
FB_GitHubBot 10:01 FireBreath: master Richard Bateman * e721def (2 files in 1 dirs): Applied patch from jtojanen to improve registration in < vs2005
FireBreath: master Richard Bateman * 8c7b211 (3 files in 2 dirs): Finished patch to axutil.cpp, fixed two warnings
FireBreath: master Richard Bateman * 3ae0573 (1 files in 1 dirs): Fixed minor memory allocation bug with IE exceptions
FireBreath: master commits af12d3d...3ae0573 -
FireBreath: firebreath-1.4 Richard Bateman * af12d3d (24 files in 6 dirs): Removed the reliance on user-supplied param names, ...
FireBreath: firebreath-1.4 Richard Bateman * e721def (2 files in 1 dirs): Applied patch from jtojanen to improve registration in < vs2005
FireBreath: firebreath-1.4 Richard Bateman * 8c7b211 (3 files in 2 dirs): Finished patch to axutil.cpp, fixed two warnings
FireBreath: firebreath-1.4 Richard Bateman * 3ae0573 (1 files in 1 dirs): Fixed minor memory allocation bug with IE exceptions
FireBreath: firebreath-1.4 commits 0c90a69...3ae0573 -
kylehuff 10:01 taxilian: are you communicating via the logs? lol.. I'm confused..
taxilian 10:01 I said it was a weird conversation, didn't I? yes, he checks the logs :-P
he's in finland, so we're not always here at the same time
kylehuff 10:01 that is awesome..
jtojanen: you are my hero..
taxilian 10:01 yeah; one of these days, I'm going to write a decent IRC log parser/viewer tool for
with full text search and the ability to save conversation strings under a tag
to supplement the docs
or maybe iaincollins will
whichever of us finally finds time soonest =]
so thanks to jtojanen, in the latest firebreath 1.4 (that I just pushed) that will be alpha 3 you no longer have to add param names to that stupid array in order for them to be useable
kylehuff 10:01 so, I just did a pull of main, which I have not done in a few weeks. have there been breaking changes? (specifically in X11/np_x11main.cpp ?)
taxilian 10:01 yes
some paths have changed!msg/firebreath-dev/BHhtdwzq860/v5PoM2kX4qcJ
I did some major project restructuring with a very positive result overall (I think) and that's the result =]
kylehuff 10:01 I thought I had seen a list somewhere! thanks taxilian.
taxilian 10:01 well, I just looked it up for scJohn an hour ago, so I had it onhand already =]
probably just before you came in
kylehuff 10:01 figures
taxilian 10:01 no worries; I'm excited that people are starting to use the latest. I really need to know that it's solid before I push it live
I think it's pretty close, but it's only been really tested by me so far
kylehuff 10:01 yeah, I try to keep it fresh, but I since I have been doing major rework myself, I have been a little behind..
(which is better than being a big behind, nobody likes an ass..)
scJohn 10:01 taxilian: fyi I just built a new project ( prep2010, etc) in Express.. the compile finished, however I get an error on the installer.
taxilian 10:01 hmm. yeah, the installer won't currently work in express
the best idea I've heard so far to fix that is to make it check to see if heat.exe is what is calling and never redirect the registry in that case :-/
which actually might not be a bad idea
I just don't know how to do it offhand, so I'll have to take the time to figure it out sometime...
unless someone else wants to?
probably not hard
scJohn 10:01 figured it was express related. i may look at it if i get time.
taxilian 10:01 well, technically it's ATL related
vs2005 has the same problem
scJohn 10:01 actually the buld failed on the installer
taxilian 10:01 what was the error?
the older ATL didn't allow you to do per-user registration, so we have to do some funny hacks to make it work
but those hacks defeat heat.exe's detection of what registry changes should be made to install the COM component
so therefore the install generator fails
scJohn 10:01 error MSB6006: "cmd.exe" exited with code -2146232576
and while i have you i got this error on a different computer when I was trying to run the plugin.. "Assertion Failed" shared_ptr.hpp "Expression: px != 0"
i have not searched to much for that error message
taxilian 10:01 that means that somewhere you are (or FireBreath is) trying to access a shared_ptr that is NULL
scJohn 10:01 thanks
taxilian 11:01 if you find it's FireBreath let me know; that could be an issue
my understanding is that dereferencing NULL pointers in C++ is considered bad practice...
scJohn 11:01 i am not aware of any shared pointers that I would be using.. and it is during the initial plugin configuration.
taxilian 11:01 are you using an JSAPIPtrs?
anything in firebreath that ends in Ptr is a shared_ptr
scJohn 11:01 no idea, don't think so.. especially at this point, I am just starting up.. I guess it could be when I am checking device connectivity
taxilian 11:01 another possibility is that you aren't correctly creating your JSAPI object
or something similar
what are you doing when that happens?
scJohn 11:01 it happens at page load, the most I am doing there is calling (via javascript) a function that returns a bool. No params
well a couple of functions that return bool..
it could be that old build
taxilian 11:01 hmm. interesting. what version?
scJohn 11:01 I am starting fron scratch with the latest build.. i'll let you know if i run into the same problems.
taxilian 11:01 ok
man.. it sure adds a lot of possibilities when we can suddenly get the DOM object for the plugin
you know, jshanab/jshanab_/jshanab_wcw2 sure makes it look like there are a lot of people in here… ;-)
jshanab_wcw2 11:01 LOL, phone, home desktop, and work. I do like that I have the whole conversation visible wherever i go
kylehuff 11:01 I still cant figure out this ScheduleOnMainThread thing... the docs say "host->ScheduleOnMainThread(obj, boost::bind(&ObjectType::method, obj, arg1, arg2));", where obj being "boost::shared_ptr<ObjectType> obj(get_object_sharedptr());" but what actually is obj? the method I am calling it from?
taxilian 11:01 hehe. you should look at setting up a ZNC server somewhere
that's what I do; I connect everywhere and all of them use a single nick =]
obj is the object that you are calling it on
kylehuff: so if you were calling it on your JSAPIPtr, you'd use host->ScheduleOnMainThread(ptr, boost::bind(MyAPIPtr::method, ptr, 1, 2, 3));
for an example, look at JSObject.cpp InvokeAsync
kylehuff 11:01 hmm, okay, I think I see what is going on..
FB_GitHubBot 16:01 FireBreath: master Richard Bateman * f0aeb30 (14 files in 6 dirs): Fixes for JSAPI<->Browser bridge, improvements in lifecycle ... -
FB_GitHubBot 16:01 FireBreath: firebreath-1.4 Richard Bateman * f0aeb30 (14 files in 6 dirs): Fixes for JSAPI<->Browser bridge, improvements in lifecycle ... -
taxilian 16:01 as of that fix that I just pushed you can now completely control the lifecycle of your JSAPI object; for example, if you need it to be destroyed before the end of your plugin's destructor. it does take a little planning, because it's still a shared_ptr, but it's now doable
don't have to wait for the browser to release all of the objects
testfire 19:01 Hi, I have installed firebreath plugin on windows. I am looking to see if I can enable logging and write logs to a file in local file system