mozilla :: #devtools

17 Mar 2017
03:35birtleshow do I get the console to show reflows?
03:35birtlesIt used to be a setting in the "CSS" dropdown but that dropdown seems to have disappeared
03:38birtlesoh, I see, I need to turn off the new console
08:41rickychienHonza, gasolin:
08:42rickychienhttps://irccloud.mozilla.com/file/69ffowVW/netmonitor%20browser%20tab%20experiment.png
08:42Honzarickychien: awesome!
08:43pbrobirtles: yeah, the new console doesn't have that. I think the plan was to get rid of this feature since the perf tool shows reflows (among other things) already
08:43pbrowas that very useful to you?
08:44Honzarickychien: Ill put that screenshot on my slides ;-)
08:45rickychienHonza: Cool.
09:55quagmireHello, I have a question. Is there an option to launch Scratchpad from a desktop shortcut? (Let say, if Firefox isn't running, the option I want should launch only Scratchpad window, possibly not related to any profile).
09:55quagmireThere's an option in Total Commander to rename multiple files using any external editor. I usually use Notepad++, but almost always end up editing file names in Scratchpad, because it works better with multiple selection. This should explain my question.
10:13pbroquagmire: no sorry, scratchpad cannot be opened without firefox running first, or from a desktop shortcut. Nice to hear that you prefer scratchpad as an editor over notepad++ though :) it sounds like you might want to look at other text editors though, like SublimeText, Atom, VSCode, ... these support multiple selection too
10:20quagmireThanks for your response
10:24ochameauquagmire: there is that 'firefox -chrome chrome://devtools/content/scratchpad/scratchpad.xul' but use that at your own risks. it may not work for real and break at anytime!
10:24pbronchevobbe: ping
10:25nchevobbepbro: pong
10:25pbroFor ElementNode Reps, is MODE.TINY the right mode for having tag#id.class, right?
10:27nchevobbeyes
10:31pbrothanks Nico
10:32nchevobbenp, i'll watch your review soon
10:33nchevobbeafter I figure out how to fix this dumb framework broken because of touch event activation in firefox 52 :(
10:36pbroahah, sure, no pressure, this is a very small fix that can wait!
12:29solejlast: working on the up for grabs -> available issue
12:29soleshould be ready soon
12:30jlastthanks
12:31solejlast: I love remark! caught a broken link :D :D
12:31solesuper thumbs up!
12:32jlastOh great!
13:22soleAhhh this available thing makes so much sense, jlast!
13:22jlasthaha
13:22soleThe more instances I replace the more sense it makes
13:22soleI love it
13:27jlastloving the PR as well. We should promote about the change
13:27jlast*promote the change
13:27jlastjryans: do you mind if i migrate some tests to source-map
13:30mkaplyWith the new architecture of devtools being completely in browser, is there any way to integrate it into another XUL Application? It seems like you can't even bring up JS console anymore without bringing along all of devtools somehow?
13:32solemkaply: I don't know specifics, but I'd adventure the JS console has been integrated on the whole of devtools for a long time already
13:34mkaplysole: Well, last time I used it separate was ESR 45, so 52 is a whole new beast. Looks like Thunderbird integrated somehow, so I'll see if I can figure out what they did.
13:35mkaplyKind of hard to debug why it's not loading when you don't have a console though :)
13:38solemkaply: do you mean the 'web console' or the 'browser console'?
13:39soleI just checked in Thunderbird (stable) and it has a "error console" option with the same shortcut as the "browser console" shortcut in Firefox
13:43mkaplysole: Browser console. Yeah, thunderbird stable is still 45. Their changes to integrate the new devtools are more recent.
13:44solemkaply: I think jryans might be best suited to answer these questions - unfortunately they elude my current knowledge so far, but I'm glad we figured out which console we were referring to :D
14:05solejdescottes: Ta-da! https://github.com/devtools-html/reps/pull/116
14:05jdescottesgreat! thanks sole
14:06solejust in time to leave haha
14:07cbauercan I throttle web traffic to a web site (to be more precise in this example an upload to web version of onedrive)? "responsive design mode" can throttle but that also emulates a higher responsive time, I only want to limit bandwith
14:08solecbauer: not at the moment but if you have access to the trickle app or some similar bandwidth shaper you could do that from the command line http://ubuntuportal.com/2012/01/how-i-manage-bandwidth-using-trickle-on-ubuntu.html
14:09soleit sits between an app and the network and slows it down
14:09cbauerI'm on windows currently
14:09cbauerand optimally I wouldn't want to throttle the whole browser
14:09solecbauer: well that's why I say some other bandwidth shaper :)
14:10solecbauer: if you're feeling brave maybe you can attempt to install it via bash in windows
14:10soleit is a sort of ubuntu environment inside windows
14:10cbaueralready thought about it but WSL might be too much overhead
14:10cbauerI know about it
14:10soleOK - worth experimenting though!
14:10cbauerand it's called wsl (windows subsystem for linux) officially btw
14:11soleyeah whatever, it's bash on the insides :P
14:11cbauerthe start menu entry is named "bash on ubuntu on windows"
14:11cbauerthanks for the suggestions though
14:14cbauerhow can I see current transfer rate (also specificly for that tab and not the whole browser)?
15:23jlastHey jryans
15:24jryansjlast: hello!
15:24jlastI added a unit test for source maps!
15:24jlasthttps://github.com/devtools-html/devtools-core/pull/255
15:24jlastwhat do you think?
15:25jryansjlast: yes, i'll take a look soon! i am watching the devtools-core repo now
16:29jryansochameau: toolbox inside about:debugging really feels like a time machine back to app manager ;)
16:47ochameaujryans: yep that's exactly what I told jdescottes who didn't knew about app-manager!!
16:47jryansochameau: we should show some screenshots ;)
16:56ochameaumdn even shows we want from toolbox in a window to toolbox in app-manager document:
16:56ochameauhttps://developer.mozilla.org/en-US/docs/Archive/B2G_OS/Using_the_App_Manager#Debugging
16:56ochameaus/want/went/
16:59julienwheh
17:30erahmfitzgen: Do you know if there's any further work planned for the devtools memory stuff or is that basically a finished project?
17:31fitzgenerahm: no idea, I'm not on devtools anymore
17:31fitzgengregtatum: ^??
17:31jryanserahm: i believe gregtatum is owning that now, though he's focused on perf tools lately.
17:31fitzgenthat was more ?s than I meant so maybe conveys the wrong thing
17:33erahmfitzgen: maybe the better question is do you feel like it was done or more MVP at the point you stopped working on it? I'm trying to do a write up of "Stuff memshrink was excited about and where it's at" (the memory tool being one fo the things 3 years ago)
17:34erahmgregtatum: The follow up being, do you have plans for any interesting changes?
17:34fitzgenerahm: I think it is a solid v1, but by no means feature complete.
17:34fitzgenerahm: big things left: see into the CC's heap and its nodes and edges
17:35jryanserahm: gregtatum is probably traveling today, should be more responsive next week
17:35fitzgenerahm: save contents of strings inline
17:35fitzgenerahm: automatic leak detection / diagnosis feature
17:36erahmjryans: thanks, sounds like an email is the way to go :)
17:36fitzgenerahm: "save contents of strings inline" into the heap snapshot, that is
17:36erahmfitzgen: oh man bridging the js/c++ gap would be great
17:36fitzgenerahm: it just requires someone who understands the CC macros... shouldn't actually be hard, the infra is all there and already generic
17:36fitzgenerahm: was designed to be generic for this purpose from the start
17:37* erahm adds to memshrink project backlog
17:38gregtatum fitzgen in paris right now, but it's on hold for the next few quarters, we're focusing on perf right now, but it's not being abandoned or anything
17:38gregtatumfitzgen: definitely need to keep the conversations going on the memory stuff
17:40fitzgengregtatum: for sure
18:06jlasthey jryans, the reason i did the worker refactor in the test is that i did not want to test the worker directly
18:07jryansjlast: right, which i understand. it's more the part about moving it to a different module. that's unrelated to testing, right?
18:07jlastahh - i think i can do that
18:11jlastjryans - pushed
18:25jlastupdated. heading out soon, so trying to get it out the door :P
20:56bkellythanks for the quick turn around on the about:debugging stuff... looks great! https://www.dropbox.com/s/y7a01187m01unpj/Screenshot%202017-03-17%2016.55.42.png?dl=0
20:57jryansah, looks good :)
21:53jryansfitzgen: WIP WIP WIP :P
21:55fitzgenjryans: ;)
21:55jryansfitzgen: anyway, sounds like exciting progress :D
21:59fitzgenjryans: yes! it is
21:59fitzgenjryans: although ofc thigns are breaking when rebasing on m-c
21:59jryansfitzgen: to be expected, i would guess...
18 Mar 2017
No messages
   
Last message: 40 days and 13 hours ago