Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

new123456

23

« on: 12 / April / 2012, 01:25:25 »

Advertisements

philmoz reyalp for Branches_r1798
rudi_de for chdkde_trunk_r984

Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug:
[ 546]= register int shift = cam_CFAPattern - 8;
==>register int shift = CAM_SENSOR_BITS_PER_PIXEL - 8;

cam_CFAPattern !=CAM_SENSOR_BITS_PER_PIXEL

Logged

philmoz

3450

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #1 on: 12 / April / 2012, 01:50:53 »

Quote from: new123456 on 12 / April / 2012, 01:25:25

philmoz reyalp for Branches_r1798
rudi_de for chdkde_trunk_r984

Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug:
[ 546]= register int shift = cam_CFAPattern - 8;
==>register int shift = CAM_SENSOR_BITS_PER_PIXEL - 8;

cam_CFAPattern !=CAM_SENSOR_BITS_PER_PIXEL

Thanks for that - does sort of mess up the DNG thumbnails a bit.

Fixed in revision 1803 (CHDK release-1.0).

Phil.

Logged

CHDK ports:
sx30is (1.00c, 1.00h, 1.00l, 1.00n & 1.00p)
g12 (1.00c, 1.00e, 1.00f & 1.00g)
sx130is (1.01d & 1.01f)
ixus310hs (1.00a & 1.01a)
sx40hs (1.00d, 1.00g & 1.00i)
g1x (1.00e, 1.00f & 1.00g)
g5x (1.00c, 1.01a, 1.01b)
g7x2 (1.01a, 1.01b, 1.10b)

new123456

23

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #2 on: 12 / April / 2012, 18:24:30 »

Thanks philmoz!

Branches_r1803 and chdkde_trunk_r985 in dng.c have a bit bug:

[258] = case 0x132 :
please fix it:[258] ==> case 0x132 : IFD_LIST[j].entry.offset= (int)cam_datetime; break; // ModifyDate
or REM ==> //case 0x132 :
or delete [258]

Logged

rudi

129
A590IS_101B, SX260HS_100B

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #3 on: 13 / April / 2012, 06:59:18 »

Thanks new123456!

fixed in chdk-de changeset 986
I hope, it is now valid.

rudi

Logged

new123456

23

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #4 on: 13 / April / 2012, 10:02:22 »

Quote from: rudi on 13 / April / 2012, 06:59:18

Thanks new123456!

fixed in chdk-de changeset 986
I hope, it is now valid.

rudi

Thanks rudi!
chdk-de changeset 986 987 in dng.c:90 {0x132, T_ASCII, 20, (int)cam_datetime},
gcc ==>dng.c:90:33: error: 'cam_datetime' undeclared here (not in a function),
moved dng.c:143= static char cam_datetime[20] = ""; to dng.c:71 ,then it work OK,please fixed.

NEW123456

« Last Edit: 13 / April / 2012, 10:08:32 by new123456 »

Logged

rudi

129
A590IS_101B, SX260HS_100B

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #5 on: 14 / April / 2012, 06:04:39 »

Hi new123456,
thank you for bug report.

I'm sorry. I pre-compiled changes normally at least until checkin. Why not for changeset 986 - I don't know.
Now we use changeset 1806 in chdk-de (r988). Thanks again also to reyalp and philmoz.

rudi

Logged

new123456

23

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #6 on: 15 / April / 2012, 04:55:33 »

Thanks rudi!
The chdk-de (r988) works OK!

Logged

new123456

23

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #7 on: 15 / April / 2012, 05:29:29 »

philmoz 's chdk_r1807/trunk/core/dng.c
[278]: IFD_LIST[GPS_IFD_INDEX].entry = 0; Crashed for sx210is!
Delete [278] it work OK,don't save GPS info for cameras without GPS in branches_release-1_0_rev=1806 work OK.
[285]: case 0x110 : // CameraName
[286]: case 0xC614: IFD_LIST[j].entry.count = strlen((char*)IFD_LIST[j].entry.offset) + 1; break; // UniqueCameraModel
DNG 's EXIF ==> UniqueCameraModel= ?臯Aビ|筿娽香 ?鸈輷?鎏檝?.瘂.撎.纘傎_8圐潔t3H蒿夘h.Fo?x?筈漮?滩5???T.哽汈?.鷏<."@]P署宮钗'x?乥"H弌狼??.Zi音鵲P?.?..鈉琙.7籤 Why?
I changed:
[285]: case 0x110 : IFD_LIST[j].entry.count = strlen((char*)IFD_LIST[j].entry.offset) + 1; break;
[286]: case 0xC614: IFD_LIST[j].entry.offset = (int)cam_name; break;
then DNG 's EXIF ==> UniqueCameraModel=CameraName==cam_name OK!
Why ? I don't know!

« Last Edit: 15 / April / 2012, 05:35:30 by new123456 »

Logged

philmoz

3450

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #8 on: 15 / April / 2012, 06:30:09 »

Quote from: new123456 on 15 / April / 2012, 05:29:29

philmoz 's chdk_r1807/trunk/core/dng.c
[278]: IFD_LIST[GPS_IFD_INDEX].entry = 0; Crashed for sx210is!
Delete [278] it work OK,don't save GPS info for cameras without GPS in branches_release-1_0_rev=1806 work OK.
[285]: case 0x110 : // CameraName
[286]: case 0xC614: IFD_LIST[j].entry.count = strlen((char*)IFD_LIST[j].entry.offset) + 1; break; // UniqueCameraModel
DNG 's EXIF ==> UniqueCameraModel= ?臯Aビ|筿娽香 ?鸈輷?鎏檝?.瘂.撎.纘傎_8圐潔t3H蒿夘h.Fo?x?筈漮?滩5???T.哽汈?.鷏<."@]P署宮钗'x?乥"H弌狼??.Zi音鵲P?.?..鈉琙.7籤 Why?
I changed:
[285]: case 0x110 : IFD_LIST[j].entry.count = strlen((char*)IFD_LIST[j].entry.offset) + 1; break;
[286]: case 0xC614: IFD_LIST[j].entry.offset = (int)cam_name; break;
then DNG 's EXIF ==> UniqueCameraModel=CameraName==cam_name OK!
Why ? I don't know!

Revision 1808 should fix the GPS bug.

I can't reproduce the camera model issue, possibly related to the GPS problem. Please retest 1808 and see how it goes.

Phil.

Logged

CHDK ports:
sx30is (1.00c, 1.00h, 1.00l, 1.00n & 1.00p)
g12 (1.00c, 1.00e, 1.00f & 1.00g)
sx130is (1.01d & 1.01f)
ixus310hs (1.00a & 1.01a)
sx40hs (1.00d, 1.00g & 1.00i)
g1x (1.00e, 1.00f & 1.00g)
g5x (1.00c, 1.01a, 1.01b)
g7x2 (1.01a, 1.01b, 1.10b)

new123456

23

Re: Branches_r1798 and chdkde_trunk_r984 in dng.c have a bug

« Reply #9 on: 15 / April / 2012, 08:35:52 »

Thanks philmoz!

Revision 1808 work OK. But dng GPS TAG a bit bug:

---- ExifTool ----ExifTool Version Number : 8.87----ExifToolgui5.0.9----
-------------TIFF-IFD0-GPS -------------
Warning : Bad format (254) for GPS entry 34
Warning : Bad format (256) for GPS entry 35
Warning : Bad format (257) for GPS entry 36
Warning : Bad format (258) for GPS entry 37
Warning : Bad format (259) for GPS entry 38
Warning : Bad format (262) for GPS entry 39
Warning : Bad format (273) for GPS entry 40
Warning : Bad format (277) for GPS entry 41
Warning : Bad format (278) for GPS entry 42
Warning : Bad format (279) for GPS entry 43
Warning : Bad format (282) for GPS entry 44
Warning : [minor] Too many warnings -- GPS parsing aborted

Logged