|
Peter Marshall
@
hooraybuffer
Munich, Bavaria
|
|
Engineer on @chromedevtools (and occasionally @v8js)
|
|
|
56
Tweets
|
92
Following
|
675
Followers
|
| Tweets |
| Peter Marshall retweeted | ||
|
Mathias Bynens
@mathias
|
Jan 28 |
|
Puppeteer v2.1.0 is out! π₯
β‘οΈ Chromium 80
β‘οΈ puppeteer.launch({product: 'firefox'})
Puppeteer can now talk to a real, unpatched Firefox binary. You no longer need to install puppeteer-firefox separately!
github.com/puppeteer/pupp⦠pic.twitter.com/Kw5UZ35gHo
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Jan 24 |
|
If you google image search 'Mathias' the first image is @mathias. I'm shook
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Jan 17 |
|
Ummmmmmmmmmm oops. We'll fix it on Monday π€¦ββοΈ
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Nov 27 |
|
/ducks cos I don't know. @leszekswirski you know stuff
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Nov 10 |
|
Snippets are also run in the new REPL-mode so this works for them too
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Oct 14 |
|
Time spent in `do` is probably time attributable to the loop itself like the back-jump at the end, condition check etc. Maybe that would be better attributed to the line with `while`. Or it's a bug in how we attribute things (e.g. line 426 looks unlikely).
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Oct 4 |
|
Check out what our interns worked on the last few months twitter.com/v8js/status/11β¦
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 28 |
|
Come to Munich it's 28 degrees at my desk all year. Especially in winter
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Eventually a spec will be the way to go, yes. We can experiment in the meantime π
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Refreshing the page will do it. There's no way to clear the Context/s we've created. It gets awkward if a page script closes over e.g. a let introduced by a REPL script
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
I think (but can't find the code to prove it) that DevTools only installs the debug function on the global object once you open it. So a <script> run after the console has been opened would see window.debug. Haven't confirmed that though
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Not sure what you mean - this still prints the debug object:
"use strict"; (0, eval)('console.log(debug)')
Or do you mean something else?
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Does writing 'use strict' change the toggle?
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Cool idea. We would need to figure out what happens if you select e.g. sloppy and then write 'use strict' - does it change the dropdown permanently or only for that input? And other interactions like that.
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Module by default sounds cool, but stops users doing
x = 7
which is a bit annoying if you just want a variable.
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Yup
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
There's no way to do that. You could use an eval :). Or we could add something (button or 'use sloppy'). You'd be stuck. But this is the same if you put 3 statements in the REPL vs. those 3 statements together in one REPL input.
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 20 |
|
Accidentally invented a new feature. Should the JS console/REPL in DevTools stay in strict mode (for later evaluations) if you set it once? It doesn't currently :(
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Sep 12 |
|
I need to go to the doctor and explain that it hurts when I run but not when I walk but these are the same word in German and the interpretation is highly regional-dependent. This is why standards work is important πΆββοΈπββοΈ
|
||
|
|
||
|
Peter Marshall
@hooraybuffer
|
Jul 10 |
|
I doubled the max length once... Apparently that was not a sustainable solution π
|
||
|
|
||