Re: release preparation: odd file permissions, #!'s, ^M's, and README

2004-02-27 Thread Leopold Toetsch 
Mitchell N Charity [EMAIL PROTECTED] wrote:
  README says

[ snip ]

There was more outdatet stuff in there. Fixed.

 Mitchell

leo

Re: release preparation: odd file permissions, #!'s, ^M's, and README

2004-02-27 Thread Leopold Toetsch 
Mitchell N Charity [EMAIL PROTECTED] wrote:
 Some quick observations on parrot_2004-02-26_08.

  Files with ^M:
[ ... ]
./t/op/00ff-dos.t

Actually that one is failing on Win32!!!1 Is this caused by the test
system? Which line endings do we support?

  The file ./testyamd probably belongs in ./tools/dev/,
  not in the top level directory.

Moved.

  These files are still at 0.0.13:

Will be updated RSN :)

  README says
The list of targetted platforms can be found in docs/parrot.pod.
  Which has a =head2 Supported Platforms
  Neither mentions PLATFORMS.

Yep.

  README says
The full changelog is available as ChangeLog.
  But ChangeLog was last updated 2003-10-31.

There are very few entries in ChangeLog recently (mainly release change
summaries).
It looks like that it got updated automatically til 2002-03-19.
How was it done?

 Mitchell

Thanks for your summary,
leo

Re: release preparation: odd file permissions, #!'s, ^M's, and README

2004-02-27 Thread Goplat 
--- Leopold Toetsch [EMAIL PROTECTED] wrote:
 Mitchell N Charity [EMAIL PROTECTED] wrote:
  Some quick observations on parrot_2004-02-26_08.
 
   Files with ^M:
 [ ... ]
 ./t/op/00ff-dos.t
 
 Actually that one is failing on Win32!!!1 Is this caused by the test
 system? Which line endings do we support?

The win32 version of cvs converts \n into \r\n... so that files endings
become \r\r\n, which makes perl5 unable to find the end of heredocs.

__
Do you Yahoo!?
Get better spam protection with Yahoo! Mail.
http://antispam.yahoo.com/tools

release preparation: odd file permissions, #!'s, ^M's, and README

2004-02-26 Thread Mitchell N Charity 
Some quick observations on parrot_2004-02-26_08.

 Odd files to be executable:
   ./docs/dev/dod.dev
   ./docs/dev/rx.dev
   ./languages/tcl/MAINTAINER
   ./languages/tcl/examples/Makefile
   ./lib/Digest/Perl/MD5.pm


 A perl by any other name, may be a different perl.
 perl and /usr/bin/perl are both common in #!'s. 

 The oddball perl #!'s are:
#!/bin/perl -w : ./icu/source/tools/genpname/preparse.pl
#!env perl : ./languages/m4/Perl5/t/001_compile.t
#!env perl : ./languages/m4/tools/frozen2yaml.pl
#!/usr/bin/perl -w^M   : ./languages/converter.pl
#!/usr/lib/perl -p : ./icu/source/tools/genprops/misc/ucdstrip.pl
#!/usr/local/bin/perl -w   : ./lib/Digest/Perl/MD5.pm

 The only /usr/bin/perl #!'s outside of languages/ are in
   ./editor/ops2vim.pl
   ./icu/source/i18n/regexcst.pl
   ./icu/source/tools/genrb/genrbjar.pl
   ./icu/source/tools/genren/genren.pl
   ./icu/source/tools/gentz/tz.pl
   ./Configure.pl
   ./editor/kate_syntax.pl
   ./icu/source/tools/cpysearch/cpysearch.pl
   ./lib/Parrot/Key.pm
   ./lib/Parrot/PackFile/Constant.pm
   ./lib/Parrot/PackFile/ConstTable.pm
   ./lib/Parrot/PackFile/FixupTable.pm
   ./lib/Parrot/PackFile.pm
   ./lib/Parrot/String.pm
   ./types/bignum_atest.pl
   ./types/bignum_test.pl
   ./languages/conversion/test.pl
   ./languages/converter.pl

 I haven't looked at them all, but I note these two
   ./types/bignum_atest.pl
   ./types/bignum_test.pl
 do Inline C, which can be perl configuration dependent.

 ./Configure.pl starts with #!/usr/bin/perl, while most coreish files
 say just perl.  But it isn't executable, and is traditionally invoked
 as perl Configure.pl.  Perhaps the #! line should be deleted?

 Lots of languages/ files use /usr/bin/perl.  Perhaps it would be
 worth having a parrot policy that all perl #!s should simply be
 'perl -w', except when there is a good reason not to.


 Files with ^M:
   ./config/gen/platform/win32/exec.c
   ./examples/benchmarks/gc_alloc_new.pasm
   ./examples/benchmarks/gc_alloc_reuse.pasm
   ./examples/benchmarks/gc_generations.pasm
   ./examples/benchmarks/gc_header_new.pasm
   ./examples/benchmarks/gc_header_reuse.pasm
   ./examples/benchmarks/gc_waves_headers.pasm
   ./examples/benchmarks/gc_waves_sizeable_data.pasm
   ./examples/benchmarks/gc_waves_sizeable_headers.pasm
   ./languages/BASIC/interpreter/basic.pasm
   ./languages/BASIC/interpreter/README.BASIC
   ./languages/BASIC/interpreter/wumpus.bas
   ./languages/converter.pl
   ./languages/perl6/t/compiler/numbers.t
   ./languages/python/python.bnf
   ./t/op/00ff-dos.t


 The file ./testyamd probably belongs in ./tools/dev/,
 not in the top level directory.

 These files are still at 0.0.13:
   ./ChangeLog
   ./DEVELOPING
   ./NEWS
   ./README
   ./VERSION
   ./parrot.spec
   ./languages/perl6/perl6  :$VERSION = '0.0.13';
   ./imcc/main.c:#define IMCC_VERSION 0.0.13.0

 README says
   The list of targetted platforms can be found in docs/parrot.pod.
 Which has a =head2 Supported Platforms
 Neither mentions PLATFORMS.

 README says
   The full changelog is available as ChangeLog.
 But ChangeLog was last updated 2003-10-31.

 README says
   the computed-goto dispatch core (core_ops_cg.c)
 Does this file still exist?


Mitchell

Re: release preparation: odd file permissions, #!'s, ^M's, and README

2004-02-26 Thread Michael Scott 
On 26 Feb 2004, at 18:57, Mitchell N Charity wrote:

 A perl by any other name, may be a different perl.
 perl and /usr/bin/perl are both common in #!'s.
I been changing them to #! perl -w when i find them, which is why 
your list covers the places I haven't visited. The good thing about the 
bang line is that it's parsed by perl and thus turns on warnings. I'll 
use your list to standardize the rest.

Mike