Eliminating the duplicate logic in S_method_common() and

Perl_gv_fetchmethod_autoload() is clearly TODO.
Move the paragraph
"strcat(), strcpy(), strncat(), strncpy(), sprintf(), vsprintf()"
to the section "Tasks that need a little C knowledge" as it doesn't
need any XS knowledge.

p4raw-id: //depot/perl@32478

diff --git a/gv.c b/gv.c
index 9ab582d..88e9993 100644 (file)
--- a/gv.c
+++ b/gv.c
@@ -575,6 +575,17 @@ S_gv_get_super_pkg(pTHX_ const char* name, I32 namelen)
     return stash;
 }
 
+/* FIXME. If changing this function note the comment in pp_hot's
+   S_method_common:
+
+   This code tries to figure out just what went wrong with
+   gv_fetchmethod.  It therefore needs to duplicate a lot of
+   the internals of that function. ...
+
+   I'd guess that with one more flag bit that could all be moved inside
+   here.
+*/
+
 GV *
 Perl_gv_fetchmethod_autoload(pTHX_ HV *stash, const char *name, I32 autoload)
 {

diff --git a/pod/perltodo.pod b/pod/perltodo.pod
index 5adc1fb..63da3f9 100644 (file)
--- a/pod/perltodo.pod
+++ b/pod/perltodo.pod
@@ -468,6 +468,18 @@ Jarkko notes that one can things morally equivalent to C<__FUNCTION__>
 See L<http://www.codeproject.com/debug/extendedtrace.asp> if you feel like
 making C<PERL_MEM_LOG> more useful on Win32.
 
+=head2 strcat(), strcpy(), strncat(), strncpy(), sprintf(), vsprintf()
+
+Maybe create a utility that checks after each libperl.a creation that
+none of the above (nor sprintf(), vsprintf(), or *SHUDDER* gets())
+ever creep back to libperl.a.
+
+  nm libperl.a | ./miniperl -alne '$o = $F[0] if /:$/; print "$o $F[1]" if $F[0] eq "U" && $F[1] =~ /^(?:strn?c(?:at|py)|v?sprintf|gets)$/'
+
+Note, of course, that this will only tell whether B<your> platform
+is using those naughty interfaces.
+
+
 =head1 Tasks that need a knowledge of XS
 
 These tasks would need C knowledge, and roughly the level of knowledge of
@@ -547,17 +559,6 @@ system() accepts a LIST syntax (and a PROGRAM LIST syntax) to avoid
 running a shell. readpipe() (the function behind qx//) could be similarly
 extended.
 
-=head2 strcat(), strcpy(), strncat(), strncpy(), sprintf(), vsprintf()
-
-Maybe create a utility that checks after each libperl.a creation that
-none of the above (nor sprintf(), vsprintf(), or *SHUDDER* gets())
-ever creep back to libperl.a.
-
-  nm libperl.a | ./miniperl -alne '$o = $F[0] if /:$/; print "$o $F[1]" if $F[0] eq "U" && $F[1] =~ /^(?:strn?c(?:at|py)|v?sprintf|gets)$/'
-
-Note, of course, that this will only tell whether B<your> platform
-is using those naughty interfaces.
-
 =head2 Audit the code for destruction ordering assumptions
 
 Change 25773 notes
@@ -608,6 +609,25 @@ from <char *> to <const char *>. It should now be possible to propagate
 const-correctness outwards to C<S_parse_body()>, C<Perl_moreswitches()>
 and C<Perl_yylex()>.
 
+=head2 Duplicate logic in S_method_common() and Perl_gv_fetchmethod_autoload()
+
+A comment in C<S_method_common> notes
+
+       /* This code tries to figure out just what went wrong with
+          gv_fetchmethod.  It therefore needs to duplicate a lot of
+          the internals of that function.  We can't move it inside
+          Perl_gv_fetchmethod_autoload(), however, since that would
+          cause UNIVERSAL->can("NoSuchPackage::foo") to croak, and we
+          don't want that.
+       */
+
+If C<Perl_gv_fetchmethod_autoload> gets rewritten to take (more) flag bits,
+then it ought to be possible to move the logic from C<S_method_common> to
+the "right" place. When making this change it would probably be good to also
+pass in at least the method name length, if not also pre-computed hash values
+when known. (I'm contemplating a plan to pre-compute hash values for common
+fixed strings such as C<ISA> and pass them in to functions.)
+
 
 =head1 Tasks that need a knowledge of the interpreter