[uniscribe] Use unique font name
When installing per-process fonts using AddFontMemResourceEx(), if a font with the same family name is already installed, sometimes that one gets used. Which is problematic for us. As such, we now mangle the font to install a new 'name' table with a unique name, which we then use to choose the font. Patch from Jonathan Kew.
This commit is contained in:
Behdad Esfahbod
parent
8751de5083
commit
8ac2e88def
|
|
@ -298,7 +298,7 @@ if test "x$with_uniscribe" = "xyes" -a "x$have_uniscribe" != "xtrue"; then
|
|||
fi
|
||||
if $have_uniscribe; then
|
||||
UNISCRIBE_CFLAGS=
|
||||
UNISCRIBE_LIBS="-lusp10 -lgdi32"
|
||||
UNISCRIBE_LIBS="-lusp10 -lgdi32 -lrpcrt4"
|
||||
AC_SUBST(UNISCRIBE_CFLAGS)
|
||||
AC_SUBST(UNISCRIBE_LIBS)
|
||||
AC_DEFINE(HAVE_UNISCRIBE, 1, [Have Uniscribe library])
|
||||
|
|
|
|||
|
|
@ -32,9 +32,11 @@
|
|||
|
||||
#include <windows.h>
|
||||
#include <usp10.h>
|
||||
#include <rpc.h>
|
||||
|
||||
#include "hb-uniscribe.h"
|
||||
|
||||
#include "hb-open-file-private.hh"
|
||||
#include "hb-ot-name-table.hh"
|
||||
#include "hb-ot-tag.h"
|
||||
|
||||
|
|
@ -267,8 +269,112 @@ HB_SHAPER_DATA_ENSURE_DECLARE(uniscribe, font)
|
|||
struct hb_uniscribe_shaper_face_data_t {
|
||||
HANDLE fh;
|
||||
hb_uniscribe_shaper_funcs_t *funcs;
|
||||
wchar_t face_name[LF_FACESIZE];
|
||||
};
|
||||
|
||||
static char *
|
||||
_hb_rename_font (const char *orig_sfnt_data, unsigned int length,
|
||||
unsigned int *new_length, wchar_t *new_name)
|
||||
{
|
||||
/* We'll create a private name for the font from a UUID using a simple,
|
||||
* somewhat base64-like encoding scheme */
|
||||
const char *enc = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+-";
|
||||
UUID id;
|
||||
UuidCreate ((UUID*) &id);
|
||||
unsigned int name_str_len = 0;
|
||||
new_name[name_str_len++] = 'F';
|
||||
new_name[name_str_len++] = '_';
|
||||
unsigned char *p = (unsigned char *) &id;
|
||||
for (unsigned int i = 0; i < 16; i += 2)
|
||||
{
|
||||
/* Spread the 16 bits from two bytes of the UUID across three chars of face_name,
|
||||
* using the bits in groups of 5,5,6 to select chars from enc.
|
||||
* This will generate 24 characters; with the 'F_' prefix we already provided,
|
||||
* the name will be 26 chars (plus the NUL terminator), so will always fit within
|
||||
* face_name (LF_FACESIZE = 32). */
|
||||
new_name[name_str_len++] = enc[p[i] >> 3];
|
||||
new_name[name_str_len++] = enc[((p[i] << 2) | (p[i + 1] >> 6)) & 0x1f];
|
||||
new_name[name_str_len++] = enc[p[i + 1] & 0x3f];
|
||||
}
|
||||
new_name[name_str_len] = 0;
|
||||
|
||||
/* Create a copy of the font data, with the 'name' table replaced by a table
|
||||
* that names the font with our private F_* name created above.
|
||||
* For simplicity, we just append a new 'name' table and update the sfnt directory;
|
||||
* the original table is left in place, but unused. */
|
||||
|
||||
/* The new table will contain just 5 name IDs: family, style, unique, full, PS.
|
||||
* All of them point to the same name data with our unique F_* name. */
|
||||
|
||||
static const uint16_t name_IDs[] = { 1, 2, 3, 4, 6 };
|
||||
|
||||
const unsigned int name_header_size = 3 * 2; /* NameTableHeader: USHORT x 3 */
|
||||
const unsigned int name_record_size = 6 * 2; /* NameRecord: USHORT x 6 */
|
||||
const unsigned int name_count = sizeof (name_IDs) / sizeof (name_IDs[0]);
|
||||
unsigned int name_table_length = name_header_size +
|
||||
name_count * name_record_size +
|
||||
name_str_len * 2; /* for name data in UTF16BE form */
|
||||
|
||||
unsigned int name_table_offset = (length + 3) & ~3;
|
||||
|
||||
*new_length = name_table_offset + (name_table_length + 3) & ~3;
|
||||
char *new_sfnt_data = (char *) calloc (1, *new_length);
|
||||
if (!new_sfnt_data)
|
||||
return NULL;
|
||||
|
||||
memcpy(new_sfnt_data, orig_sfnt_data, length);
|
||||
|
||||
unsigned char *name_table_data = (unsigned char *) (new_sfnt_data + name_table_offset);
|
||||
|
||||
p = name_table_data;
|
||||
*p++ = 0x00; *p++ = 0x00; /* format = 0 */
|
||||
*p++ = 0x00; *p++ = name_count; /* number of name records */
|
||||
*p++ = 0x00; *p++ = name_header_size + name_count * name_record_size; /* offset to string data */
|
||||
|
||||
for (unsigned int i = 0; i < name_count; i++) {
|
||||
*p++ = 0x00; *p++ = 0x03; /* platformID */
|
||||
*p++ = 0x00; *p++ = 0x01; /* encodingID */
|
||||
*p++ = 0x04; *p++ = 0x09; /* languageID = 0x0409 */
|
||||
*p++ = 0x00; *p++ = name_IDs[i]; /* nameID */
|
||||
*p++ = 0x00; *p++ = name_str_len * 2; /* string length (bytes) */
|
||||
*p++ = 0x00; *p++ = 0x00; /* string offset */
|
||||
}
|
||||
|
||||
for (unsigned int i = 0; i < name_str_len; i++) {
|
||||
/* copy string data from face_name, converting wchar_t to UTF16BE */
|
||||
*p++ = new_name[i] >> 8;
|
||||
*p++ = new_name[i] & 0xff;
|
||||
}
|
||||
|
||||
/* calculate new name table checksum */
|
||||
uint32_t checksum = 0;
|
||||
while (name_table_data < p) {
|
||||
checksum += (name_table_data[0] << 24) +
|
||||
(name_table_data[1] << 16) +
|
||||
(name_table_data[2] << 8) +
|
||||
name_table_data[3];
|
||||
name_table_data += 4;
|
||||
}
|
||||
|
||||
/* adjust name table entry to point to new name table */
|
||||
OT::OpenTypeFontFace *face = (OT::OpenTypeFontFace *) (new_sfnt_data);
|
||||
unsigned int index;
|
||||
if (face->find_table_index (HB_OT_TAG_name, &index))
|
||||
{
|
||||
const OT::TableRecord& record = face->get_table (index);
|
||||
OT::TableRecord *rp = const_cast<OT::TableRecord *> (&record);
|
||||
rp->checkSum.set(checksum);
|
||||
rp->offset.set(name_table_offset);
|
||||
rp->length.set(name_table_length);
|
||||
}
|
||||
|
||||
/* The checkSumAdjustment field in the 'head' table is now wrong,
|
||||
but AFAIK that doesn't actually cause any problems so I haven't
|
||||
bothered to fix it. */
|
||||
|
||||
return new_sfnt_data;
|
||||
}
|
||||
|
||||
hb_uniscribe_shaper_face_data_t *
|
||||
_hb_uniscribe_shaper_face_data_create (hb_face_t *face)
|
||||
{
|
||||
|
|
@ -289,9 +395,13 @@ _hb_uniscribe_shaper_face_data_create (hb_face_t *face)
|
|||
if (unlikely (!blob_length))
|
||||
DEBUG_MSG (UNISCRIBE, face, "Face has empty blob");
|
||||
|
||||
DWORD num_fonts_installed;
|
||||
data->fh = AddFontMemResourceEx ((void *) blob_data, blob_length, 0, &num_fonts_installed);
|
||||
unsigned int new_length;
|
||||
char *new_font_data = _hb_rename_font (blob_data, blob_length, &new_length, data->face_name);
|
||||
hb_blob_destroy (blob);
|
||||
|
||||
DWORD num_fonts_installed;
|
||||
data->fh = AddFontMemResourceEx (new_font_data, new_length, 0, &num_fonts_installed);
|
||||
free (new_font_data);
|
||||
if (unlikely (!data->fh)) {
|
||||
DEBUG_MSG (UNISCRIBE, face, "Face AddFontMemResourceEx() failed");
|
||||
free (data);
|
||||
|
|
@ -328,26 +438,10 @@ populate_log_font (LOGFONTW *lf,
|
|||
lf->lfHeight = -font->y_scale;
|
||||
lf->lfCharSet = DEFAULT_CHARSET;
|
||||
|
||||
hb_blob_t *blob = OT::Sanitizer<OT::name>::sanitize (hb_face_reference_table (font->face, HB_TAG ('n','a','m','e')));
|
||||
const OT::name *name_table = OT::Sanitizer<OT::name>::lock_instance (blob);
|
||||
unsigned int len = name_table->get_name (3, 1, 0x409, 4,
|
||||
lf->lfFaceName,
|
||||
sizeof (lf->lfFaceName[0]) * LF_FACESIZE)
|
||||
/ sizeof (lf->lfFaceName[0]);
|
||||
hb_blob_destroy (blob);
|
||||
hb_face_t *face = font->face;
|
||||
hb_uniscribe_shaper_face_data_t *face_data = HB_SHAPER_DATA_GET (face);
|
||||
|
||||
if (unlikely (!len)) {
|
||||
DEBUG_MSG (UNISCRIBE, NULL, "Didn't find English name table entry");
|
||||
return false;
|
||||
}
|
||||
if (unlikely (len >= LF_FACESIZE)) {
|
||||
DEBUG_MSG (UNISCRIBE, NULL, "Font name too long");
|
||||
return false;
|
||||
}
|
||||
|
||||
for (unsigned int i = 0; i < len; i++)
|
||||
lf->lfFaceName[i] = hb_be_uint16 (lf->lfFaceName[i]);
|
||||
lf->lfFaceName[len] = 0;
|
||||
memcpy (lf->lfFaceName, face_data->face_name, sizeof (lf->lfFaceName));
|
||||
|
||||
return true;
|
||||
}
|
||||
|
|
|
|||
Loading…
Reference in New Issue