[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index][Thread Index][Top&Search][Original]

Re: v5.5.640 plans



From: Gurusamy Sarathy <gsar@ActiveState.com>
>FYI, I've dealt with most of the patches that came in over the last few
>weeks.
>
>I'm hotly pursuing the utf8 trail right now, and expect to have v5.5.640
>out some time this weekend.
>
>Meanwhile, the following todo items are still sorely in need of volunteers:
>
[snip]
>
>   * round out Compiler support additions (save END blocks; write a small
>     testsuite and driver for it; investigate bugs reported so far)

I've been toying with a module to give perl-level access to PL_beginav, 
PL_initav, PL_soptav, and PL_endav for an idea I had on how to trim the size 
of bytecompiled programs (perl -MO=Bytecode -e 'print "Greetings, 
mortals!\n";' is ~600K with 5.5.63). I've been seeing if converting 'use's 
to INIT blocks when under PL_minus_c would do it, and it seems to. If I can 
get it to work, I'll send in a patch. Oh, also - would it be ok for me to 
turn PL_minus_c into a set of flags (PERL_COMPILEONLY | PERL_BYTECOMPILE, 
for instance?) so that different backends can enable different changes to 
the compilation process? (ie, changing 'use's to INIT blocks makes virtually 
no sense whatsoever for B::CC)

>
>   * round out documentation
>      - perldelta upkeep (see TODO markers)
>      - ensure new stuff in perldelta is documented elsewhere
>      - rework API listing to reflect the current canonical form
>

One thing: I've been musing over the API, and the fact that some of it seems 
(to me) to be things that noone really has any business using 
(Perl_boot_core_UNIVERSAL, Perl_find_script, most of the Perl_do_*'s, etc.) 
Would it be acceptable to remove them from perlapi.c and global.sym, 
declaring them to be internal to perl and off-limits to anyone else?

-- BKS
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com


Follow-Ups from:
Gurusamy Sarathy <gsar@activestate.com>
Ilya Zakharevich <ilya@math.ohio-state.edu>

[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index][Thread Index][Top&Search][Original]