develooper Front page | perl.perl5.porters | Postings from August 2001

Re: Master's Thesis/Project

Thread Previous | Thread Next
From:
Hugo van der Sanden
Date:
August 21, 2001 07:10
Subject:
Re: Master's Thesis/Project
Message ID:
200108211401.f7LE10319997@crypt.compulink.co.uk
Gibbs Tanton - tgibbs <Tanton.Gibbs@acxiom.com> wrote:
:I'm working on a Master's in computer science and I thought there might be
:some good projects to be had from Perl.  Does anyone have any recomendations
:as to Perl needs that would make good master's projects or even a master's
:thesis?

A useful first step would be to look at 'man perltodo' (pod/perltodo.pod
in the source tree).

For a master's project, a likely area of interest would be optimisation -
the todo list mentions optimisations such as shortcircuiting assignment
of @_:
  sub foo {
    my($self, $arg, %rest) = @_;
    ...
  }
  foo($object, 'test', -cache => 1, -order => 'numeric');
  $object->foo('test', -cache => 1, -order => 'numeric');
.. which can perhaps set up the variables directly if we already know
which function will be called (and know that it hasn't changed).

Another area of particular interest to me is the optimisations
done for regular expression matching, currently implemented by the
horrendous re_intuit_start. It would be nice to have that rewritten
so that it was a) clear enough to be maintainable and b) flexible
enough to make it easy to add new optimisations.

I'd also like at some point to see a mechanism added that allows a
program to request more expensive optimisations, for the times you
expect a program to run for a long time. I'm sure there are things
we could optimise further if we were prepared to take the time to
do some substantial analysis of the optree.

There are doubtless many others; I hope this gives you a starting
point.

Hugo

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About