Re: inaccurate spec count when running test runner html from browser vs running command line via phantomjs

[Jasmine Hegman]

No, that seems very strange.

When I run htmlReporter via phantomJS all I am doing is pulling stuff from its DOM, so it would seem like it should be identical with the browser's results. If there were fewer specs, then it might be a funky non-browser-quirk from phantomJS, but if you are getting more specs, then I am not sure. I am not familiar with Chutzpah myself, and do not expect a test runner to cause such an issue, but perhaps it deserves closer inspection?

Sorry for not being any help really, just wanted to give you some feedback with my experiences. The only really weird issue I've run into is missing stack traces, but that was a long cause of PhantomJs <-> jasmine quirks, and it was fixed, but is now broken again I just realized *facepalm* but my spec count is still accurate.

Good luck in resolving your problem!

On Wednesday, April 10, 2013 5:50:04 PM UTC-7, d48 wrote:

When I run a testrunner html file using jasmine htmlReporter my final spec counts are higher (about 4 specs) then when I run that same test runner file via Chutzpah with phantomjs. Has anyone else ran into this

[d48]

Awesome, thanks for the reply. Still looking into it so need to verify which numbers are incorrect when there's multiple specs being ran. Will post what I find once I resolve