| ImageData inconsistency? [message #529111] |
Fri, 23 April 2010 05:21  |
Eclipse User |
|
|
|
Originally posted by: totolaricot.mac.com
This is a multi-part message in MIME format.
--------------050505090205050709080504
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Using the attached image, I created an ImageData two different ways:
1)
final Image load = new Image(display, "close.png");
final ImageData idLoad = load.getImageData();
System.out.println("ImageData (image load) " +
idLoad.width+ " " + idLoad.height + " " + idLoad.depth);
2)
ImageData id = new ImageData("close.png");
System.out.println("ImageData (name) " + id.width+ " " +
id.height + " " + id.depth);
this is the outcome
ImageData (image load) 18 18 32
ImageData (name) 18 18 24
the second ImageData structure is misleading as the alphaData array is
actually populated properly and the image is really a 32bpp image (this
is visible when the image is painted on a GC)
LM/
--------------050505090205050709080504
Content-Type: image/png;
name="close.png"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
filename="close.png"
iVBORw0KGgoAAAANSUhEUgAAABIAAAASCAYAAABWzo5XAAAC7mlDQ1BJQ0Mg UHJvZmlsZQAA
eAGFVM9rE0EU/jZuqdAiCFprDrJ4kCJJWatoRdQ2/RFiawzbH7ZFkGQzSdZu NuvuJrWliOTi
0SreRe2hB/+AHnrwZC9KhVpFKN6rKGKhFy3xzW5MtqXqwM5+8943731vdt8A DXLSNPWABOQN
x1KiEWlsfEJq/IgAjqIJQTQlVdvsTiQGQYNz+Xvn2HoPgVtWw3v7d7J3rZrS tpoHhP1A4Eea
2Sqw7xdxClkSAog836Epx3QI3+PY8uyPOU55eMG1Dys9xFkifEA1Lc5/TbhT zSXTQINIOJT1
cVI+nNeLlNcdB2luZsbIEL1PkKa7zO6rYqGcTvYOkL2d9H5Os94+wiHCCxmt P0a4jZ71jNU/
4mHhpObEhj0cGDX0+GAVtxqp+DXCFF8QTSeiVHHZLg3xmK79VvJKgnCQOMpk YYBzWkhP10xu
+LqHBX0m1xOv4ndWUeF5jxNn3tTd70XaAq8wDh0MGgyaDUhQEEUEYZiwUECG PBoxNLJyPyOr
BhuTezJ1JGq7dGJEsUF7Ntw9t1Gk3Tz+KCJxlEO1CJL8Qf4qr8lP5Xn5y1yw 2Fb3lK2bmrry
4DvF5Zm5Gh7X08jjc01efJXUdpNXR5aseXq8muwaP+xXlzHmgjWPxHOw+/Et X5XMlymMFMXj
VfPqS4R1WjE3359sfzs94i7PLrXWc62JizdWm5dn/WpI++6qvJPmVflPXvXx /GfNxGPiKTEm
dornIYmXxS7xkthLqwviYG3HCJ2VhinSbZH6JNVgYJq89S9dP1t4vUZ/DPVR lBnM0lSJ93/C
KmQ0nbkOb/qP28f8F+T3iuefKAIvbODImbptU3HvEKFlpW5zrgIXv9F98LZu a6N+OPwEWDyr
Fq1SNZ8gvAEcdod6HugpmNOWls05Uocsn5O66cpiUsxQ20NSUtcl12VLFrOZ VWLpdtiZ0x1u
HKE5QvfEp0plk/qv8RGw/bBS+fmsUtl+ThrWgZf6b8C8/UXAeIuJAAAACXBI WXMAAAsTAAAL
EwEAmpwYAAABYklEQVQ4EdVSMWvCQBR+FwpqHAQR11AKYocitHSus4tLHf0B buLkoqOTP0HB
vaskLqUJHbuWbIWErq5JQYKv9669I8llERzag8u9fO+979773jFEhHMs4xwk xPEPiBhjSNus
VDAMQyVg4DholkrCZxiGwpU0JHZ6cwcmmw0+T6dYr9XQdV18Wa2wXq0KjHwU k84h+0Ix/hqV
chnC/R4e2m14m83gqtsVno/lEqxGAz6bTTBNM5+mEzm7Hdz3+/A0GgkyXoFK ek0SeBwOwbZt
hUmDUVn5Rdpct1rwvliIKshPVd7M5+D7PliWxfI5heNHHly08HgsggWmEbnr Nd4NBrCdTFQ1
FEn6bMdjuO10wPM8rQ2tNRqxbImEvez1xI1SbNliFEWZ9rSpfR0O4nYpLB+/ IEoPII5jgWU+
+fdAj5EHIB8xBkHA3T/vjGzCyEcxEpen1lrmlhN+NLFPyM2E/j2ibx6fxEEm p2U6AAAAAElF
TkSuQmCC
--------------050505090205050709080504--
|
|
|
|
| Re: ImageData inconsistency? [message #529725 is a reply to message #529572] |
Mon, 26 April 2010 23:58  |
| Eclipse User |
|
|
|
Originally posted by: totolaricot.mac.com
On 4/26/10 2:54 PM, Praveen wrote:
> which platform you are using ? When I run the reported scenario on
> Linux distro - Ubuntu, I have seen the depth is reported as 24 in both
> the cases, which seems to be expected. I believe alpha is not reported
> in depth value, as it would be carried in a separate alphaData array.
Sorry, I should have mentioned it originally. I ran the same code on XP
(3.5 latest stable) and Cocoa 64 swt_3641. In both case the reported
color depth was 24 when loading the ImageData by name, while building an
ImageData from an already loaded Image reported 32. As I mentioned, in
both cases the images were actually 32 bpp (the ImageData.alphaData[]
contained the right alpha values).
|
|
|
| Re: ImageData inconsistency? [message #529727 is a reply to message #529725] |
Tue, 27 April 2010 00:11 |
| Eclipse User |
|
|
|
Originally posted by: totolaricot.mac.com
On 4/27/10 1:58 AM, L. Mihalkovic wrote:
> On 4/26/10 2:54 PM, Praveen wrote:
>> which platform you are using ? When I run the reported scenario on
>> Linux distro - Ubuntu, I have seen the depth is reported as 24 in
>> both the cases, which seems to be expected. I believe alpha is not
>> reported in depth value, as it would be carried in a separate
>> alphaData array.
> Sorry, I should have mentioned it originally. I ran the same code on
> XP (3.5 latest stable) and Cocoa 64 swt_3641. In both case the
> reported color depth was 24 when loading the ImageData by name, while
> building an ImageData from an already loaded Image reported 32. As I
> mentioned, in both cases the images were actually 32 bpp (the
> ImageData.alphaData[] contained the right alpha values).
I see. so the color depth only deals with the main array, and does not
include the alphaData?! so in order to convert an image to 32 bits, I
have to check the depth to figure how to deal with the main array, and
separately check for the presence of alphaData
|
|
|
Powered by
FUDForum. Page generated in 0.02288 seconds
] 
Search
Help
Register
Login
Home
