incubator-flex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Roland Zwaga <>
Subject Re: Flex CodeScape ...
Date Mon, 01 Oct 2012 12:07:41 GMT
> Yeah took a look at the PDF, there are some things that need to be thought
> about. I'm not really at the point where I could even begin to think about
> this part.
> Hopefully others might take an interest when the framework is up and could
> help contribute then.
> BTW the "framework" here uses the parser to load source files and swcs,
> create the AST nodes and IDefinitions.
> Once the parsing is done a processor takes all the compilation units and
> sorts all member data, meta data into an accessor map API. From there
> processing is done to resolve inherited comments, accessor get-set pairs
> and various other things.
> At this point, anything is possible, using the accessor map dependencies
> are set and it's just a matter of finessing the API to create the output
> you want.
> IE With the first iteration here I have created the asdoc HTML API docs
> using Velocity templates. Works very fast. I am parsing and outputting the
> spark and framework html in 35 seconds.

Hey Mike,

do you think your framework could also form a basis for static analysis
tools such as FlexPMD/CPD et al? If the parsing is so much faster than
perhaps these existing
tools could be re-written to use the new compiler. I myself know absolutely
nothing about static code analysis, but perhaps folks on the list here do
and leveraging your
existing work could be beneficial.



  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message