mewin/libpng
1
0
Fork 0
mirror of https://git.code.sf.net/p/libpng/code.git synced 2025-07-10 18:04:09 +02:00
Code Issues Projects Releases Wiki Activity

[libpng15] Fixed png_save_int_32 when int is not 2's complement (John Bowler).

Browse Source
This commit is contained in:
John Bowler 2015-08-21 14:50:38 -05:00 committed by Glenn Randers-Pehrson
parent 7730265b27
commit 8c4b258355
3 changed files with 15 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
ANNOUNCE
View File

@ -1,5 +1,5 @@
Libpng 1.5.24beta02 - August 19, 2015
Libpng 1.5.24beta02 - August 21, 2015
This is not intended to be a public release. It will be replaced
within a few weeks by a public version or by another test version.
@ -43,7 +43,8 @@ Version 1.5.24beta01 [August 19, 2015]
in ANSI-C (unlike 0x80000000 in the signed case) the checking that
occurs later can catch them (John Bowler).
Version 1.5.24beta02 [August 19, 2015]
Version 1.5.24beta02 [August 21, 2015]
Fixed png_save_int_32 when int is not 2's complement (John Bowler).
Send comments/corrections/commendations to png-mng-implement at lists.sf.net
(subscription required; visit

3
CHANGES
View File

@ -4389,7 +4389,8 @@ Version 1.5.24beta01 [August 19, 2015]
in ANSI-C (unlike 0x80000000 in the signed case) the checking that
occurs later can catch them (John Bowler).
Version 1.5.24beta02 [August 19, 2015]
Version 1.5.24beta02 [August 21, 2015]
Fixed png_save_int_32 when int is not 2's complement (John Bowler).
Send comments/corrections/commendations to png-mng-implement at lists.sf.net
(subscription required; visit

19
pngwutil.c
View File

@ -30,19 +30,20 @@ png_save_uint_32(png_bytep buf, png_uint_32 i)
}
#ifdef PNG_SAVE_INT_32_SUPPORTED
/* The png_save_int_32 function assumes integers are stored in two's
* complement format. If this isn't the case, then this routine needs to
* be modified to write data in two's complement format. Note that,
* the following works correctly even if png_int_32 has more than 32 bits
* (compare the more complex code required on read for sign extention.)
/* PNG signed integers are saved in 32-bit 2's complement format. ANSI C-90
* defines a cast of a signed integer to an unsigned integer either to preserve
* the value, if it is positive, or to calculate:
*
* (UNSIGNED_MAX+1) + integer
*
* Where UNSIGNED_MAX is the appropriate maximum unsigned value, so when the
* negative integral value is added the result will be an unsigned value
* correspnding to the 2's complement representation.
*/
void PNGAPI
png_save_int_32(png_bytep buf, png_int_32 i)
{
buf[0] = (png_byte)((i >> 24) & 0xff);
buf[1] = (png_byte)((i >> 16) & 0xff);
buf[2] = (png_byte)((i >> 8) & 0xff);
buf[3] = (png_byte)(i & 0xff);
png_save_uint_32(buf, i);
}
#endif