In perl.git, the branch sprout/cvgv has been created

<http://perl5.git.perl.org/perl.git/commitdiff/2c8c8eb6dc0ddd2049932c0228684a2f26c3640d?hp=0000000000000000000000000000000000000000>

        at  2c8c8eb6dc0ddd2049932c0228684a2f26c3640d (commit)

- Log -----------------------------------------------------------------
commit 2c8c8eb6dc0ddd2049932c0228684a2f26c3640d
Author: Father Chrysostomos <[email protected]>
Date:   Tue Sep 2 22:39:11 2014 -0700

    test fix

M       t/op/caller.t

commit 06feb47c986d2c5f759c9550cc113b18779f5931
Author: Father Chrysostomos <[email protected]>
Date:   Sun Aug 31 20:33:16 2014 -0700

    Anonymous
    
    ---------------------
    
    sub foo { print +(caller(0))[3],"\n" }
    my $fooref = delete $::{foo};
    $fooref -> ();
    
    That piece of code used to crash in 5.7.3 or thereabouts.
    
    It was fixed by 16658 (aka 07b8c804e8) to produce ‘(unknown)’ instead.
    
    Then in 5.13.3 it was changed (by 803f274) to produce ‘main::__ANON__’ 
instead.
    
    I am trying to save memory by removing the need for CVs to carry GVs around 
with them.  It just happens to change the behaviour of the code above to 
produce ‘main::foo’ instead.
    
    Is that a bug fix?

M       t/op/caller.t

commit a5a57da43e162161fb4ead1b1b43e72b660aa1fd
Author: Father Chrysostomos <[email protected]>
Date:   Sun Aug 31 20:13:38 2014 -0700

    ntoehusnt

M       gv.c
M       op.c

commit 66c68f936e8bf8f44ef18ee5230cc5a819da4476
Author: Father Chrysostomos <[email protected]>
Date:   Sun Aug 31 18:05:49 2014 -0700

    sv.h: Expand comment about potential SVf_UTF8 conflict

M       sv.h

commit da56eae71d1fe8e8a059359f480fe70daa664168
Author: Father Chrysostomos <[email protected]>
Date:   Sat Aug 30 10:21:04 2014 -0700

    ntheontu
    
    stuff

M       embed.fnc
M       gv.c
M       op.c
M       proto.h

commit 09e74ed93fc904bd262e899d20601f3942640b66
Author: Father Chrysostomos <[email protected]>
Date:   Sat Aug 30 10:26:58 2014 -0700

    op.c: Calculate hash for CvNAME_HEK
    
    I assumed when I wrote that code that share_hek would calculate the
    hash, like most hash functions; but this internal function assumes
    the caller does it.
    
    Hence, CVs were not sharing their heks with other types of thingies
    that have heks.  A CV named foo and a GV named foo would cause two
    heks of the same name to be present in the shared string table.

M       op.c
M       pad.c

commit 61ac183a5514d6697cb9210c6d98d89f02270f15
Author: Father Chrysostomos <[email protected]>
Date:   Sun Aug 31 20:13:21 2014 -0700

    oeuntho

M       pp_hot.c

commit b8f28fcd66fd9adcdbf65c7566ad7d7e62d70f77
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 18:26:36 2014 -0700

    For lexical subs, reify CvGV from CvSTASH and CvNAME_HEK
    
    From now on, the presence of a name hek implies a GV.  Any access to
    CvGV will cause that implicit GV to be reified.

M       cv.h
M       embed.fnc
M       ext/B/t/b.t
M       gv.c
M       inline.h
M       op.c
M       pp_hot.c
M       proto.h

commit 89b34b210872251131c338e8c47e6ffb85e43e2a
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 17:40:23 2014 -0700

    Increase $XS::APItest::VERSION to 0.64

M       ext/XS-APItest/APItest.pm

commit 73119bd68ec0d85092c2dadc70bc324015f01da1
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 17:39:48 2014 -0700

    Test cv_name

M       MANIFEST
M       ext/XS-APItest/APItest.xs
A       ext/XS-APItest/t/cv_name.t

commit 5c7f103142504401fda21301c0838818018a4eb9
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 16:03:22 2014 -0700

    pad.c: Document cv_name

M       pad.c

commit 7360e32be65974421ebe5e55160074fc2453fbb9
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 15:59:05 2014 -0700

    sv_cathek
    
    This macro, intended for internal use, simplifies the code in
    a couple of places.

M       pad.c
M       sv.h
M       util.c

commit bf1bccf332e0b77ac214ad9d7b918c0adcd0dfc5
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 15:56:30 2014 -0700

    cv_name
    
    An API function for getting the name of a CV.  Docs to follow.

M       embed.fnc
M       embed.h
M       pad.c
M       proto.h

commit 8805c9694f104e9c988ddc62cba0f2e7b7c18f68
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 17:37:55 2014 -0700

    Turn on CVf_LEXICAL for lexical subs
    
    This flag will signify that lexical subs should not have package names
    associated with them in error messages, etc.

M       gv.c
M       op.c
M       pad.c
M       scope.c

commit d2315fc74da531f27cfa5735ed12f4fbc744ae28
Author: Father Chrysostomos <[email protected]>
Date:   Thu Aug 28 12:55:56 2014 -0700

    Add CVf_LEXICAL flag
    
    Lexical subs will use this instead of CvNAMED to indicate that the
    name should not include the package.

M       cv.h
-----------------------------------------------------------------------

--
Perl5 Master Repository

Reply via email to