Quoting Alex Harui <aha...@adobe.com>:
On 12/4/12 9:53 AM, "Michael Schmalle" <apa...@teotigraphix.com> wrote:
Quoting Alex Harui <aha...@adobe.com>:
On 12/4/12 2:24 AM, "Michael Schmalle" <apa...@teotigraphix.com> wrote:
What makes that code REALLY confusing is although you say he is using
AS, there is a huge amount of ABC opcode in that code he wrote.
Are you talking about JSEmmitter?
I'm talking about the classes in general.
You know AST better than me, but it seems like JSGeneratingReducer has an
AST API and doesn't require knowing about ABC or SWF?
My question is, who is going
to maintain this code base in a year? Adobe, our committers that
totally understand all the classes and their implementation?
Adobe won't be maintaining this code base. Sure they pay me, but really,
I'm just another community member regarding FalconJS. I'm learning the code
base just like you.
I haven't really thought about this, but other than bug fixes, unless we try
to evolve the language, what other changes would need to be done (assuming
performance is acceptable)?
Ok, I get what you are saying but that is like buying a car with no
mechanic within 3000 miles. The way that code is in the project, it
feels like a prototype implementation. There are notes and todos
everywhere.
But I am probably just over thinking... I just have a hard time
counting on things I didn't make or can't even talk to the people that
made it to ask questions about why and how they did things.
At least in most opensource projects you can at least look at the
development history.
By getting MXML to output data instead of code, there is a lot lower chance
we'll need to change the compiler going forward.
And in no way shape or form am I saying don't use what you already
have. I have my own reservations and those are just my opinions.
Mike
--
Alex Harui
Flex SDK Team
Adobe Systems, Inc.
http://blogs.adobe.com/aharui
--
Michael Schmalle - Teoti Graphix, LLC
http://www.teotigraphix.com
http://blog.teotigraphix.com
