X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regcomp.pl;h=17472cc8110f1494430a4e8d35afe7dd65a416a5;hb=cd7f9af76f0d337e0b6ef5b49db2576247d0d50f;hp=d7d07330109c9432c16551d88c63064a8d281187;hpb=9b155405e1006cdd8b8d8ffbf82a0cdd100aa3a4;p=p5sagit%2Fp5-mst-13.2.git

diff --git a/regcomp.pl b/regcomp.pl
index d7d0733..17472cc 100644
--- a/regcomp.pl
+++ b/regcomp.pl
@@ -1,43 +1,110 @@
+BEGIN {
+    # Get function prototypes
+    require 'regen_lib.pl';
+}
 #use Fatal qw(open close rename chmod unlink);
+use strict;
+use warnings;
+
 open DESC, 'regcomp.sym';
-$ind = 0;
 
+my $ind = 0;
+my (@name,@rest,@type,@code,@args,@longj);
+my ($desc,$lastregop);
 while (<DESC>) {
-  next if /^\s*($|\#)/;
-  $ind++;
-  chomp;
-  ($name[$ind], $desc, $rest[$ind]) = split /\t+/, $_, 3;
-  ($type[$ind], $code[$ind], $args[$ind], $longj[$ind]) 
-    = split /[,\s]\s*/, $desc, 4;
+    s/#.*$//;
+    next if /^\s*$/;
+    s/\s*\z//;
+    if (/^-+\s*$/) {
+        $lastregop= $ind;
+        next;
+    }
+    unless ($lastregop) {
+        $ind++;
+        ($name[$ind], $desc, $rest[$ind]) = split /\t+/, $_, 3;  
+        ($type[$ind], $code[$ind], $args[$ind], $longj[$ind]) 
+          = split /[,\s]\s*/, $desc, 4;
+    } else {
+        my ($type,@lists)=split /\s*\t+\s*/, $_;
+        die "No list? $type" if !@lists;
+        foreach my $list (@lists) {
+            my ($names,$special)=split /:/, $list , 2;
+            $special ||= "";
+            foreach my $name (split /,/,$names) {
+                my $real= $name eq 'resume' 
+                        ? "resume_$type" 
+                        : "${type}_$name";
+                my @suffix;
+                if (!$special) {
+                   @suffix=("");
+                } elsif ($special=~/\d/) {
+                    @suffix=(1..$special);
+                } elsif ($special eq 'FAIL') {
+                    @suffix=("","_fail");
+                } else {
+                    die "unknown :type ':$special'";
+                }
+                foreach my $suffix (@suffix) {
+                    $ind++;
+                    $name[$ind]="$real$suffix";
+                    $type[$ind]=$type;
+                    $rest[$ind]="state for $type";
+                }
+            }
+        }
+        
+    }
 }
+# use fixed width to keep the diffs between regcomp.pl recompiles
+# as small as possible.
+my ($width,$rwidth,$twidth)=(22,12,9);
+$lastregop ||= $ind;
+my $tot = $ind;
 close DESC;
-$tot = $ind;
+die "Too many regexp/state opcodes! Maximum is 256, but there are $lastregop in file!"
+    if $lastregop>256;
 
-$tmp_h = 'tmp_reg.h';
+my $tmp_h = 'tmp_reg.h';
 
 unlink $tmp_h if -f $tmp_h;
 
 open OUT, ">$tmp_h";
+#*OUT=\*STDOUT;
+binmode OUT;
 
-print OUT <<EOP;
-/* !!!!!!!   DO NOT EDIT THIS FILE   !!!!!!!
+printf OUT <<EOP,
+/* -*- buffer-read-only: t -*-
+   !!!!!!!   DO NOT EDIT THIS FILE   !!!!!!!
    This file is built by regcomp.pl from regcomp.sym.
    Any changes made here will be lost!
 */
 
-EOP
+/* Regops and State definitions */
+
+#define %*s\t%d
+#define %*s\t%d
 
-$ind = 0;
-while (++$ind <= $tot) {
-  $oind = $ind - 1;
-  $hind = sprintf "%#4x", $oind;
-  print OUT <<EOP;
-#define	$name[$ind]	$oind	/* $hind $rest[$ind] */
 EOP
+    -$width, REGNODE_MAX        => $lastregop - 1,
+    -$width, REGMATCH_STATE_MAX => $tot - 1
+;
+
+
+for ($ind=1; $ind <= $lastregop ; $ind++) {
+  my $oind = $ind - 1;
+  printf OUT "#define\t%*s\t%d\t/* %#04x %s */\n",
+    -$width, $name[$ind], $ind-1, $ind-1, $rest[$ind];
+}
+print OUT "\t/* ------------ States ------------- */\n";
+for ( ; $ind <= $tot ; $ind++) {
+  printf OUT "#define\t%*s\t(REGNODE_MAX + %d)\t/* %s */\n",
+    -$width, $name[$ind], $ind - $lastregop, $rest[$ind];
 }
 
 print OUT <<EOP;
 
+/* PL_regkind[] What type of regop or state is this. */
+
 #ifndef DOINIT
 EXTCONST U8 PL_regkind[];
 #else
@@ -46,74 +113,112 @@ EOP
 
 $ind = 0;
 while (++$ind <= $tot) {
-  print OUT <<EOP;
-	$type[$ind],		/* $name[$ind] */
-EOP
+  printf OUT "\t%*s\t/* %*s */\n",
+             -1-$twidth, "$type[$ind],", -$width, $name[$ind];
+  print OUT "\t/* ------------ States ------------- */\n"
+    if $ind == $lastregop and $lastregop != $tot;
 }
 
 print OUT <<EOP;
 };
 #endif
 
+/* regarglen[] - How large is the argument part of the node (in regnodes) */
 
 #ifdef REG_COMP_C
-const static U8 regarglen[] = {
+static const U8 regarglen[] = {
 EOP
 
 $ind = 0;
-while (++$ind <= $tot) {
-  $size = 0;
+while (++$ind <= $lastregop) {
+  my $size = 0;
   $size = "EXTRA_SIZE(struct regnode_$args[$ind])" if $args[$ind];
   
-  print OUT <<EOP;
-	$size,		/* $name[$ind] */
-EOP
+  printf OUT "\t%*s\t/* %*s */\n",
+	-37, "$size,",-$rwidth,$name[$ind];
 }
 
 print OUT <<EOP;
 };
 
-const static char reg_off_by_arg[] = {
+/* reg_off_by_arg[] - Which argument holds the offset to the next node */
+
+static const char reg_off_by_arg[] = {
 EOP
 
 $ind = 0;
-while (++$ind <= $tot) {
-  $size = $longj[$ind] || 0;
+while (++$ind <= $lastregop) {
+  my $size = $longj[$ind] || 0;
 
-  print OUT <<EOP;
-	$size,		/* $name[$ind] */
-EOP
+  printf OUT "\t%d,\t/* %*s */\n",
+	$size, -$rwidth, $name[$ind]
 }
 
 print OUT <<EOP;
 };
 
-#ifdef DEBUGGING
-const static char * const reg_name[] = {
+#endif /* REG_COMP_C */
+
+/* reg_name[] - Opcode/state names in string form, for debugging */
+
+#ifndef DOINIT
+EXTCONST char * PL_reg_name[];
+#else
+EXTCONST char * const PL_reg_name[] = {
 EOP
 
 $ind = 0;
+my $ofs = 1;
+my $sym = "";
 while (++$ind <= $tot) {
-  $hind = sprintf "%#4x", $ind-1;
-  $size = $longj[$ind] || 0;
-
-  print OUT <<EOP;
-	"$name[$ind]",		/* $hind */
-EOP
+  my $size = $longj[$ind] || 0;
+
+  printf OUT "\t%*s\t/* $sym%#04x */\n",
+	-3-$width,qq("$name[$ind]",), $ind - $ofs;
+  if ($ind == $lastregop and $lastregop != $tot) {
+    print OUT "\t/* ------------ States ------------- */\n";
+    $ofs = $lastregop;
+    $sym = 'REGNODE_MAX +';
+  }
+    
 }
 
 print OUT <<EOP;
 };
+#endif /* DOINIT */
 
-const static int reg_num = $tot;
-
-#endif /* DEBUGGING */
-#endif /* REG_COMP_C */
+/* PL_reg_extflags_name[] - Opcode/state names in string form, for debugging */
 
+#ifndef DOINIT
+EXTCONST char * PL_reg_extflags_name[];
+#else
+EXTCONST char * const PL_reg_extflags_name[] = {
 EOP
 
-close OUT;
+open my $fh,"<","regexp.h" or die "Can't read regexp.h: $!";
+my %rxfv;
+my $val;
+while (<$fh>) {
+    if (/#define\s+(RXf_\w+)\s+(0x[A-F\d]+)/i) {
+        $rxfv{$1}= eval $2;
+        $val|=$rxfv{$1};
+    }
+}    
+my %vrxf=reverse %rxfv;
+printf OUT "\t/* Bits in extflags defined: %032b */\n",$val;
+for (0..31) {
+    my $n=$vrxf{2**$_}||"UNUSED_BIT_$_";
+    $n=~s/^RXf_(PMf_)?//;
+    printf OUT qq(\t%-20s/* 0x%08x */\n), 
+        qq("$n",),2**$_;
+}  
+ 
+print OUT <<EOP;
+};
+#endif /* DOINIT */
+
+/* ex: set ro: */
+EOP
+close OUT or die "close $tmp_h: $!";
 
-chmod 0666, 'regnodes.h';
-unlink 'regnodes.h';
-rename $tmp_h, 'regnodes.h';
+safer_rename $tmp_h, 'regnodes.h';