SDL/build-scripts/makecasefoldhashtable.pl

323 lines
9.7 KiB
Perl
Raw Permalink Normal View History

stdlib: Improve Unicode support and consistency in string comparison functions. SDL_strcasecmp (even when calling into a C runtime) does not work with Unicode chars, and depending on the user's locale, might not work with even basic ASCII strings. This implements the function from scratch, using "case-folding," which is a more robust method that deals with various languages. It involves a hashtable of a few hundred codepoints that are "uppercase" and how to map them to lowercase equivalents (possibly increasing the size of the string in the process). The vast majority of human languages (and Unicode) do not have letters with different cases, but still, this static table takes about 10 kilobytes on a 64-bit machine. Even this will fail in one known case: the Turkish 'i' folds differently if you're writing in Turkish vs other languages. Generally this is seen as unfortunate collateral damage in cases where you can't specify the language in use. In addition to case-folding the codepoints, the new functions also know how to decode the various formats to turn them into codepoints in the first place, instead of blindly stepping by one byte (or one wchar_t) per character. Also included is casefolding.txt from the Unicode Consortium and a perl script to generate the hashtable from that text file, so we can trivially update this if new languages are added in the future. A simple test using the new function: ```c #include <SDL3/SDL.h> int main(void) { const char *a = "α ε η"; const char *b = "Α Ε Η"; SDL_Log(" strcasecmp(\"%s\", \"%s\") == %d\n", a, b, strcasecmp(a, b)); SDL_Log("SDL_strcasecmp(\"%s\", \"%s\") == %d\n", a, b, SDL_strcasecmp(a, b)); return 0; } ``` Produces: ``` INFO: strcasecmp("α ε η", "Α Ε Η") == 32 INFO: SDL_strcasecmp("α ε η", "Α Ε Η") == 0 ``` glibc strcasecmp() fails to compare a Greek lowercase string to its uppercase equivalent, even with a UTF-8 locale, but SDL_strcasecmp() works. Other SDL_stdinc.h functions are changed to be more consistent, which is to say they now ignore any C runtime and often dictate that only English-based low-ASCII works with them. Fixes Issue #9313.
2024-03-26 11:22:38 -06:00
#!/usr/bin/perl -w
# Simple DirectMedia Layer
# Copyright (C) 1997-2024 Sam Lantinga <slouken@libsdl.org>
#
# This software is provided 'as-is', without any express or implied
# warranty. In no event will the authors be held liable for any damages
# arising from the use of this software.
#
# Permission is granted to anyone to use this software for any purpose,
# including commercial applications, and to alter it and redistribute it
# freely, subject to the following restrictions:
#
# 1. The origin of this software must not be misrepresented; you must not
# claim that you wrote the original software. If you use this software
# in a product, an acknowledgment in the product documentation would be
# appreciated but is not required.
# 2. Altered source versions must be plainly marked as such, and must not be
# misrepresented as being the original software.
# 3. This notice may not be removed or altered from any source distribution.
# This script was originally written by Ryan C. Gordon for PhysicsFS
# ( https://icculus.org/physfs/ ), under the zlib license: the same license
# that SDL itself uses).
use warnings;
use strict;
my $HASHBUCKETS1_16 = 256;
my $HASHBUCKETS1_32 = 16;
my $HASHBUCKETS2_16 = 16;
my $HASHBUCKETS3_16 = 4;
my $mem_used = 0;
print <<__EOF__;
/*
Simple DirectMedia Layer
Copyright (C) 1997-2024 Sam Lantinga <slouken\@libsdl.org>
This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
*/
/*
* This data was generated by SDL/build-scripts/makecasefoldhashtable.pl
*
* Do not manually edit this file!
*/
#ifndef SDL_casefolding_h_
#define SDL_casefolding_h_
/* We build three simple hashmaps here: one that maps Unicode codepoints to
a one, two, or three lowercase codepoints. To retrieve this info: look at
case_fold_hashX, where X is 1, 2, or 3. Most foldable codepoints fold to one,
a few dozen fold to two, and a handful fold to three. If the codepoint isn't
in any of these hashes, it doesn't fold (no separate upper and lowercase).
Almost all these codepoints fit into 16 bits, so we hash them as such to save
memory. If a codepoint is > 0xFFFF, we have separate hashes for them,
since there are (currently) only about 120 of them and (currently) all of them
map to a single lowercase codepoint. */
typedef struct CaseFoldMapping1_32
{
Uint32 from;
Uint32 to0;
} CaseFoldMapping1_32;
typedef struct CaseFoldMapping1_16
{
Uint16 from;
Uint16 to0;
} CaseFoldMapping1_16;
typedef struct CaseFoldMapping2_16
{
Uint16 from;
Uint16 to0;
Uint16 to1;
} CaseFoldMapping2_16;
typedef struct CaseFoldMapping3_16
{
Uint16 from;
Uint16 to0;
Uint16 to1;
Uint16 to2;
} CaseFoldMapping3_16;
typedef struct CaseFoldHashBucket1_16
{
const CaseFoldMapping1_16 *list;
const Uint8 count;
} CaseFoldHashBucket1_16;
typedef struct CaseFoldHashBucket1_32
{
const CaseFoldMapping1_32 *list;
const Uint8 count;
} CaseFoldHashBucket1_32;
typedef struct CaseFoldHashBucket2_16
{
const CaseFoldMapping2_16 *list;
const Uint8 count;
} CaseFoldHashBucket2_16;
typedef struct CaseFoldHashBucket3_16
{
const CaseFoldMapping3_16 *list;
const Uint8 count;
} CaseFoldHashBucket3_16;
__EOF__
my @foldPairs1_16;
my @foldPairs2_16;
my @foldPairs3_16;
my @foldPairs1_32;
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
$foldPairs1_16[$i] = '';
}
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
$foldPairs1_32[$i] = '';
}
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
$foldPairs2_16[$i] = '';
}
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
$foldPairs3_16[$i] = '';
}
open(FH,'<','casefolding.txt') or die("failed to open casefolding.txt: $!\n");
while (<FH>) {
chomp;
# strip comments from textfile...
s/\#.*\Z//;
# strip whitespace...
s/\A\s+//;
s/\s+\Z//;
next if not /\A([a-fA-F0-9]+)\;\s*(.)\;\s*(.+)\;/;
my ($code, $status, $mapping) = ($1, $2, $3);
my $hexxed = hex($code);
#print("// code '$code' status '$status' mapping '$mapping'\n");
if (($status eq 'C') or ($status eq 'F')) {
my ($map1, $map2, $map3) = (undef, undef, undef);
$map1 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
$map2 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
$map3 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
die("mapping space too small for '$code'\n") if ($mapping ne '');
die("problem parsing mapping for '$code'\n") if (not defined($map1));
if ($hexxed < 128) {
# Just ignore these, we'll handle the low-ASCII ones ourselves.
} elsif ($hexxed > 0xFFFF) {
# We just need to add the 32-bit 2 and/or 3 codepoint maps if this die()'s here.
die("Uhoh, a codepoint > 0xFFFF that folds to multiple codepoints! Fixme.") if defined($map2);
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS1_32-1));
#print("// hexxed '$hexxed' hashed1 '$hashed'\n");
$foldPairs1_32[$hashed] .= " { 0x$code, 0x$map1 },\n";
$mem_used += 8;
} elsif (not defined($map2)) {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS1_16-1));
#print("// hexxed '$hexxed' hashed1 '$hashed'\n");
$foldPairs1_16[$hashed] .= " { 0x$code, 0x$map1 },\n";
$mem_used += 4;
} elsif (not defined($map3)) {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS2_16-1));
#print("// hexxed '$hexxed' hashed2 '$hashed'\n");
$foldPairs2_16[$hashed] .= " { 0x$code, 0x$map1, 0x$map2 },\n";
$mem_used += 6;
} else {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS3_16-1));
#print("// hexxed '$hexxed' hashed3 '$hashed'\n");
$foldPairs3_16[$hashed] .= " { 0x$code, 0x$map1, 0x$map2, 0x$map3 },\n";
$mem_used += 8;
}
}
}
close(FH);
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
$foldPairs1_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs1_16[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_16_${num}";
print("static const CaseFoldMapping1_16 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
$foldPairs1_32[$i] =~ s/,\n\Z//;
my $str = $foldPairs1_32[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_32_${num}";
print("static const CaseFoldMapping1_32 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
$foldPairs2_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs2_16[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold2_16_${num}";
print("static const CaseFoldMapping2_16 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
$foldPairs3_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs3_16[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold3_16_${num}";
print("static const CaseFoldMapping3_16 ${sym}[] = {\n$str\n};\n\n");
}
print("static const CaseFoldHashBucket1_16 case_fold_hash1_16[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
my $str = $foldPairs1_16[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_16_${num}";
print(" { $sym, SDL_arraysize($sym) },\n");
}
$mem_used += 12;
}
print("};\n\n");
print("static const CaseFoldHashBucket1_32 case_fold_hash1_32[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
my $str = $foldPairs1_32[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_32_${num}";
print(" { $sym, SDL_arraysize($sym) },\n");
}
$mem_used += 12;
}
print("};\n\n");
print("static const CaseFoldHashBucket2_16 case_fold_hash2_16[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
my $str = $foldPairs2_16[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold2_16_${num}";
print(" { $sym, SDL_arraysize($sym) },\n");
}
$mem_used += 12;
}
print("};\n\n");
print("static const CaseFoldHashBucket3_16 case_fold_hash3_16[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
my $str = $foldPairs3_16[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold3_16_${num}";
print(" { $sym, SDL_arraysize($sym) },\n");
}
$mem_used += 12;
}
print("};\n\n");
print <<__EOF__;
#endif /* SDL_casefolding_h_ */
__EOF__
print STDERR "Memory required for case-folding hashtable: $mem_used bytes\n";
exit 0;
# end of makecashfoldhashtable.pl ...