marko10_000
/
wine-wine
forked from Mirrors/wine-wine
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

winedump: Add dumping of sortkey NLS files. 

Signed-off-by: Alexandre Julliard <julliard@winehq.org>
feature/deterministic
Alexandre Julliard 2020-03-13 10:14:49 +01:00
parent 602f20826d
commit 8207bdfecb
3 changed files with 409 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
tools/winedump/dump.c
View File

@ -207,6 +207,34 @@ const char* get_guid_str(const GUID* guid)
return str;
}
const char *get_unicode_str( const WCHAR *str, int len )
{
char *buffer;
int i = 0;
if (len == -1) len = strlenW( str );
buffer = dump_want_n( len * 6 + 3);
buffer[i++] = '"';
while (len-- > 0 && *str)
{
WCHAR c = *str++;
switch (c)
{
case '\n': strcpy( buffer + i, "\\n" ); i += 2; break;
case '\r': strcpy( buffer + i, "\\r" ); i += 2; break;
case '\t': strcpy( buffer + i, "\\t" ); i += 2; break;
case '"': strcpy( buffer + i, "\\\"" ); i += 2; break;
case '\\': strcpy( buffer + i, "\\\\" ); i += 2; break;
default:
if (c >= ' ' && c <= 126) buffer[i++] = c;
else i += sprintf( buffer + i, "\\u%04x",c);
}
}
buffer[i++] = '"';
buffer[i] = 0;
return buffer;
}
const void* PRD(unsigned long prd, unsigned long len)
{
return (prd + len > dump_total_len) ? NULL : (const char*)dump_base + prd;

396
tools/winedump/nls.c
View File

@ -35,7 +35,7 @@ static const void *read_data( unsigned int *pos, unsigned int size )
return ret;
}
static unsigned short casemap( const unsigned short *table, unsigned int len, unsigned short ch )
static unsigned short mapchar( const unsigned short *table, unsigned int len, unsigned short ch )
{
unsigned int off = table[ch >> 8] + ((ch >> 4) & 0x0f);
if (off >= len) return 0;
@ -44,9 +44,64 @@ static unsigned short casemap( const unsigned short *table, unsigned int len, un
return ch + table[off];
}
static void dump_casemap(void)
static void dump_offset_table( const unsigned short *table, unsigned int len )
{
int i, ch;
for (i = 0; i < 0x10000; i++)
{
if (!(i % 16)) printf( "\n%04x:", i );
ch = mapchar( table, len, i );
if (ch == i) printf( " ...." );
else printf( " %04x", ch );
}
}
struct ctype
{
WORD c1, c2, c3;
};
static const char *get_ctype( const struct ctype *ctype )
{
static char buffer[100];
static const char *c1[] = { "up ", "lo ", "dg ", "sp ", "pt ", "cl ", "bl ", "xd ", "al " };
static const char *c2[] = { " ", "L ", "R ", "EN", "ES", "ET",
"AN", "CS", "B ", "S ", "WS", "ON" };
static const char *c3[] = { "ns ", "di ", "vo ", "sy ", "ka ", "hi ", "hw ", "fw ",
"id ", "ks ", "lx ", "hi ", "lo ", " ", " ", "al " };
int i;
strcpy( buffer, "| " );
for (i = 0; i < ARRAY_SIZE(c1); i++)
strcat( buffer, (ctype->c1 & (1 << i)) ? c1[i] : "__ " );
strcat( buffer, "| " );
strcat( buffer, ctype->c2 < ARRAY_SIZE(c2) ? c2[ctype->c2] : "??" );
strcat( buffer, " | " );
for (i = 0; i < ARRAY_SIZE(c3); i++)
strcat( buffer, (ctype->c3 & (1 << i)) ? c3[i] : "__ " );
strcat( buffer, "|" );
return buffer;
}
static void dump_ctype_table( const USHORT *ptr )
{
const struct ctype *ctypes = (const struct ctype *)(ptr + 2);
const BYTE *types = (const BYTE *)ptr + ptr[1] + 2;
int i, len = (ptr[1] - 2) / sizeof(*ctypes);
printf( " CTYPE1 CTYPE2 CTYPE3\n" );
for (i = 0; i < 0x10000; i++)
{
const BYTE *b = types + ((const WORD *)types)[i >> 8];
b = types + ((const WORD *)b)[(i >> 4) & 0x0f] + (i & 0x0f);
if (*b < len) printf( "%04x %s\n", i, get_ctype( ctypes + *b ));
else printf( "%04x ??? %02x\n", i, *b );
}
printf( "\n" );
}
static void dump_casemap(void)
{
unsigned int pos = 0, upper_len, lower_len;
const unsigned short *header, *upper, *lower;
@ -62,21 +117,9 @@ static void dump_casemap(void)
printf( "Magic: %04x\n", header[0] );
printf( "Upper-case table:\n" );
for (i = 0; i < 0x10000; i++)
{
if (!(i % 16)) printf( "\n%04x:", i );
ch = casemap( upper, upper_len, i );
if (ch == i) printf( " ...." );
else printf( " %04x", ch );
}
dump_offset_table( upper, upper_len );
printf( "\n\nLower-case table:\n" );
for (i = 0; i < 0x10000; i++)
{
if (!(i % 16)) printf( "\n%04x:", i );
ch = casemap( lower, lower_len, i );
if (ch == i) printf( " ...." );
else printf( " %04x", ch );
}
dump_offset_table( lower, lower_len );
printf( "\n\n" );
}
@ -346,6 +389,325 @@ static void dump_norm(void)
printf( "\n" );
}
struct sortguid
{
GUID id; /* sort GUID */
DWORD flags; /* flags */
DWORD compr; /* offset to compression table */
DWORD except; /* exception table offset in sortkey table */
DWORD ling_except; /* exception table offset for linguistic casing */
DWORD casemap; /* linguistic casemap table offset */
};
#define FLAG_HAS_3_BYTE_WEIGHTS 0x01
#define FLAG_REVERSEDIACRITICS 0x10
#define FLAG_DOUBLECOMPRESSION 0x20
#define FLAG_INVERSECASING 0x40
struct language_id
{
DWORD offset;
WCHAR name[32];
};
struct compression
{
DWORD offset;
WCHAR minchar, maxchar;
WORD len[8];
};
struct comprlang
{
struct compression compr;
WCHAR name[32];
};
static const char *get_sortkey( DWORD key )
{
static char buffer[16];
if (!key) return "....";
if ((WORD)key == 0x200)
sprintf( buffer, "expand %04x", key >> 16 );
else
sprintf( buffer, "%u.%u.%u.%u", (BYTE)(key >> 8), (BYTE)key, (BYTE)(key >> 16), (BYTE)(key >> 24) );
return buffer;
}
static const void *dump_expansions( const DWORD *ptr )
{
DWORD i, count = *ptr++;
printf( "\nExpansions: (count=%04x)\n\n", count );
for (i = 0; i < count; i++)
{
const WCHAR *p = (const WCHAR *)(ptr + i);
printf( " %04x: %04x %04x\n", i, p[0], p[1] );
}
return ptr + count;
}
static void dump_exceptions( const DWORD *sortkeys, DWORD offset )
{
int i, j;
const DWORD *table = sortkeys + offset;
for (i = 0; i < 0x100; i++)
{
if (table[i] == i * 0x100) continue;
for (j = 0; j < 0x100; j++)
{
if (sortkeys[table[i] + j] == sortkeys[i * 0x100 + j]) continue;
printf( " %04x: %s\n", i * 0x100 + j, get_sortkey( sortkeys[table[i] + j] ));
}
}
}
static const void *dump_compression( const struct compression *compr, const WCHAR *table )
{
int i, j, k;
const WCHAR *p = table + compr->offset;
printf( " min=%04x max=%04x counts=%u,%u,%u,%u,%u,%u,%u,%u\n",
compr->minchar, compr->maxchar,
compr->len[0], compr->len[1], compr->len[2], compr->len[3],
compr->len[4], compr->len[5], compr->len[6], compr->len[7] );
for (i = 0; i < 8; i++)
{
for (j = 0; j < compr->len[i]; j++)
{
printf( " " );
for (k = 0; k < i + 2; k++) printf( " %04x", *p++ );
p = (const WCHAR *)(((ULONG_PTR)p + 3) & ~3);
printf( " -> %s\n", get_sortkey( *(const DWORD *)p ));
p += 2;
}
}
return p;
}
static const void *dump_multiple_weights( const DWORD *ptr )
{
int i, count = *ptr++;
const WCHAR *p;
printf( "\nMultiple weights: (count=%u)\n\n", count );
p = (const WCHAR *)ptr;
for (i = 0; i < count; i++)
{
BYTE weight = p[i];
BYTE count = p[i] >> 8;
printf( "%u - %u\n", weight, weight + count );
}
return ptr + (count + 1) / 2;
}
static void dump_sort( int old_version )
{
const struct
{
DWORD sortkeys;
DWORD casemaps;
DWORD ctypes;
DWORD sortids;
} *header;
const struct compression *compr;
const struct sortguid *guids;
const struct comprlang *comprlangs;
const struct language_id *language_ids = NULL;
const WORD *casemaps, *map;
const DWORD *sortkeys, *ptr;
const WCHAR *p = NULL;
int i, j, size, len;
int nb_casemaps = 0, casemap_offsets[16];
if (!(header = PRD( 0, sizeof(*header) ))) return;
if (!(sortkeys = PRD( header->sortkeys, header->casemaps - header->sortkeys ))) return;
printf( "\nSort keys:\n" );
for (i = 0; i < 0x10000; i++)
{
if (!(i % 8)) printf( "\n%04x:", i );
printf( " %16s", get_sortkey( sortkeys[i] ));
}
printf( "\n\n" );
size = (header->ctypes - header->casemaps) / sizeof(*casemaps);
if (!(casemaps = PRD( header->casemaps, size * sizeof(*casemaps) ))) return;
len = 0;
if (old_version)
{
ptr = (const DWORD *)casemaps;
len = *ptr++;
language_ids = (const struct language_id *)ptr;
casemaps = (const WORD *)(language_ids + len);
}
map = casemaps;
while (size)
{
const WORD *upper = map + 2;
const WORD *lower = map + 2 + map[1];
const WORD *end = map + map[1] + 1 + map[map[1] + 1];
if (map[0] != 1) break;
printf( "\nCase mapping table %u:\n", nb_casemaps );
casemap_offsets[nb_casemaps++] = map - casemaps;
for (j = 0; j < len; j++)
{
if (language_ids[j].offset != map - casemaps) continue;
printf( "Language: %s\n", get_unicode_str( language_ids[j].name, -1 ));
break;
}
printf( "\nUpper-case table:\n" );
dump_offset_table( upper, lower - upper );
printf( "\n\nLower-case table:\n" );
dump_offset_table( lower, end - lower );
printf( "\n\n" );
size -= (end - map);
map = end;
}
if (!(p = PRD( header->ctypes, header->sortids - header->ctypes ))) return;
printf( "\nCTYPE table:\n\n" );
dump_ctype_table( p );
printf( "\nSort tables:\n\n" );
size = (dump_total_len - header->sortids) / sizeof(*ptr);
if (!(ptr = PRD( header->sortids, size * sizeof(*ptr) ))) return;
if (old_version)
{
len = *ptr++;
for (i = 0; i < len; i++, ptr += 2) printf( "NLS version: %08x %08x\n", ptr[0], ptr[1] );
len = *ptr++;
for (i = 0; i < len; i++, ptr += 2) printf( "Defined version: %08x %08x\n", ptr[0], ptr[1] );
len = *ptr++;
printf( "\nReversed diacritics:\n\n" );
for (i = 0; i < len; i++)
{
const WCHAR *name = (const WCHAR *)ptr;
printf( "%s\n", get_unicode_str( name, -1 ));
ptr += 16;
}
len = *ptr++;
printf( "\nDouble compression:\n\n" );
for (i = 0; i < len; i++)
{
const WCHAR *name = (const WCHAR *)ptr;
printf( "%s\n", get_unicode_str( name, -1 ));
ptr += 16;
}
ptr = dump_expansions( ptr );
printf( "\nCompressions:\n" );
size = *ptr++;
comprlangs = (const struct comprlang *)ptr;
for (i = 0; i < size; i++)
{
printf( "\n %s\n", get_unicode_str( comprlangs[i].name, -1 ));
ptr = dump_compression( &comprlangs[i].compr, (const WCHAR *)(comprlangs + size) );
}
ptr = dump_multiple_weights( ptr );
size = *ptr++;
printf( "\nJamo sort:\n\n" );
for (i = 0; i < size; i++, ptr += 2)
{
const struct jamo { BYTE val[5], off, len; } *jamo = (const struct jamo *)ptr;
printf( "%04x: %02x %02x %02x %02x %02x off=%02x len=%02x\n", 0x1100 + i, jamo->val[0],
jamo->val[1], jamo->val[2], jamo->val[3], jamo->val[4],
jamo->off, jamo->len );
}
size = *ptr++;
printf( "\nJamo second chars:\n\n" );
for (i = 0; i < size; i++, ptr += 2)
{
const struct jamo { WORD ch; BYTE val[5], len; } *jamo = (const struct jamo *)ptr;
printf( "%02x: %04x: %02x %02x %02x %02x %02x len=%02x\n", i, jamo->ch, jamo->val[0],
jamo->val[1], jamo->val[2], jamo->val[3], jamo->val[4], jamo->len );
}
size = *ptr++;
printf( "\nExceptions:\n" );
language_ids = (const struct language_id *)ptr;
for (i = 0; i < size; i++)
{
printf( "\n %08x %s\n", language_ids[i].offset, get_unicode_str( language_ids[i].name, -1 ));
dump_exceptions( sortkeys, language_ids[i].offset );
}
}
else
{
int guid_count = ptr[1];
printf( "NLS version: %08x\n\n", ptr[0] );
printf( "Sort GUIDs:\n\n" );
guids = (const struct sortguid *)(ptr + 2);
for (i = 0; i < guid_count; i++)
{
for (j = 0; j < nb_casemaps; j++) if (casemap_offsets[j] == guids[i].casemap) break;
printf( " %s flags=%08x compr=%08x casemap=%d\n", get_guid_str( &guids[i].id ),
guids[i].flags, guids[i].compr, j < nb_casemaps ? j : -1 );
}
ptr = dump_expansions( (const DWORD *)(guids + guid_count) );
size = *ptr++;
printf( "\nCompressions:\n" );
compr = (const struct compression *)ptr;
for (i = 0; i < size; i++)
{
printf( "\n" );
for (j = 0; j < guid_count; j++)
if (guids[j].compr == i) printf( " %s\n", get_guid_str( &guids[j].id ));
ptr = dump_compression( compr + i, (const WCHAR *)(compr + size) );
}
ptr = dump_multiple_weights( ptr );
size = *ptr++;
printf( "\nJamo sort:\n\n" );
for (i = 0; i < size; i++)
{
static const WCHAR hangul_chars[] =
{
0xa960, 0xa961, 0xa962, 0xa963, 0xa964, 0xa965, 0xa966, 0xa967,
0xa968, 0xa969, 0xa96a, 0xa96b, 0xa96c, 0xa96d, 0xa96e, 0xa96f,
0xa970, 0xa971, 0xa972, 0xa973, 0xa974, 0xa975, 0xa976, 0xa977,
0xa978, 0xa979, 0xa97a, 0xa97b, 0xa97c,
0xd7b0, 0xd7b1, 0xd7b2, 0xd7b3, 0xd7b4, 0xd7b5, 0xd7b6, 0xd7b7,
0xd7b8, 0xd7b9, 0xd7ba, 0xd7bb, 0xd7bc, 0xd7bd, 0xd7be, 0xd7bf,
0xd7c0, 0xd7c1, 0xd7c2, 0xd7c3, 0xd7c4, 0xd7c5, 0xd7c6,
0xd7cb, 0xd7cc, 0xd7cd, 0xd7ce, 0xd7cf,
0xd7d0, 0xd7d1, 0xd7d2, 0xd7d3, 0xd7d4, 0xd7d5, 0xd7d6, 0xd7d7,
0xd7d8, 0xd7d9, 0xd7da, 0xd7db, 0xd7dc, 0xd7dd, 0xd7de, 0xd7df,
0xd7e0, 0xd7e1, 0xd7e2, 0xd7e3, 0xd7e4, 0xd7e5, 0xd7e6, 0xd7e7,
0xd7e8, 0xd7e9, 0xd7ea, 0xd7eb, 0xd7ec, 0xd7ed, 0xd7ee, 0xd7ef,
0xd7f0, 0xd7f1, 0xd7f2, 0xd7f3, 0xd7f4, 0xd7f5, 0xd7f6, 0xd7f7,
0xd7f8, 0xd7f9, 0xd7fa, 0xd7fb
};
const BYTE *b = (const BYTE *)(ptr + 2 * i);
WCHAR wc = i < 0x100 ? 0x1100 + i : hangul_chars[i - 0x100];
printf( "%04x: %02x %02x %02x %02x %02x\n", wc, b[0], b[1], b[2], b[3], b[4] );
}
printf( "\nExceptions:\n" );
for (i = 0; i < guid_count; i++)
{
if (!guids[i].except) continue;
printf( "\n %s\n", get_guid_str( &guids[i].id ));
dump_exceptions( sortkeys, guids[i].except );
if (!guids[i].ling_except) continue;
printf( "\n %s LINGUISTIC_CASING\n", get_guid_str( &guids[i].id ));
dump_exceptions( sortkeys, guids[i].ling_except );
}
}
printf( "\n" );
}
void nls_dump(void)
{
const char *name = strrchr( globals.input_name, '/' );
@ -354,6 +716,8 @@ void nls_dump(void)
if (!strcasecmp( name, "l_intl.nls" )) return dump_casemap();
if (!strncasecmp( name, "c_", 2 )) return dump_codepage();
if (!strncasecmp( name, "norm", 4 )) return dump_norm();
if (!strcasecmp( name, "sortdefault.nls" )) return dump_sort( 0 );
if (!strncasecmp( name, "sort", 4 )) return dump_sort( 1 );
fprintf( stderr, "Unrecognized file name '%s'\n", globals.input_name );
}

1
tools/winedump/winedump.h
View File

@ -229,6 +229,7 @@ const char* get_time_str( unsigned long );
unsigned int strlenW( const unsigned short *str );
void dump_unicode_str( const unsigned short *str, int len );
const char* get_guid_str(const GUID* guid);
const char* get_unicode_str( const WCHAR *str, int len );
const char* get_symbol_str(const char* symname);
void print_fake_dll(void);
void dump_file_header(const IMAGE_FILE_HEADER *);