[photran-dev] Status Report from UIUC

[Jeffrey Overbey <subscribe-photran-dev@xxxxxxxxxxxx>]

I apologize for the project seeming "dead" recently.  It is not.  It's
just slowed down a bit.

Since September, Spiros took a job with a startup, and our grant from
IBM expired, so I have been working as a teaching assistant.  Our
undergrads (Cheah and Nicholas) also moved on to other projects.

However, we are actively seeking funding, and in the mean time, we have
several people at UIUC still working on Photran, although their work is
not as immediately visible as we might like.

- We have three undergrads (Ken, Feanil, and Shawn) working with Walt
Brainerd to make the CDT debugger more Fortran-friendly.  (GDB actually
has pretty good Fortran support built-in.)  They will graduate in May,
so that is a fairly fixed deadline.

- One graduate student (Stas) worked with me last semester to put
together the start of an Fortran indexer, mostly to support our
refactoring efforts.  It is not fast, and we are not using as much of
the CDT indexer as I would like, but our main goal was to put something
together quickly so that we can (finally) release our refactoring tool.

- I am still working on getting the Rename refactoring ready for
release.  We are using an interesting fusion of abstract and concrete
syntax as the internal representation; more recently, adding cross-file
binding support and module loading has been the focus of my efforts
(aside from fighting with Eclipse APIs as usual).  Now it's on to
testing.  I hope a sample Rename will be ready by the end of the month,
or shortly after... it depends on how many bugs I can find in it.

- Introduce Implicit None is also not far off, but the version we demoed
last year needs to be updated to use the new program representation.

- My next efforts will focus on C Preprocessor support in the
refactoring engine.  One of my colleagues wrote a very nice PhD thesis
on the topic; I plan to combine some of her ideas and a few of my own.

- For reasons described above, my work on "administrative" tasks and
coding outside the refactoring engine has basically gone to zero.  Since
my time is severely constrained, I can only afford to work on the pieces
that can contribute to research publications and eventually get me a
PhD.  I hope you understand.

- Feanil also hopes to work on some loop parallelization transformations
using OpenMP.  I just learned about that today.

Is anything new at LANL?  Intel?

We are all still excited to be working on the project, and I apologize
again for not sending this update sooner.

Jeff