Re: [photran] Global Variables

["Ralph Johnson" <johnson@xxxxxxxxxxx>]

> Not clear to me where (free) Photran is going and the speed with which it
> might get there…

I'm head of the group at UIUC that produced Photran.  I have an idea
where Photran is going, but I don't know how long it will take to get
there.  That is the way it is with
low-budget software development.

The hard-core development is being done by one person, Jeff Overbey.
He is focusing on program representation and program transformation.
He is not thinking much about better user interface or how the
debugger works.  Jeff is trying to do work that will lead to papers
and to a PhD thesis, and there is only so much that one person can do.

The debugger is a separate module from the rest of CDT/Photran. My
graduate students have all shyed away from it because it was not
related to their research.  There were three undergrad who worked on
it last year and discovered a number of ways to improve it.
Unfortunately, the work they did will need to be redone to work with
the latest version of the CDT.  Jeff is going to make a version of
Photran based on the older CDT that will include their work, so you
can see a better debugger.  But what we'd REALLY like is for someone
out there to take responsibility for the debugger and focus on making
it useful.  What we call "the debugger" is really just a Java shell
around gdb.  gdb does almost all the work, and it doesn't need to be
changed.  However, the Java shell needs to be changed a little, as the
student's project showed.

I am teaching a software engineering course this semester, and using
Eclipse and Photran in it.  There are going to be about a hundred
students, divided into groups of four or five, who will be adding
features to Photran.  I don't expect that many of the projects will be
good enough that we will want to take their code, but I could be
wrong.  In any case, a lot of students will learn the design of
Eclipse, CDT and Photran, so they will be able to do more serious work
on it.  I'll try to get some of them to work on it next semester, and
they will be more likely to make a contribution then.

One of the handicaps we face is that the people in my group are not
Fortran programmers.  So, we don't use Photran on a daily basis.  We
need to find out what needs to be improved if we are going to put our
effort in the right direction.

A lot of students in my class are looking for projects to improve the
GUI of Photran.  What are some things that could make it a much better
Fortran editor?

-Ralph Johnson