perl 5.0 alpha 3

[editor's note: the sparc executables have not been included,
and emacs backup files have been removed]

Author: Larry Wall

Bugs/minmax

@@ -0,0 +1,24 @@
+Article 433 of comp.os.386bsd.apps:
+Path: netlabs!news.cerf.net!usc!howland.reston.ans.net!spool.mu.edu!bloom-beacon.mit.edu!ai-lab!life.ai.mit.edu!mycroft
+From: mycroft@trinity.gnu.ai.mit.edu (Charles Hannum)
+Newsgroups: comp.os.386bsd.apps
+Subject: Re: Perl-4.036?
+Date: 06 Sep 1993 19:01:10 GMT
+Organization: MIT Artificial Intelligence Lab
+Lines: 9
+Message-ID: <MYCROFT.93Sep6150110@trinity.gnu.ai.mit.edu>
+References: <26fptu$1q1@terminator.rs.itd.umich.edu> <26fve4$ivf@homer.cs.mcgill.ca>
+NNTP-Posting-Host: trinity.gnu.ai.mit.edu
+In-reply-to: storm@cs.mcgill.ca's message of 6 Sep 1993 18:27:16 GMT
+
+
+   Perl 4.036 comipled without a single hitch under NetBSD 0.9 last
+   week.  It failed the db test, but I suspect that is due to the new
+   db stuff under NetBSD and the like...
+
+Yes.  The perl test seems to expect the database to be put in
+`foo.pag' and `foo.dir', which isn't the case any more.  I suspect
+lwall will fix this soon.
+
+
+

Bugs/misparse

@@ -0,0 +1 @@
+s/\w/[\u$&\l$&]/gi;

Bugs/pagdir

@@ -0,0 +1,64 @@
+Article 13355 of comp.lang.perl:
+Newsgroups: comp.lang.perl
+Path: netlabs!news.cerf.net!usc!cs.utexas.edu!uunet!fmrco!fmrco!asherman
+From: asherman@fmrco.com (Aaron Sherman)
+Subject: Re: perl 5a2 cannot "die" (plus current list o' bugs)
+In-Reply-To: wjm@feenix.metronet.com's message of Fri, 20 Aug 1993 21:32:10 GMT
+Message-ID: <ASHERMAN.93Aug23094250@uboat.fmrco.com>
+Sender: news@fmrco.uucp
+X-Quote: "...accepting is much harder than giving." -Mike Smith
+Reply-To: asherman@fmrco.COM
+Organization: I-Kinetics, 19 Bishop-Allen Dr., Cambridge, MA
+References: <ASHERMAN.93Aug20102959@uboat.fmrco.com> <CC2uHM.6Hq@feenix.metronet.com>
+Date: Mon, 23 Aug 1993 14:42:50 GMT
+Lines: 47
+
+
+>>>>> wjm@feenix.metronet.com (Bill Middleton) said:
+
+wjm> asherman@fmrco.COM writes:
+
+>An interesting pair of bugs can be seen in the following output:
+
+wjm> I dont think so.  Could be in the compilation or something.  Did it
+wjm> pass all tests?  Each of the following work fine here on this HP.
+
+I tried compiling with Sun's native CC and GCC. Both worked fine, but
+caused this problem. I'll try it with Larry's original version when I
+get a chance.
+
+wjm> perl5 -e 'die "hello $. \n";
+wjm> hello  
+
+Ah. But, note that the $. STILL isn't working. So only ONE of those
+bugs did not show.
+
+This is my current list of bugs (not complete, but what I've had time
+to note). Hope it helps:
+
+"perl -e die" will cause a seg-fault
+
+$. is not updated
+
+Memory leak for anonymous arrays:
+	while(1) { @a = (1, 2, 3, [4, 5], 6); @a = (); }
+	Will keep allocating and not freeing memory.
+
+"perl -e 'sub foo {print 1} foo'" should either complain or call foo,
+	but does neither. Or, did Larry not impliment the &-less
+	function calling that he was talking about?
+
+"perl -le 'sub foo {1} $a = \&foo; print &{$a}; print &{$a} + 1'" should
+	not fail to parse.
+
+
+			-AJS
+
+--
+Aaron Sherman			I-Kinetics, Inc.
+Systems Engineer		  "Open Systems Stepstones"
+Voice: (617)661-8181		19 Bishop Allen Dr.
+Fax:   (617)661-8625		Cambridge, MA 02139
+Pager: (508)545-0584		asherman@i-kinetics.com
+
+

Bugs/replacecase

@@ -1,18 +1,75 @@
-
+Incompatibilities
+-----------------
     s'$lhs'$rhs' now does no interpolation on either side.  It used to
     interplolate $lhs but not $rhs.
 
     The second and third arguments of splice are now evaluated in scalar
     context (like the book says) rather than list context.
 
-    Saying shift @foo + 20 is now a semantic error.
+    Saying "shift @foo + 20" is now a semantic error because of precedence.
+
+    "open FOO || die" is now incorrect.  You need parens around the filehandle.
 
     The elements of argument lists for formats are now evaluated in list
-    context.
+    context.  This means you can interpolate list values now.
 
-    You can't do a goto into a block that is optimized away.
+    You can't do a goto into a block that is optimized away.  Darn.
 
     It is no longer syntactically legal to use whitespace as the name
     of a variable.
 
     Some error messages will be different.
+
+    The caller function now a false value in a scalar context if there is
+    no caller.  This lets library files determine if they're being required.
+
+    m//g now attaches its state to the searched string rather than the
+    regular expression.
+
+New things
+----------
+    The -w switch is much more informative.
+
+    References.  See t/op/ref.t for examples.
+
+    Objects.  See t/op/ref.t for examples.
+
+    => is now a synonym for comma.  This is useful as documentation for
+    arguments that come in pairs, such as initializers for associative arrays,
+    or named arguments to a subroutine.
+
+    All functions have been turned into list operators or unary operators,
+    meaning the parens are optional.  Even subroutines may be called as
+    list operators if they've already been declared.
+
+    More embeddible.  See main.c and embed_h.SH.
+
+    The interpreter is now flattened out.  Compare Perl 4's eval.c with
+    the perl 5's pp.c.  Compare Perl 4's 900 line interpreter loop in cmd.c
+    with Perl 5's 1 line interpreter loop in run.c.  Eventually we'll make
+    everything non-blocking so we can interface nicely with a scheduler.
+
+    eval is now treated more like a subroutine call.  Among other things,
+    this means you can return from it.
+
+    Format value lists may be spread over multiple lines by enclosing in
+    curlies.
+
+    You may now define BEGIN and END subroutines for each package.  The BEGIN
+    subroutine executes the moment it's parsed.  The END subroutine executes
+    just before exiting.
+
+    Flags on the #! line are interpreted even if the script wasn't
+    executed directly.  (And even if the script was located by "perl -x"!)
+
+    The ?: operator is now legal as an lvalue.
+
+    List context now propagates to the right side of && and ||, as well
+    as the 2nd and 3rd arguments to ?:.
+
+    The "defined" function can now take a general expression.
+
+    Lexical scoping available via "my".  eval can see the current lexical
+    variables.
+
+    Saying "package;" requires explicit package name on global symbols.

Bugs/stuff

@@ -360,6 +360,7 @@ shlist:
 # AUTOMATICALLY GENERATED MAKE DEPENDENCIES--PUT NOTHING BELOW THIS LINE
 $(obj) hv.o:
 	@ echo "You haven't done a "'"make depend" yet!'; exit 1
+
 makedepend: makedepend.SH
 	/bin/sh $(shellflags) makedepend.SH
 

Changes

@@ -1,18 +1,31 @@
+POSIX compatibility
+hash binding
+pack(P)
+fix gv refcnts
+
+/\Afoo/ (beginning of string, or where previous g match left off)
+/foo\Z/ (end of string only)
+Make specialized allocators
+Optimize switch statements
+Optimize foreach on array
+Optimize foreach (1..1000000)
 Set KEEP on constant split
-Optimize foreach on array.
-Execute all BEGINs and ENDs.
-Make a good way to determine if *.pl is being executed directly.
-Make specialized allocators.
-Optimize switches.
-Do debugger
-Cache eval tree
-Implement eval once
-Cache m//g state
+Cache eval tree (unless lexical outer scope used (mark in &compiling?))
 rcatmaybe
-Fix length($&)
-eval {} coredump
+Cache method lookup
+Shrink opcode tables via multiple implementations selected in peep
+Straighten out the RS stuff in BEGIN
+Make yyparse recursion longjmp() proof.
 
+sub mysplice(@, $, $, ...)?
+pretty function?  (or is it, as I suspect, a lib routine?)
+perlex function?
+X Perl?  Motif Perl?
+give DOG $bone?
+Nested destructors?
 make tr/// return histogram in list context?
-Do anything with "hint"?
+Implement eval once?  (Unnecessary with cache?)
 When does split() go to @_?
 undef wantarray in void context?
+goto &coroutine?
+filehandle references?

Makefile

@@ -1 +1,5 @@
-###############################################################################
+#!./perl
+
+sub foo;
+
+foo;

Todo

@@ -233,8 +233,8 @@ medium=''
 large=''
 huge=''
 optimize='-g'
-ccflags=' -I/usr/include/sun -I/usr/ucbinclude -DDEBUGGING'
-cppflags=' -I/usr/include/sun -I/usr/ucbinclude -DDEBUGGING'
+ccflags='-DDEBUGGING'
+cppflags='-DDEBUGGING'
 ldflags=''
 cc='cc'
 nativegcc=''

bar

@@ -49,6 +49,7 @@ struct cop {
 	struct ccop ccop;	/* compound command */
 	struct scop scop;	/* switch command */
     } uop;
+    U32		cop_seq;	/* parse sequence number */
     short	cop_slen;	/* len of cop_short, if not null */
     VOL short	cop_flags;	/* optimization flags--see above */
     HV *	cop_stash;	/* package line was compiled in */
@@ -70,6 +71,7 @@ struct block_sub {
     GV *	defgv;
     AV *	savearray;
     AV *	argarray;
+    AV *	comppad;
     U16		olddepth;
     U8		hasargs;
 };

config.sh

@@ -236,7 +236,7 @@ I32 len;
 	}
     }
     if (!fp) {
-	if (dowarn && io->type == '<' && index(name, '\n'))
+	if (dowarn && io->type == '<' && strchr(name, '\n'))
 	    warn(warn_nl, "open");
 	Safefree(myname);
 	goto say_false;
@@ -909,7 +909,7 @@ dARGS
 	sv_setpv(statname,SvPVn(sv));
 	laststype = OP_STAT;
 	laststatval = stat(SvPVn(sv),&statcache);
-	if (laststatval < 0 && dowarn && index(SvPVn(sv), '\n'))
+	if (laststatval < 0 && dowarn && strchr(SvPVn(sv), '\n'))
 	    warn(warn_nl, "stat");
 	return laststatval;
     }
@@ -941,7 +941,7 @@ dARGS
 #else
     laststatval = stat(SvPVn(sv),&statcache);
 #endif
-    if (laststatval < 0 && dowarn && index(SvPVn(sv), '\n'))
+    if (laststatval < 0 && dowarn && strchr(SvPVn(sv), '\n'))
 	warn(warn_nl, "lstat");
     return laststatval;
 }
@@ -1033,7 +1033,7 @@ char *cmd;
     if (*s == '=')
 	goto doshell;
     for (s = cmd; *s; s++) {
-	if (*s != ' ' && !isALPHA(*s) && index("$&*(){}[]'\";\\|?<>~`\n",*s)) {
+	if (*s != ' ' && !isALPHA(*s) && strchr("$&*(){}[]'\";\\|?<>~`\n",*s)) {
 	    if (*s == '\n' && !s[1]) {
 		*s = '\0';
 		break;

cop.h

@@ -89,7 +89,7 @@ OP *arg;
 
     tbl = (short*) cPVOP->op_pv;
     s = SvPVn(sv);
-    send = s + SvCUR(sv);
+    send = s + SvCUROK(sv);
     if (!tbl || !s)
 	fatal("panic: do_trans");
     DEBUG_t( deb("2.TBL\n"));
@@ -139,7 +139,7 @@ register SV **sp;
     register I32 items = sp - mark;
     register char *delim = SvPVn(del);
     register STRLEN len;
-    I32 delimlen = SvCUR(del);
+    I32 delimlen = SvCUROK(del);
 
     mark++;
     len = (items > 0 ? (delimlen * (items - 1) ) : 0);
@@ -207,7 +207,7 @@ register SV **sarg;
     sv_setpv(sv,"");
     len--;			/* don't count pattern string */
     t = s = SvPVn(*sarg);
-    send = s + SvCUR(*sarg);
+    send = s + SvCUROK(*sarg);
     sarg++;
     for ( ; ; len--) {
 
@@ -316,7 +316,7 @@ register SV **sarg;
 		    break;		/* so handle simple cases */
 		}
 		else if (f[1] == '-') {
-		    char *mp = index(f, '.');
+		    char *mp = strchr(f, '.');
 		    I32 min = atoi(f+2);
 
 		    if (mp) {
@@ -330,7 +330,7 @@ register SV **sarg;
 		    break;
 		}
 		else if (isDIGIT(f[1])) {
-		    char *mp = index(f, '.');
+		    char *mp = strchr(f, '.');
 		    I32 min = atoi(f+1);
 
 		    if (mp) {
@@ -438,7 +438,7 @@ register SV *sv;
 	return;
     }
     tmps = SvPVn(sv);
-    if (tmps && SvCUR(sv)) {
+    if (tmps && SvCUROK(sv)) {
 	tmps += SvCUR(sv) - 1;
 	sv_setpvn(astr,tmps,1);	/* remember last char */
 	*tmps = '\0';				/* wipe it out */
@@ -466,10 +466,10 @@ SV *right;
     register char *lc = SvPVn(left);
     register char *rc = SvPVn(right);
     register I32 len;
+    I32 leftlen = SvCUROK(left);
+    I32 rightlen = SvCUROK(right);
 
-    len = SvCUR(left);
-    if (len > SvCUR(right))
-	len = SvCUR(right);
+    len = leftlen < rightlen ? leftlen : rightlen;
     if (SvTYPE(sv) < SVt_PV)
 	sv_upgrade(sv, SVt_PV);
     if (SvCUR(sv) > len)
@@ -545,10 +545,10 @@ SV *right;
 	    *dc++ = *lc++ | *rc++;
       mop_up:
 	len = SvCUR(sv);
-	if (SvCUR(right) > len)
-	    sv_catpvn(sv,SvPV(right)+len,SvCUR(right) - len);
-	else if (SvCUR(left) > len)
-	    sv_catpvn(sv,SvPV(left)+len,SvCUR(left) - len);
+	if (rightlen > len)
+	    sv_catpvn(sv, SvPV(right) + len, rightlen - len);
+	else if (leftlen > len)
+	    sv_catpvn(sv, SvPV(left) + len, leftlen - len);
 	break;
     }
 }