perldebguts.pod description of scalar ${"_<$filename"} doesnt match code

[p5pRT]

Migrated from rt.perl.org#125492 (status was 'open')

Searchable as RT125492$

[p5pRT]

From @bulk88

Created by @bulk88

This ticket is related to
https://rt.perl.org/Public/Bug/Display.html?id=125296

perldebguts says

---------------------------------------------------
=item *

Each scalar C<${"_<$filename"}> contains C<"_<$filename">. This is
also the case for evaluated strings that contain subroutines, or
which are currently being executed. The $filename for C<eval>ed
strings looks like C<(eval 34)>.
---------------------------------------------------

This is not true.

---------------------------------------------------
BEGIN {$^P |=2;}
use warnings;
use Data​::Dumper;
use Devel​::Peek 'Dump';
$Data​::Dumper​::Sortkeys = 1;

print Dumper(${'​::_<Peek.c'});
print Dumper(${'​::_<C​:/perl521/srcnewb4opt/lib/Data/Dumper.pm'});
eval q|
print Dumper(${'​::_<(eval 3)'});|;
---------------------------------------------------
$VAR1 = 'Peek.c';
$VAR1 = 'C​:/perl521/srcnewb4opt/lib/Data/Dumper.pm';
$VAR1 = '(eval 3)';
---------------------------------------------------

That perldebguts item came from commit
http​://perl5.git.perl.org/perl.git/commitdiff/36477c247f3c188fb8cc7e276c87b739d3e6ab7c
"[inseparable changes from patch from perl5.003_10 to perl5.003_11]" and
specifically "10+ debugger patch" . I think that POD was based off a
comment in
http​://perl5.git.perl.org/perl.git/commitdiff/55497cffdd24c959994f9a8ddd56db8ce85e1c5b
  "[inseparable changes from patch from perl5.003_07 to perl5.003_08]"
specifically "Major '..' and debugger patches".

The actual C code that determines {SCALAR} contains a normal path and
not a "_<" path came from day 1 of gv_fetchfile in "perl 5.0 alpha 2"
http​://perl5.git.perl.org/perl.git/commitdiff/79072805bf63abe5b5978b5928ab00d360ea3e7f
which came from a function called fstab which is a near clone of
gv_fetchfile. fstab came from "perl 3.0 patch #34 patch #29, continued"
http​://perl5.git.perl.org/perl.git/commitdiff/0a12ae7dee71b6eb0609c35185096ab75c95b2da
. fstab was first used 1 commit ahead in at
http​://perl5.git.perl.org/perl.git/commitdiff/395c379347344a50494d2458b3a5e38ebdeac851
.

Most shocking is changing gv_fetchfile to store the "_<" path in SCALAR,
does not cause any tests to fail on a harness run.

------------------------BEFORE----------------------------------
GV *
Perl_gv_fetchfile_flags(pTHX_ const char *const name, const STRLEN namelen,
  const U32 flags)
{
  char smallbuf[128];
  char *tmpbuf;
  const STRLEN tmplen = namelen + 2;
  GV *gv;

  PERL_ARGS_ASSERT_GV_FETCHFILE_FLAGS;
  PERL_UNUSED_ARG(flags);

  if (!PL_defstash)
  return NULL;

  if (tmplen <= sizeof smallbuf)
  tmpbuf = smallbuf;
  else
  Newx(tmpbuf, tmplen, char);
  /* This is where the debugger's %{"​::_<$filename"} hash is created */
  tmpbuf[0] = '_';
  tmpbuf[1] = '<';
  memcpy(tmpbuf + 2, name, namelen);
  gv = *(GV**)hv_fetch(PL_defstash, tmpbuf, tmplen, TRUE);
  if (!isGV(gv)) {
  gv_init(gv, PL_defstash, tmpbuf, tmplen, FALSE);
#ifdef PERL_DONT_CREATE_GVSV
  GvSV(gv) = newSVpvn(name, namelen);
#else
  sv_setpvn(GvSV(gv), name, namelen);
#endif

------------------------AFTER----------------------------------
GV *
Perl_gv_fetchfile_flags(pTHX_ const char *const name, const STRLEN namelen,
  const U32 flags)
{
  char smallbuf[128];
  char *tmpbuf;
  const STRLEN tmplen = namelen + 2;
  GV *gv;

  PERL_ARGS_ASSERT_GV_FETCHFILE_FLAGS;
  PERL_UNUSED_ARG(flags);

  if (!PL_defstash)
  return NULL;

  if (tmplen <= sizeof smallbuf)
  tmpbuf = smallbuf;
  else
  Newx(tmpbuf, tmplen, char);
  /* This is where the debugger's %{"​::_<$filename"} hash is created */
  tmpbuf[0] = '_';
  tmpbuf[1] = '<';
  memcpy(tmpbuf + 2, name, namelen);
  gv = *(GV**)hv_fetch(PL_defstash, tmpbuf, tmplen, TRUE);
  if (!isGV(gv)) {
  gv_init(gv, PL_defstash, tmpbuf, tmplen, FALSE);
#ifdef PERL_DONT_CREATE_GVSV
  GvSV(gv) = newSVpvn(tmpbuf, tmplen);
#else
  sv_setpvn(GvSV(gv), tmpbuf, tmplen);
#endif
  }
----------------------------------------------------------

So it seems either the SCALAR part of the "_<" glob is unused by core
(maybe it should be deleted to save memory leaving only the hash and
array parts of the "_<" glob?), or is never used during a harness run
(perl5db.pl has no tests).

So my question is, fix the POD to match the code, or fix the code to
match the POD, or delete the code/feature entirely since its been unused
for 22 years?

Perl Info

---	
Flags:
                 category=core
                 severity=low

Site configuration information for perl 5.21.4:

Configured by Owner at Thu Sep 18 12:08:58 2014.

Summary of my perl5 (revision 5 version 21 subversion 4) configuration:
               Derived from: 7d2b2edb94ab56333b9049a3e26d15ea18445512
               Ancestor: 19be3be6968e2337bcdfe480693fff795ecd1304
               Platform:
                 osname=MSWin32, osvers=5.1,
archname=MSWin32-x86-multi-thread
                 uname=''
                 config_args='undef'
                 hint=recommended, useposix=true, d_sigaction=undef
                 useithreads=define, usemultiplicity=define
                 use64bitint=undef, use64bitall=undef, uselongdouble=undef
                 usemymalloc=n, bincompat5005=undef
               Compiler:
                 cc='cl', ccflags ='-nologo -GF -W3 -O1 -MD -Zi -DNDEBUG
-DWIN32
-D_CONSOLE -DNO_STRICT  -DPERL_TEXTMODE_SCRIPTS
-DPERL_HASH_FUNC_ONE_AT_A_TIME -DPERL_IMPLICIT_CONTEXT
-DPERL_IMPLICIT_SYS -DUSE_PERLIO -D_USE_32BIT_TIME_T',
                 optimize='-O1 -MD -Zi -DNDEBUG',
                 cppflags='-DWIN32'
                 ccversion='12.00.8168', gccversion='', gccosandvers=''
                 intsize=4, longsize=4, ptrsize=4, doublesize=8,
byteorder=1234
                 d_longlong=undef, longlongsize=8, d_longdbl=define,
longdblsize=8,
longdblkind=0
                 ivtype='long', ivsize=4, nvtype='double', nvsize=8,
Off_t='__int64',
lseeksize=8
                 alignbytes=8, prototype=define
               Linker and Libraries:
                 ld='link', ldflags ='-nologo -nodefaultlib -debug
-opt:ref,icf
-libpath:"c:\perl521\lib\CORE"  -machine:x86'
                 libpth=C:\PROGRA~1\MIAF9D~1\VC98\lib
                 libs=oldnames.lib kernel32.lib user32.lib gdi32.lib
winspool.lib
comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib
netapi32.lib uuid.lib ws2_32.lib mpr.lib winmm.lib  version.lib
odbc32.lib odbccp32.lib comctl32.lib msvcrt.lib
                 perllibs=oldnames.lib kernel32.lib user32.lib gdi32.lib
winspool.lib  comdlg32.lib advapi32.lib shell32.lib ole32.lib
oleaut32.lib  netapi32.lib uuid.lib ws2_32.lib mpr.lib winmm.lib
version.lib odbc32.lib odbccp32.lib comctl32.lib msvcrt.lib
                 libc=msvcrt.lib, so=dll, useshrplib=true,
libperl=perl521.lib
                 gnulibc_version=''
               Dynamic Linking:
                 dlsrc=dl_win32.xs, dlext=dll, d_dlsymun=undef,
ccdlflags=' '
                 cccdlflags=' ', lddlflags='-dll -nologo -nodefaultlib
-debug
-opt:ref,icf  -libpath:"c:\perl521\lib\CORE"  -machine:x86'

Locally applied patches:
                 uncommitted-changes
                 a0fe7a7e75de29e59f1da0d6822dc06e5be658fe
                 a261faffee83d0145642ab5d1d046c9f813bc497
                 6506ab86ad1602a9ca720fcd30446dce1461d23d
                 7d2b2edb94ab56333b9049a3e26d15ea18445512


@INC for perl 5.21.4:
                 lib
                 C:/perl521/srcnew/lib
                 .


Environment for perl 5.21.4:
                 HOME (unset)
                 LANG (unset)
                 LANGUAGE (unset)
                 LD_LIBRARY_PATH (unset)
                 LOGDIR (unset)
                 PATH=
                 PERL_BADLANG (unset)
                 PERL_JSON_BACKEND=Cpanel::JSON::XS
                 PERL_YAML_BACKEND=YAML
                 SHELL (unset)

[p5pRT]

From @rjbs

* bulk88 <perlbug-followup@​perl.org> [2015-06-26T18​:40​:56]

So it seems either the SCALAR part of the "_<" glob is unused by core
(maybe it should be deleted to save memory leaving only the hash and
array parts of the "_<" glob?), or is never used during a harness run
(perl5db.pl has no tests).

Shooting from the hip​: fix the docs to match the code. The current behavior
seems more useful.

Oh, and add a test. :-)

--
rjbs

[p5pRT]

The RT System itself - Status changed from 'new' to 'open'