Issue still exists using iMacros 7.0.3.0, for both:
- Firefox 4.0 beta 7
- Latest Minefield [Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0b8pre) Gecko/20101202 Firefox/4.0b8pre]
In case it helps, I've tracked down the regression ranges for this issue, using both the mozilla-central branch and the tracemonkey branch.
By cross-referencing between the two, it should be fairly easy to find the common changeset, which hopefully may speed up the diagnosis.
Mozilla-central:
- Last good nightly: 2010-10-31
- First bad nightly: 2010-11-01
- Pushlog:
http://hg.mozilla.org/mozilla-central/p ... ea6992e1c6
Tracemonkey:
- Last good nightly: 2010-10-30
- First bad nightly: 2010-10-31
- Pushlog:
http://hg.mozilla.org/tracemonkey/pushl ... f6608b2871
Also, given that I was under the impression that the new Jaegermonkey JS engine was responsible for this breakage, I was surprised to find that setting:
javascript.options.methodjit.content = false
...ie: turning off the new Jaegermonkey JS features, didn't solve the problem.
So therefore the breakage is either due to other unrelated JS engine optimisations - or else an unintentional change to non method-jit code, when Jaegermonkey was merged.
Hope that helps - and looking forwards to a fix so I can finally use iMacros in Minefield/the latest Fx4.0 betas again.
Thanks,
Ed