On 2/17/14, 5:25 AM, Jan de Mooij wrote:
On Sat, Feb 15, 2014 at 3:36 AM, Chris Peterson <[email protected]> wrote:
Jan, any more news from the JSC developers about collaborating on Yarr?
In your original mail starting this thread, you said that fixing Yarr
ourselves or upstream would be hard without help from someone very familiar
with Yarr. Are the JSC developers actively maintaining Yarr now? Do you feel
they will commit new resources to fixing Yarr performance and crashes?
AFAIK, this is not a priority for them. When we talked to them they
were willing to discuss changes with us etc, but they had no plans to
improve Yarr themselves. Sorry, I forgot to reply to your previous
mail, I'll send you their contact info.
hi Jan, sorry for my late reply!
Reading the other YARR emails you forwarded, the JSC developers sound
upbeat but won't commit any engineering help themselves.
In your original email about replacing YARR, you pointed out that no one
at Mozilla is (currently) familiar enough with YARR to implement major
features like nested parenthesized groups. If the JSC developers are not
going to help, what do you think we should do?
Do you want to resurrect your YARR thread on js-engine.internals
newsgroup? The discussion ended with Luke suggesting:
- fork irregexp
- significantly refactor the code to use SM rooting, assembler,
Vector, LifoAlloc, etc APIs
- declare open season on stylistic refactorings to make irregexp match SM
Andreas and Brendan seemed to support the plan, but Brendan suggested we
contact Apple first ("one ping only"). Which brings us to the present
day. :)
chris
_______________________________________________
dev-tech-js-engine-internals mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-tech-js-engine-internals
