00:20:30  * jugglinmikequit (Ping timeout: 272 seconds)
00:25:43  * ewanasjoined
00:41:32  * unixpicklequit (Quit: My Mac has gone to sleep. ZZZzzz…)
00:42:09  * ewanasquit (Ping timeout: 246 seconds)
01:14:44  <jwolfe>does anyone know what "harmony" means in v8? it looks like it's used to identify "Features that are still work in progress". so is that just an arbitrary name for that kind of thing?
01:25:06  <caitp>jwolfe: it used to mean stuff that was on the standards track in general, afaik? I think harmony was one of the nicknames for a future version of the spec
01:25:24  <jwolfe>caitp, ah. that makes sense. thanks.
01:28:32  * unixpicklejoined
01:41:16  * rhalffjoined
01:49:23  * bradleymeckquit (Quit: bradleymeck)
01:56:22  * unixpicklequit (Quit: My Mac has gone to sleep. ZZZzzz…)
02:18:56  * plutoniixquit (Quit: Leaving)
02:24:26  * Garbeequit (Quit: Connection closed for inactivity)
02:25:40  <jwolfe>in parser-base.h: ParseFormalParameterList(), there's a very pretty comment showing the grammar that the function is implementing. if i add trailing comma support behind a flag, what should i do about that grammar comment?
02:32:00  * rhalffquit (Ping timeout: 244 seconds)
03:37:06  * zvquit (Ping timeout: 276 seconds)
03:37:27  * rhalffjoined
03:55:54  * zvjoined
03:57:22  * plutoniixjoined
03:57:24  * plutoniixquit (Max SendQ exceeded)
03:57:51  * plutoniixjoined
03:58:57  * plutoniixquit (Max SendQ exceeded)
03:59:25  * plutoniixjoined
04:00:33  * plutoniixquit (Max SendQ exceeded)
04:01:01  * plutoniixjoined
04:02:03  * plutoniixquit (Max SendQ exceeded)
04:02:31  * plutoniixjoined
04:03:28  * plutoniixquit (Max SendQ exceeded)
04:03:56  * plutoniixjoined
04:04:53  * plutoniixquit (Max SendQ exceeded)
04:05:21  * plutoniixjoined
04:40:05  <trungl-bot>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap%20-%20debug/builds/7474 "V8 Linux - nosnap - debug" from f6facbb2106ffc2918dd249166233a7fa95dc449: [email protected])
05:34:41  <trungl-bot>Tree opened by [email protected]: Tree is open
05:44:15  * thefourtheyejoined
06:04:42  * zvquit (Ping timeout: 244 seconds)
06:23:33  * zvjoined
06:39:28  * BobGneujoined
06:39:59  * BobGneuquit (Read error: Connection reset by peer)
06:40:36  * BobGneujoined
06:49:44  * _Gettyjoined
06:52:03  * BobGneuquit (Read error: Connection reset by peer)
06:55:07  * APKjoined
07:00:14  * saurik_joined
07:04:45  * AKPWDquit (*.net *.split)
07:04:45  * Gettyquit (*.net *.split)
07:04:46  * saurikquit (*.net *.split)
07:06:48  * Net147_quit (Ping timeout: 246 seconds)
07:08:17  * Net147joined
08:02:38  * mostynbjoined
08:21:44  * ewanasjoined
08:55:05  * plutoniixquit (Quit: Leaving)
08:58:48  * plutoniixjoined
09:06:00  * rhalffquit (Ping timeout: 276 seconds)
09:21:27  * rhalffjoined
09:22:32  * mostynbquit (Quit: Leaving)
09:30:16  * davijoined
09:44:16  * daviquit (Ping timeout: 252 seconds)
10:03:24  * rhalffquit (Ping timeout: 244 seconds)
10:28:00  * rhalffjoined
10:47:56  * ghjfhfjoined
10:48:30  * ghjfhfquit (Client Quit)
10:55:10  * plutoniixquit (Quit: จรลี จรลา)
11:20:12  * rhalffquit (Ping timeout: 260 seconds)
11:22:51  * rhalffjoined
12:04:15  * bobmcwjoined
12:16:46  * Garbeejoined
12:23:13  * rhalffquit (Ping timeout: 252 seconds)
12:32:41  * bradleymeckjoined
12:41:48  * Venemojoined
12:44:58  * bradleymeckquit (Quit: bradleymeck)
12:50:22  * bradleymeckjoined
13:10:36  * unixpicklejoined
13:11:29  * rhalffjoined
13:23:01  * jugglinmikejoined
13:33:22  <caitp>jwolfe: well, I guess you'd update it
13:33:36  <caitp>it's tricky to do that with the 80c limit :(
13:56:33  * plutoniixjoined
14:02:37  * unixpicklequit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:05:34  * bobmcwquit (Read error: Connection reset by peer)
14:05:56  * bobmcwjoined
14:06:51  * Venemoquit (Ping timeout: 240 seconds)
14:28:53  * unixpicklejoined
15:22:36  * seventhjoined
15:27:46  <trungl-bot>Tree closed by [email protected]: Tree is closed (Automatic: "compile" on http://build.chromium.org/p/client.v8/builders/V8%20Win64%20-%20debug/builds/10560 "V8 Win64 - debug" from 815da79694218031e8219a4bef037e9499cffe2a: [email protected])
15:50:04  * rhalffquit (Ping timeout: 260 seconds)
15:55:24  * rhalffjoined
16:04:28  * davijoined
16:04:28  * daviquit (Changing host)
16:04:28  * davijoined
16:06:06  <trungl-bot>Tree opened by [email protected]: Open
16:07:32  * Venemojoined
16:13:52  * RT|Chatzillaquit (Quit: ChatZilla 0.9.86.1 [Firefox 2.0.0.22pre/2010030309])
16:26:31  * daviquit (Ping timeout: 240 seconds)
16:44:04  * ewanasquit (Ping timeout: 272 seconds)
16:49:40  * unixpicklequit (Quit: My Mac has gone to sleep. ZZZzzz…)
17:10:41  * thefourtheyequit (Quit: Connection closed for inactivity)
17:16:21  * rhalffquit (Remote host closed the connection)
17:18:44  <jwolfe>caitp, does this grammar seem right to you: https://jeffmo.github.io/es-trailing-function-commas/ ? it seems like it's wrong to me. there's even a TODO in there that i don't understand at all.
17:19:02  <jwolfe>so my question now is, if i'm supposed to put grammar in the comments, what grammar do i put? do i make something up that i think is reasonable?
17:19:47  <caitp>seems wrong how?
17:20:06  <jwolfe>i'd kinda like to submit a WIP CL to get review feedback. do you think it's too soon to do that? i believe all the code and tests are working properly, but maybe not the comments.
17:20:30  <caitp>go ahead and submit it
17:23:54  <caitp>the Modified Productions bit looks fine to me
17:24:20  <jwolfe>do you know why it removes FormalsList?
17:25:22  * unixpicklejoined
17:40:31  * davijoined
17:50:59  * daviquit (Ping timeout: 244 seconds)
17:51:07  <caitp>jwolfe: it just refactors it really
17:52:34  <caitp>FormalParameters seems to replace FormalParameterList, which itself replaces FormalsList
18:01:40  * davijoined
18:05:19  <jwolfe>here's my CL WIP: https://codereview.chromium.org/2094463002
18:05:29  <jwolfe>i haven't touched the comments yet.
18:07:53  * hferreirojoined
18:14:50  * hferreiro_joined
18:16:42  <jwolfe>is there a way to stop/cancel a commit bot in progress?
18:17:54  <jwolfe>i want to add extra trybot jobs, but i've already started the commit process. it's not too big of a deal if it's not possible.
18:19:16  * ewanasjoined
18:20:01  * hferreiroquit (Remote host closed the connection)
18:20:07  * hferreiro_changed nick to hferreiro
18:20:42  <aklein>jwolfe: nope, trybots can't be canceled. I'd say don't worry about it :)
18:21:09  <aklein>caitp: if you wouldn't mind doing a first pass on jwolfe's patch that'd be helpful, I'm firefighting a crash today :(
18:21:28  <caitp>sgtm
18:21:28  <jwolfe>aklein, ok thanks. yeah, i'm not worried too much, but i obviously would have wanted to put the blink tests in that patch.
18:24:26  * hferreiroquit (Quit: hferreiro)
18:26:24  * hferreirojoined
18:26:55  * hferreiroquit (Client Quit)
18:27:12  * hferreirojoined
18:41:51  <jwolfe>(erm, wait. that's already a chromium patch. i'm not worrying correctly >.< )
18:42:37  <caitp>remember to relax =)
18:45:45  * daviquit (Ping timeout: 246 seconds)
18:47:20  * davijoined
18:57:51  * daviquit (Ping timeout: 240 seconds)
18:58:26  * hferreiroquit (Quit: hferreiro)
19:52:56  <trungl-bot>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap/builds/9423 "V8 Linux - nosnap" from 0e14baf712955a1993f742647bb2adc293702b80: [email protected])
19:54:57  <trungl-bot>Tree opened by [email protected]: Tree is open (attempt retry for apparent infra issue)
19:55:58  <trungl-bot>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap%20-%20debug/builds/7497 "V8 Linux - nosnap - debug" from a53b9bf02f31e5647c37e0392afa19f74df1a3ba: [email protected],[email protected])
20:03:01  <trungl-bot>Tree opened by [email protected]: open (reverted patch)
20:07:04  <trungl-bot>Tree closed by [email protected]: Tree is closed (Automatic: "Check" on http://build.chromium.org/p/client.v8/builders/V8%20Linux64%20-%20debug/builds/10045 "V8 Linux64 - debug" from 8d830a5aab1c688dc0c3a1690e2ceab145b23bc3: [email protected],[email protected])
20:12:57  * seventhquit (Remote host closed the connection)
20:17:05  * Venemoquit (Quit: Communi 3.5.0 - http://communi.github.com)
20:19:24  * jugglinmikequit (Ping timeout: 272 seconds)
20:21:36  * jugglinmikejoined
20:22:34  * zvquit (Ping timeout: 272 seconds)
20:46:39  * zvjoined
20:48:24  <trungl-bot>Tree opened by [email protected]: Tree is open (infra outage appears to be over)
21:21:43  <jwolfe>caitp, regarding https://codereview.chromium.org/2094463002/diff/1/src/parsing/parser.cc#newcode820 , why doesn't that omission cause test failures? i thought i was testing the preparser by using RunParserSyncTest().
21:24:02  <jwolfe>i found the code you're talking about, but i can't get it to matter.
21:24:23  <jwolfe>could it be that the preparser doesn't ever check for that kind of error, so there's no need to tell it to allow it sometimes?
21:36:10  <caitp>jwolfe: i haven't looked at the test yet, gimme a sec
21:39:17  <caitp>jwolfe: it's because of how the preparser tests are run
21:39:40  <caitp>they take the Parser constructor path, rather than the lazy path
21:39:53  <caitp>which is arguably a bug in the testing infrastructure, although it does catch some things
21:48:40  <jwolfe>caitp, i can't seem to reproduce the problem where the preparser doesn't know about the flag. all my attempts indicate that i don't actually need that code, which doesn't seem right to me.
21:48:54  <jwolfe>my most recent attempt was: ./out/x64.debug/d8 --harmony_trailing_commas_in_parameters -e 'function a() { print(1,); } a();'
21:49:08  <caitp>you definitely need the code
21:49:30  <caitp>try adding --min-preparse-length=0
21:49:58  <caitp>or alternatively, adding a bunch of comments to the function body
21:50:19  * bobmcwquit (Remote host closed the connection)
21:50:39  <jwolfe>thanks, that was it.
21:50:43  <jwolfe>ok
21:51:00  <jwolfe>i didn't even know about a min preparse length.
21:51:21  <caitp>yeah
21:51:46  <caitp>I think we need to fix test-parsing.cc so that it also tests that lazy parsing path
21:51:57  <caitp>aklein / littledan__: ^
21:54:09  <littledan__>caitp, sounds like a good idea (I'm a little surprised we don't already)
21:54:32  <littledan__>test-parsing.cc already runs the code in various modes; I think it'd be reasonable to blanket-run it with various modes like that
21:54:45  <littledan__>we already test both the parser and preparser
21:55:36  * APKchanged nick to AKPWD
21:55:51  <aklein>what littledan__ said, I'm not sure there's anything to add to test-parsing.cc for this
21:56:30  <aklein>caitp: is there an error that only crops up with full parsing on the outside of a function and pre-parsing on the innards? I can't think of what that would be.
22:05:18  * plutoniixquit (Quit: Leaving)
22:25:43  * RT|Chatzillajoined
22:35:02  <jwolfe>test-parsing did not catch the bug that patch set 2 fixes here: https://codereview.chromium.org/2094463002/
22:35:17  <jwolfe>flags were not being forwarded to the pre parser.
22:36:59  <jwolfe>maybe the issue is that the code that forwards flags to the pre parser isn't run in the parser tests. meaning Parser::ParseLazyFunctionBodyWithPreParser() isn't tested.
22:37:26  <jwolfe>oh, maybe i just need to add context tests that surround the code in a function? trying now...
22:38:28  * unixpicklequit (Quit: My Mac has gone to sleep. ZZZzzz…)
22:40:56  <jwolfe>...no that didn't seem to do it.
22:43:55  <jwolfe>ok i got it. i needed to include {"function() {", "}"}, in my context_data. then it catches the error.
22:47:46  <jwolfe>erm, i mean {"function foo() {", "}"},
22:52:26  <jwolfe>ok, next question: why is there a "GeneratorParameter" grammar parameter in the code comments? i don't see it in the current spec or the jeffmo proposal. should i delete it?
22:57:28  * unixpicklejoined
22:59:15  <caitp>jwolfe: looks like that was Andy's invention
23:19:21  * bobmcwjoined
23:24:52  <jugglinmike>jwolfe: caitp https://twitter.com/abargull/status/603258950163763202
23:25:11  <jugglinmike>That was inches from being in es2015
23:25:45  <caitp>interesting
23:27:32  * bobmcwquit (Remote host closed the connection)
23:34:33  * jugglinmikequit (Ping timeout: 240 seconds)
23:40:04  <jwolfe>should like i should delete it.
23:46:38  <jwolfe>*sounds like i should delete it.
23:54:39  <caitp>jwolfe: i mean on the one hand yes, on the other its unrelated to your patch
23:54:53  <caitp>messes up blame a bit
23:55:19  <jwolfe>i mean, i'm already going to be editing this one comment. i don't think it's out of scope in that sense.
23:55:36  <jwolfe>there's another comment that mentions GeneratorParameter that I could leave alone
23:55:49  <jwolfe>or i could delete it while i'm in here.
23:57:05  <jwolfe>to me this is a matter of choosing between bad code or bad history. i think i would choose bad history every time, but of course that's just my analysis of the situation.