nvm, saw your notes, didn't realize 4.9 was that fast! -----Original Message----- From: Jason Taylor [mailto:ja...@dedoose.com] Sent: Wednesday, December 09, 2015 3:22 PM To: dev@flex.apache.org Subject: RE: Next Flex SDK release
Harbs, Flex SDK 13.0 was the last release where TLF worked correctly without major performance issues. I know because we are still stuck on that release due to this bug. -----Original Message----- From: Harbs [mailto:harbs.li...@gmail.com] Sent: Monday, December 07, 2015 7:06 AM To: dev@flex.apache.org Subject: Re: Next Flex SDK release It looks like I don't need 4.6. There's a 3.0 branch in the Git repo which seems to have ti working correctly. On Dec 7, 2015, at 3:19 PM, Harbs <harbs.li...@gmail.com> wrote: > OK. I'm working on this, and I just added a comment to the JIRA with my > preliminary observations. Kind of interesting... > > I'm going to need to find the source of the old TLF to compare. I don't want > to use 4.9.1 because that code just had a band-aid. I assume Adobe's 4.6 code > was working correctly. > > I'm downloading 4.6 now from Adobe's site. Hopefully I'll be able to > use and build that... > > If anyone has any thoughts to help me on this, I'd appreciate it! ;-) > > Harbs > > On Nov 18, 2015, at 8:43 AM, Harbs <harbs.li...@gmail.com> wrote: > >> You had created such a flag, but enabling it causes lots of RTEs with the >> current code. >> >> I'd rather find the underlying cause of the problem which seems to be way >> too much recursion. >> >> I will try to take another look at this issue next week. >> >> On Nov 18, 2015, at 7:53 AM, Alex Harui <aha...@adobe.com> wrote: >> >>> 2) TLF Performance >>> https://issues.apache.org/jira/browse/FLEX-34769 >>> >>> I'd like to get an update from Harbs. I haven't spent any serious >>> thinking on the issue, but my recollection is that there is some >>> snippet of code we could disable or enable with a flag so folks can >>> get old behavior back if they don't need whatever that new behavior >>> was meant to solve (which I think may have been related to table support). >> >
