Another week, another summary, while we're getting unnoticeably older. (Yes, I just remarked that I started those summaries two years ago.)
David Favor and Rafael investigated the implementation of filetests operators on AIX (bug #30885), notably regarding access control lists and how AIX system calls deal with them.
http://groups.google.com/groups?selm=rt-3.0.11-30885-92880.10.8131991882473%40pe rl.org
Ovid asked whether there was a way, internally, to tell the difference
between a variable that holds the undef value, and a variable that
hasn't been initialized at all. Dave Mitchell says that the short answer
is no, but elaborates a bit.
http://groups.google.com/groups?selm=20040804222250.54875.qmail%40web60810.mail. yahoo.com
John Peacock sent the Final Version Object Core Patch. (Minus a small leak found by Steve Hay.)
Ken Williams initiated a debate on the meaning of the UNIVERSAL::VERSION()
method, when the version module has been loaded. It overloads
UNIVERSAL::VERSION() to return a version object. This annoys
Module::Build. John Peacock argues that $VERSION should not be compared as
a string. Fergal Daly disagrees, saying that $VERSION should not be
compared as a floating point number.
http://groups.google.com/groups?selm=3BA4E2DA-E680-11D8-A3F5-000A95BD9874%40math forum.org
Marcus Holland-Moritz released two beta versions of Devel::PPPort 3. Incidentally, we learn that he has 87 perls installed on his laptop for his testing needs. However, more tests are welcome.
http://groups.google.com/groups?selm=20040808194504.3338d959%40r2d2
Paul Marquess released DB_File 1.810, to solve Cygwin problems.
Rafael removed the ``Newline in left-justified string for sprintf'' warning.
Fergal Daly added to the diagnostics module the ability to display
stack traces on errors and warnings.
http://groups.google.com/groups?selm=20040803233309.GA239%40dyn.fergaldaly.com
Fergal, actually, is apparently writing a stack trace module. While doing so, he found and reported a bug concerning what happens when you tie %SIG : it doesn't work until you insert a value in it. (Bug #30926.)
http://groups.google.com/groups?selm=rt-3.0.11-30926-93061.2.56136439978242%40pe rl.org
Andy Lyttle reported that using array indices near 2**32 wraps arounds
and results in accesses to the first elements of the array. (Bug #30979.)
As he says, that's the sort of weird behavior we expect from C, not from
perl.
This summary was written by Rafael Garcia-Suarez. Weekly summaries are published on http://use.perl.org/ and posted on a mailing list, which subscription address is perl5-summary-subscribe@perl.org . Comments and corrections welcome.