00:05:42  * xaxxonjoined
00:49:25  * xaxxonquit (Quit: xaxxon)
02:37:51  * zvquit (Ping timeout: 248 seconds)
03:05:53  * oleavrquit (Quit: Connection closed for inactivity)
03:18:14  * plutoniixjoined
03:19:23  * plutoniixquit (Max SendQ exceeded)
03:19:49  * plutoniixjoined
04:04:20  * xaxxonjoined
04:14:24  * Guest59quit (Quit: My Mac has gone to sleep. ZZZzzz…)
04:31:30  * bradleymeckquit (Quit: bradleymeck)
04:32:29  * xaxxonquit (Read error: Connection reset by peer)
04:44:55  * bradleymeckjoined
04:52:11  * zvjoined
05:03:16  * bradleymeckquit (Quit: bradleymeck)
05:13:23  * bradleymeckjoined
05:14:23  * zvquit (Ping timeout: 276 seconds)
05:17:48  * bradleymeckquit (Client Quit)
05:40:17  * xaxxonjoined
05:48:39  * zvjoined
06:20:58  * xaxxonquit (Ping timeout: 256 seconds)
07:04:34  * xaxxonjoined
07:45:36  * xaxxonquit (Quit: xaxxon)
09:00:01  * plutoniixquit (Quit: Leaving)
11:25:11  * mylesborinsquit (Quit: farewell for now)
11:25:42  * mylesborinsjoined
11:26:47  * AtumTjoined
15:19:22  * bradleymeckjoined
15:44:20  * bradleymeckquit (Quit: bradleymeck)
17:08:08  * bradleymeckjoined
17:08:40  * RT|Chatzillaquit (Quit: ChatZilla [Firefox])
17:14:35  * bradleymeckquit (Quit: bradleymeck)
17:18:04  * seventhjoined
17:30:34  <caitp>aklein: I don't think I'm going to find a simpler fix for the bug you CC'd me on... the simplest things are going to be either the rewindable parser, or the "get rid of PatternRewriter" things. so, I'm looking at seeing if parts of it are fixable with just the Destructuring Assignment portion of that CL
17:30:53  <caitp>but I don't think I am going to find a smaller/hackier fix for it
18:54:40  * bradleymeckjoined
18:59:10  * bradleymeckquit (Ping timeout: 256 seconds)
18:59:38  * seventhquit (Remote host closed the connection)
19:00:11  * bradleymeckjoined
21:24:00  <aklein>caitp: thanks for the heads-up. I may spend a bit of time this afternoon looking for a hack to fix it. but I do think that changing how we do destructuring assignment would be a nice fix for it
21:33:52  <bradleymeck>does HeapProfiler::FindObjectById return things that definitely should never actually hit userland code even though they pass local->IsObject() == true ?
21:36:57  <bradleymeck>" 2: formatPrimitive(aka formatPrimitive) [util.js:616] [bytecode=0x1cdbe036e261 offset=99](this=0x1cdbd4a822d1 <undefined>,fn=0x1cdb39383a41 <JSFunction stylizeNoColor (sfi = 0x1cdbe0343f79)>#1#,value=0x1cdbd4a82251 <Odd Oddball: uninitialized>)"
21:37:12  <bradleymeck>Odd Oddball: uninitialized seems very weird to me to get out of that API
21:42:52  <aklein>bradleymeck: that's in the object graph of the object you searched for, though, not the object itself, from what I can tell
21:43:10  <aklein>it's to be expected that some of the heap objects kept alive by a v8::Object might be internal things
21:43:15  <bradleymeck>aklein: correct, i only found this once I started to print the object
21:43:30  <bradleymeck>i'm unsure how to detect these oddballs that blow up if touched though
21:43:41  <bradleymeck>i can use C++ but don't see any obvious API
21:43:53  <aklein>how are you "touching" them?
21:44:08  <bradleymeck>util.format in node... sec
21:45:38  <bradleymeck>https://github.com/nodejs/node/blob/master/lib/util.js#L378 appears to be where it gets the angry thing , will gist a full trace
21:46:01  <bradleymeck>https://gist.github.com/bmeck/f8090346c760665b4916d436e523c3d2
21:47:30  <bradleymeck>no clue why that line is numbered that since it doesn't seem to match the fn call names
21:47:37  <bradleymeck>but... i'm pretty lost
21:49:32  <aklein>bradleymeck: I'm pretty lost too, I don't understand how you're feeding objects into that code
21:49:38  <aklein>I'm not very familiar with the heap profiler API
21:50:14  <bradleymeck>i'll keep poking around
21:51:17  <bradleymeck>i'm basically just having a huge loop of `1 ... 20_000` that I'm iterating on and trying to make sure my code doesn't blow up when the profiler doesn't have valid results for that API
21:52:02  <bradleymeck>~= for (var i = 0; i < 20000; i++) console.log(profiler.getObjectBySnapshotId(i))
22:33:35  <bradleymeck>aklein: yup, it got a hold of some Object with a JS property edge to a hidden `system / Oddball` node
22:34:00  <bradleymeck>no real way to test this that I can tell from JS/C++, it just looks like Undefined
22:34:23  <aklein>bradleymeck: I know of at least one place where we use a JSObject for an internal purpose
22:34:28  <aklein>so it's not inconceivable
22:35:13  <bradleymeck>aklein: it seems fine for now, they would have to just be guessing random Ids to get a hold of this
22:35:30  <bradleymeck>and since they already are dumping snapshots for my use case :shrug:
22:55:00  <cloudshu>is there a way to get prevent Chrome Infra Package Deployer from verifying certificates, or a way to pass it a cacert
22:55:16  <cloudshu>getting depot_tools behind a corporate proxy is unending torment
23:14:39  * bradleymeckquit (Quit: bradleymeck)
23:26:20  <aklein>cloudshu: dunno. you might have luck asking on #chromium; dpranke is who I'd bug :)
23:26:50  * RT|Chatzillajoined
23:29:07  <cloudshu>aklein: thanks
23:33:05  * AtumTquit (Remote host closed the connection)