search

browserify / crypto-browserify

partial implementation of node's `crypto` for the browser
MIT License
656 stars 199 forks source link

Browser tests #139

Closed calvinmetcalf closed 9 years ago

calvinmetcalf commented 9 years ago

at it again, inspired by crypto-browserify/createHash#8

dcousens commented 9 years ago

:+1:

dcousens commented 9 years ago

The browser tests still aren't passing on travis?

calvinmetcalf commented 9 years ago

Yeah not yet

On Sun, Oct 4, 2015, 2:35 AM Daniel Cousens notifications@github.com wrote:

The browser tests still aren't passing on travis?

— Reply to this email directly or view it on GitHub https://github.com/crypto-browserify/crypto-browserify/pull/139#issuecomment-145322574 .

calvinmetcalf commented 9 years ago

we apparently do not support opera because it fails to be able to generate random numbers, I believe they didn't support that api because the result of Math.random() was actually cryptographically secure there, not a big deal because it's just the earlier pre blink based operae.

calvinmetcalf commented 9 years ago

hm also ie9 fails due to the poor RNG, we may need to shim a non cryptographically secure one for the tests

calvinmetcalf commented 9 years ago

ok all pasing now except android (due to known bug) and safari due to...it being the new IE?

dcousens commented 9 years ago

Can't be merged unless tests are passing :S

calvinmetcalf commented 9 years ago

Oh I know

On Sun, Oct 4, 2015, 6:04 PM Daniel Cousens notifications@github.com wrote:

Can't be merged unless tests are passing :S

— Reply to this email directly or view it on GitHub https://github.com/crypto-browserify/crypto-browserify/pull/139#issuecomment-145392118 .

calvinmetcalf commented 9 years ago

getting weird flakiness on ipads...it's never the same version that fails

calvinmetcalf commented 9 years ago

that being said it's pretty consistently md5 tests that are failing

calvinmetcalf commented 9 years ago

its consistently the md5 test that the ipad (not iphone) is failing, so this could be real...

calvinmetcalf commented 9 years ago

I've been trying all day to reproduce on an ipad simulator (like in saucelabs) and can't making me suspect its a saucelabs issue

calvinmetcalf commented 9 years ago

Ping @dcousens on this I think we can merge without iPad s

dcousens commented 9 years ago

ACK on merge, but, if you could outline where you got up with the iPad testing, that would be useful for future.

calvinmetcalf commented 9 years ago

basically I set up the emulator with xcode on my mac and tested against all the failing version several times and was unable to reproduce the issue.

dcousens commented 9 years ago

:+1: cheers