Bump Wuffs, add TGA support through it
This commit is contained in:
10
fiv-io.c
10
fiv-io.c
@@ -74,6 +74,7 @@
|
||||
#define WUFFS_CONFIG__MODULE__GIF
|
||||
#define WUFFS_CONFIG__MODULE__LZW
|
||||
#define WUFFS_CONFIG__MODULE__PNG
|
||||
#define WUFFS_CONFIG__MODULE__TGA
|
||||
#define WUFFS_CONFIG__MODULE__ZLIB
|
||||
#include "wuffs-mirror-release-c/release/c/wuffs-v0.3.c"
|
||||
|
||||
@@ -2403,7 +2404,7 @@ fiv_io_open(
|
||||
{
|
||||
// TODO(p): Don't always load everything into memory, test type first,
|
||||
// so that we can reject non-pictures early. Wuffs only needs the first
|
||||
// 16 bytes (soon 12) to make a guess right now.
|
||||
// 17 bytes to make a guess right now.
|
||||
//
|
||||
// LibRaw poses an issue--there is no good registry for identification
|
||||
// of supported files. Many of them are compliant TIFF files.
|
||||
@@ -2436,7 +2437,7 @@ fiv_io_open_from_data(const char *data, size_t len, const gchar *uri,
|
||||
wuffs_base__make_slice_u8((uint8_t *) data, len);
|
||||
|
||||
cairo_surface_t *surface = NULL;
|
||||
switch (wuffs_base__magic_number_guess_fourcc(prefix)) {
|
||||
switch (wuffs_base__magic_number_guess_fourcc(prefix, true /* closed */)) {
|
||||
case WUFFS_BASE__FOURCC__BMP:
|
||||
// Note that BMP can redirect into another format,
|
||||
// which is so far unsupported here.
|
||||
@@ -2454,6 +2455,11 @@ fiv_io_open_from_data(const char *data, size_t len, const gchar *uri,
|
||||
wuffs_png__decoder__alloc_as__wuffs_base__image_decoder, data, len,
|
||||
profile, error);
|
||||
break;
|
||||
case WUFFS_BASE__FOURCC__TGA:
|
||||
surface = open_wuffs_using(
|
||||
wuffs_tga__decoder__alloc_as__wuffs_base__image_decoder, data, len,
|
||||
profile, error);
|
||||
break;
|
||||
case WUFFS_BASE__FOURCC__JPEG:
|
||||
surface = enhance
|
||||
? open_libjpeg_enhanced(data, len, profile, error)
|
||||
|
||||
Reference in New Issue
Block a user