perl.h: don't include charclass_invlists.h indiscriminately

charclass_invlists.h is a generated file with a lot of code (4.5 MB).
The vast majority of this code is skipped (masked by #ifdef .. #endif),
but since perl.h includes it, the preprocessor has to slog
through all 4.5 MB every time a *.c core file is compiled, slowing down
the build. This also affects XS modules, which normally include perl.h.

This commit rearranges things a bit in order to speed up compilation.
For details, see below.

Conceptually speaking, charclass_invlists.h consists of 4 parts:

 1. Code only active in regcomp.c
 2. Code only active in regexec.c
 3. Code only active in utf8.c
 4. Other code (always active)

As it turns out, part 4 consists of two constants (NUM_ANYOF_CODE_POINTS
and MAX_FOLD_FROMS) and nothing else. Furthermore, these are only needed
in regexp.h.

This commit splits off part 4 into a new header file
(regexp_constants.h), which (like charclass_invlists.h) is generated by
regen/mk_invlists.pl and included by regexp.h, its only consumer.

Ideally, the rest of charclass_invlists.h (parts 1-3) should not be
included in perl.h, but only in regcomp.c/regexec.c/utf8.c. However,
this causes problems in practice: The generated code uses symbols like
NULL, U8, or U32, so it cannot be included before perl.h, but it defines
types that are needed by other headers (proto.h), so it cannot be
included after perl.h either. I couldn't figure out how to disentangle
this sanely, so now perl.h still includes charclass_invlists.h, but
hidden behind an

    #if defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_REGEXEC_C) || defined(PERL_IN_UTF8_C)

guard. This way most consumers of perl.h will never have to touch
charclass_invlists.h.

Fixes #22678.

Author: mauke

MANIFEST

@@ -176,6 +176,7 @@ regen.pl				Run all scripts that (re)generate files
 regen_perly.pl				generate perly.{act,h,tab} from perly.y
 regexec.c				Regular expression evaluator
 regexp.h				Public declarations for the above
+regexp_constants.h			Generated by regen/mk_invlists.pl
 reginline.h				Inline subs for the RE engine.
 regnodes.h				Description of nodes of the RE engine
 run.c					The interpreter loop

Makefile.SH

@@ -534,7 +534,8 @@ h = \
     handy.h hv.h hv_func.h iperlsys.h keywords.h l1_char_class_tab.h \
     mg.h mydtrace.h op.h op_reg_common.h opcode.h pad.h patchlevel.h \
     perl.h perlapi.h perly.h pp.h proto.h regcomp.h regcomp_internal.h \
-    regexp.h scope.h sv.h thread.h unixish.h utf8.h util.h warnings.h \
+    regexp.h regexp_constants.h scope.h sv.h thread.h unixish.h utf8.h \
+    util.h warnings.h \
     $(CONFIGH)
 
 c_base = \

charclass_invlists.h

@@ -3307,7 +3307,9 @@ typedef struct padname PADNAME;
 #endif
 
 #include "handy.h"
-#include "charclass_invlists.h"
+#if defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_REGEXEC_C) || defined(PERL_IN_UTF8_C)
+#  include "charclass_invlists.h"
+#endif
 
 #if defined(USE_LARGE_FILES) && !defined(NO_64_BIT_RAWIO)
 #   if LSEEKSIZE == 8 && !defined(USE_64_BIT_RAWIO)

lib/unicore/uni_keywords.pl

@@ -1,6 +1,5 @@
-#!perl -w
-use 5.015;
-use strict;
+#!perl
+use v5.16;
 use warnings;
 use Unicode::UCD qw(prop_aliases
                     prop_values
@@ -59,13 +58,20 @@
                       {style => '*', by => 'regen/mk_invlists.pl',
                       from => "Unicode::UCD"});
 
+my $regexp_constants_fh = open_new('regexp_constants.h', '>',
+                      {style => '*', by => 'regen/mk_invlists.pl',
+                      from => "Unicode::UCD"});
+
 my $in_file_pound_if = "";
 
 my $max_hdr_len = 3;    # In headings, how wide a name is allowed?
 
-print $out_fh "/* See the generating file for comments */\n\n";
-
 print $out_fh <<'EOF';
+/* See the generating file for comments */
+
+EOF
+
+print $regexp_constants_fh <<'EOF';
 /* This gives the number of code points that can be in the bitmap of an ANYOF
  * node.  The shift number must currently be one of: 8..12.  It can't be less
  * than 8 (256) because some code relies on it being at least that.  Above 12
@@ -84,7 +90,7 @@
 
 my $num_anyof_code_points = '(1 << 8)';
 
-print $out_fh "#define NUM_ANYOF_CODE_POINTS   $num_anyof_code_points\n\n";
+print $regexp_constants_fh "#define NUM_ANYOF_CODE_POINTS   $num_anyof_code_points\n\n";
 
 $num_anyof_code_points = eval $num_anyof_code_points;
 
@@ -3300,7 +3306,7 @@ sub token_name
 
 print "Computing fold data\n" if DEBUG;
 
-print $out_fh <<"EOF";
+print $regexp_constants_fh <<"EOF";
 
 /* More than one code point may have the same code point as their fold.  This
  * gives the maximum number in the current Unicode release.  (The folded-to
@@ -3337,6 +3343,8 @@ sub token_name
     }
 }
 
+read_only_bottom_close_and_rename($regexp_constants_fh, \@sources);
+
 read_only_bottom_close_and_rename($out_fh, \@sources);
 
 my %name_to_index;

perl.h

@@ -680,6 +680,8 @@ and check for NULL.
 
 /* Stuff that needs to be included in the pluggable extension goes below here */
 
+#include "regexp_constants.h"
+
 #ifdef PERL_ANY_COW
 #  define RXp_MATCH_COPY_FREE(prog)                                 \
     STMT_START {                                                    \

regen/mk_invlists.pl

@@ -0,0 +1,86 @@
+/* -*- mode: C; buffer-read-only: t -*-
+ * !!!!!!!   DO NOT EDIT THIS FILE   !!!!!!!
+ * This file is built by regen/mk_invlists.pl from Unicode::UCD.
+ * Any changes made here will be lost!
+ */
+
+/* This gives the number of code points that can be in the bitmap of an ANYOF
+ * node.  The shift number must currently be one of: 8..12.  It can't be less
+ * than 8 (256) because some code relies on it being at least that.  Above 12
+ * (4096), and you start running into warnings that some data structure widths
+ * have been exceeded, though the test suite as of this writing still passes
+ * for up through 16, which is as high as anyone would ever want to go,
+ * encompassing all of the Unicode BMP, and thus including all the economically
+ * important world scripts.  At 12 most of them are: including Arabic,
+ * Cyrillic, Greek, Hebrew, Indian subcontinent, Latin, and Thai; but not Han,
+ * Japanese, nor Korean.  The regnode sizing data structure in regnodes.h currently
+ * uses a U8, and the trie types TRIEC and AHOCORASICKC are larger than U8 for
+ * shift values above 12.)  Be sure to benchmark before changing, as larger sizes
+ * do significantly slow down the test suite. */
+
+#define NUM_ANYOF_CODE_POINTS   (1 << 8)
+
+
+/* More than one code point may have the same code point as their fold.  This
+ * gives the maximum number in the current Unicode release.  (The folded-to
+ * code point is not included in this count.)  For example, both 'S' and
+ * \x{17F} fold to 's', so the number for that fold is 2.  Another way to
+ * look at it is the maximum length of all the IVCF_AUX_TABLE's */
+#define MAX_FOLD_FROMS 3
+
+/* Generated from:
+ * 0e8307ab7c654d9c133ea885f5413a4eb5c0123ed2178f7e1cbabed36b67792c lib/Unicode/UCD.pm
+ * eb840f36e0a7446293578c684a54c6d83d249abde7bdd4dfa89794af1d7fe9e9 lib/unicore/ArabicShaping.txt
+ * 333ae1e99db0504ca8a046a07dc45b5e7aa91869c685e6bf955ebe674804827a lib/unicore/BidiBrackets.txt
+ * b4b9e1d87d8ea273613880de9d2b2f0b0b696244b42152bfa0a3106e7d983a20 lib/unicore/BidiMirroring.txt
+ * 529dc5d0f6386d52f2f56e004bbfab48ce2d587eea9d38ba546c4052491bd820 lib/unicore/Blocks.txt
+ * cdd49e55eae3bbf1f0a3f6580c974a0263cb86a6a08daa10fbf705b4808a56f7 lib/unicore/CaseFolding.txt
+ * 3b019c0a33c3140cbc920c078f4f9af2680ba4f71869c8d4de5190667c70b6a3 lib/unicore/CompositionExclusions.txt
+ * 7570877e0fa197c45338f7c41a02636da4e14c8dba6a3611a01cd30bf329d5ca lib/unicore/DAge.txt
+ * d367290bc0867e6b484c68370530bdd1a08b6b32404601b8c7accaf83e05628d lib/unicore/DCoreProperties.txt
+ * d5687a48c95c7d6e1ec59cb29c0f2e8b052018eb069a4371b7368d0561e12a29 lib/unicore/DNormalizationProps.txt
+ * 743e7bc435c04ab1a8459710b1c3cad56eedced5b806b4659b6e69b85d0adf2a lib/unicore/EastAsianWidth.txt
+ * f2e04bae8c856fad3a16353a99d4cc2de6c72770260379f5e4974a97548aad2a lib/unicore/EquivalentUnifiedIdeograph.txt
+ * 9a3ab36d36a22bdb84de7a17b17e9b9c242134f0080f0a8b4b28d209465a8fc8 lib/unicore/HangulSyllableType.txt
+ * 790bc9595795c0e0a3860a21a7f97157a134b61a4fc4ab03c7d315d07c9a6eb7 lib/unicore/IdStatus.txt
+ * 71d3ed8f15cd5d8cd00cdebe62015ff26356462774b261b4a2b83d3bf46b1639 lib/unicore/IdType.txt
+ * 0ce56c1294da405c0a0a0071582ac839fd229bbf97bdd260462ee571309d4ec4 lib/unicore/IndicPositionalCategory.txt
+ * ffae561a51b47ddbbe267fdd8505ac3776b85b2932268809127acee84200b573 lib/unicore/IndicSyllabicCategory.txt
+ * 14733bcb6731ae0c07485bf59a41cb3db08785a50bd2b46b836b4341eab7ee46 lib/unicore/Jamo.txt
+ * 012bca868e2c4e59a5a10a7546baf0c6fb1b2ef458c277f054915c8a49d292bf lib/unicore/LineBreak.txt
+ * 3e39509e8fae3e5d50ba73759d0b97194501d14a9c63107a6372a46b38be18e8 lib/unicore/NameAliases.txt
+ * 1d5202155f14841973aa540b1625f4befbde185ac77ce5aceaaaa0501a68bd66 lib/unicore/NamedSequences.txt
+ * fb9ac8cc154a80cad6caac9897af55a4e75176af6f4e2bb6edc2bf8b1d57f326 lib/unicore/NormTest.txt
+ * e05c0a2811d113dae4abd832884199a3ea8d187ee1b872d8240a788a96540bfd lib/unicore/PropList.txt
+ * 13a7666843abea5c6b7eb8c057c57ab9bb2ba96cfc936e204224dd67d71cafad lib/unicore/PropValueAliases.txt
+ * e4935149af407fa455901832b710bccb63d2453e46d09190e234d019bcfbba45 lib/unicore/PropertyAliases.txt
+ * 7e07313d9d0bee42220c476b64485995130ae30917bbcf7780b602d677d7e33f lib/unicore/ScriptExtensions.txt
+ * cca85d830f46aece2e7c1459ef1249993dca8f2e46d51e869255be140d7ea4b0 lib/unicore/Scripts.txt
+ * 78b29c64b5840d25c11a9f31b665ee551b8a499eca6c70d770fcad7dd710f494 lib/unicore/SpecialCasing.txt
+ * 806e9aed65037197f1ec85e12be6e8cd870fc5608b4de0fffd990f689f376a73 lib/unicore/UnicodeData.txt
+ * ca6d332f485a6f5f452b29b4a74146af0f2c17b7577aa4c821d597210f70611a lib/unicore/VerticalOrientation.txt
+ * 0d2080d0def294a4b7660801cc03ddfe5866ff300c789c2cc1b50fd7802b2d97 lib/unicore/auxiliary/GCBTest.txt
+ * 5a0f8748575432f8ff95e1dd5bfaa27bda1a844809e17d6939ee912bba6568a1 lib/unicore/auxiliary/GraphemeBreakProperty.txt
+ * 371bde4052aa593b108684ae292d8ea2dbb93c19990e0cdf416fa7239557aac3 lib/unicore/auxiliary/LBTest.txt
+ * f62279d8fd10935ba0cf0d8417a1dcbe7ab0d4e62f59c17e02cbe40f580c4162 lib/unicore/auxiliary/SBTest.txt
+ * 61e4ba975b0a5bc1a76ee931b94914395d7289ef624e3c0d4d6b9460ee387bea lib/unicore/auxiliary/SentenceBreakProperty.txt
+ * 2a676130c71194245e7c74a837e58330f202600d8ddcf4518129dd476f26e18e lib/unicore/auxiliary/WBTest.txt
+ * 5188a56e91593467c2e912601ebc78750e6adc9b04541b8c5becb5441e388ce2 lib/unicore/auxiliary/WordBreakProperty.txt
+ * 29071dba22c72c27783a73016afb8ffaeb025866740791f9c2d0b55cc45a3470 lib/unicore/emoji/emoji.txt
+ * 4841f2090c2dbc592d3ce43bb74c2191b3da50fb9a0d00274f1448c202851b02 lib/unicore/extracted/DBidiClass.txt
+ * f10a35451429137f7348825f22d624b6390c526ead3d8e756d2af9e5ed5b2b67 lib/unicore/extracted/DBinaryProperties.txt
+ * ca54f6360cd288ad92113415bf1f77749015abe11cbd6798d21f7fa81f04205d lib/unicore/extracted/DCombiningClass.txt
+ * db059ce45e3cec49bfda56e262fa658b3a5561b1648de266c818d2a08a85b78a lib/unicore/extracted/DDecompositionType.txt
+ * d62e6950f086e53f47c593a38342621f8838f48c49a1de070cf83d3959bd1688 lib/unicore/extracted/DEastAsianWidth.txt
+ * fe29a45c0882500e591140aaa5c4f5067e6a5d746806148af34400c48b9c06f9 lib/unicore/extracted/DGeneralCategory.txt
+ * e13ca1344b16023aa38c6ada39f9658536fc6bb7c3c24d579f0bc316a4f4f1e0 lib/unicore/extracted/DJoinGroup.txt
+ * c4870b11e2b8b7d0eb70b99ce85608e5c28a399efa316cca97238a58ae160e5e lib/unicore/extracted/DJoinType.txt
+ * 3f4f32ed2a577344a508114527e721d7a8b633d32f38945d47fe0c743650c585 lib/unicore/extracted/DLineBreak.txt
+ * 710abf2d581ac9c57f244c0834f9d9969d9781e0396adccd330eaae658ac7d6b lib/unicore/extracted/DNumType.txt
+ * 6bd30f385f3baf3ab5d5308c111a81de87bea5f494ba0ba69e8ab45263b8c34d lib/unicore/extracted/DNumValues.txt
+ * 8968a5ee00063fa31ff15474b95ccb6b9c228e3ebad6c20bc77f4225187e2023 lib/unicore/mktables
+ * 55d90fdc3f902e5c0b16b3378f9eaa36e970a1c09723c33de7d47d0370044012 lib/unicore/version
+ * 0a6b5ab33bb1026531f816efe81aea1a8ffcd34a27cbea37dd6a70a63d73c844 regen/charset_translations.pl
+ * c7ff8e0d207d3538c7feb4a1a152b159e5e902d20293b303569ea8323e84633e regen/mk_PL_charclass.pl
+ * e7ba2c6c0577fbb8b767a1305dbebcfeec166d11aa010cfbad9001c5f5971ee6 regen/mk_invlists.pl
+ * ex: set ro ft=c: */

regexp.h

@@ -31,7 +31,7 @@ my $tests = 28; # I can't see a clean way to calculate this automatically.
 
 my %skip = ("regen_perly.pl"    => [qw(perly.act perly.h perly.tab)],
             "regen/keywords.pl" => [qw(keywords.c keywords.h)],
-            "regen/mk_invlists.pl" => [qw(charclass_invlists.h uni_keywords.h)],
+            "regen/mk_invlists.pl" => [qw(charclass_invlists.h regexp_constants.h uni_keywords.h)],
             "regen/regcharclass.pl" => [qw(regcharclass.h)],
            );