IRC Log Viewer » #firebreath » 2015-03-21

IRC Nick Time (GMT-7) Message
kylehuff 10:03 there doesn't seem to be any NP_GetProductVendor, or any NP_GetValue cases such as "NPPVpluginProductVendorString". any ideas on how we can get this info on linux taxilian? (looking to populate vendor and product name in PluginInfo)
if not, it probably doesn't really matter. those will just be empty in PluginInfo on linux
taxilian 10:03 kylehuff: I don't know of any way to get that information on linux
kylehuff 10:03 it's all good. I don't think it is a big deal
it's all good. I don't think it is a big deal
can you see any issue with calling NP_GetValue from dlopen?
taxilian 14:03 I don't think so
kylehuff 14:03 taxilian: do you want the file structure of src/NativeMessageHost to match src/PluginAuto in regards to OS specific files?
i.e. src/NativeMessageHost/{Win,X11,Mac}/PluginLoader{Win,X11,Mac}.cpp
(also, I have NativeMessageHost building on linux, and I can communicate with it via the api-tests extension)
taxilian 15:03 hang on, I'm swamped righ tnow
hang on, I'm swamped righ tnow
taxilian 16:03 kylehuff: sorry, this has been nuts. yeah, we probably should use that structure. it's easier in cmake
nice!
kylehuff 16:03 good, 'cause I already did it! lol
whenever things quiet down, I've got a few questions about some problem areas re: building on linux
taxilian 16:03 =]
it may be a bit
this test session has been insane
31 people have tested so far, 3 of us running the tests
kylehuff 16:03 no problem. I have some workarounds in place for testing other things. whenever you have time
taxilian 17:03 sounds good
finally headed home
I expect to be on again tonight, but I dont' know if you'll still be around
taxilian 19:03 kylehuff: If by chance you're still up I am now at my computer for while and not crazy swamped
kylehuff 20:03 taxilian: okay. now I just need to remember what the items were.... =c )
ah, okay, I remember.
in MainLoop.cpp, there is `root["c"] = c;` and `root["n"] = i + 1;`, which g++ bombs out on, because "long unsigned int" and "size_t" is too ambiguous for Json::Value... as a workaround, I swapped it with `root["c"] = std::to_string(c);` so I can get past compilation, do you have any advice?
taxilian 20:03 yeah, change the type from size_t to whatever it should be for Json::Value
probably uint32_t
kylehuff 20:03 okay, I couldn't readily tell what they were being used for, so I didn't mess with them
taxilian 20:03 it's for message fragmentation
c = count of total, n = which one this is
so if c = 5, n = 1, n = 2, n = 3, n = 4, n = 5 from the other messages
kylehuff 20:03 gotcha
the other thing is, in fbjson.cpp, there is `throw std::bad_cast("Invalid JSON document");`, and as far as I can tell, the gcc implementation of std::bad_cast does not accept arguments.
taxilian 20:03 interesting
could probably just make it std::bad_cast I guess
or use bad_variant_cast