home *** CD-ROM | disk | FTP | other *** search
- Path: uni-paderborn.de!usenet
- From: laire@basis.owl.de (Ralph Schmidt)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: V43 picture datatype
- Date: 28 Mar 1996 04:35:12 GMT
- Organization: Universitaet Paderborn, Germany
- Message-ID: <4jd4u0$kvj@news.uni-paderborn.de>
- References: <1918.6660T1145T254@sn.no>
- NNTP-Posting-Host: basis.owl.de
- In-Reply-To: <1918.6660T1145T254@sn.no>
- X-Newsreader: NewsFlash [$Revision: 2.104 $] NF-U-00029
-
- On 03/27/96, Christopher Naas wrote:
- >I have a couple of questions about transparency and the picture.datatype.
- >
- >
- >1. Picture datatype returns the wrong transparency color in the BitMapHeader
- > when the picture is remapped to a screen.
- > - Is this intentional?
- > - Is there a way to obtain the new transparency color?
- >
- >2. I have noticed that the new picture.datatype has an interesting tag:
- >
-
- No..it┤s not the wrong colour because the transparency colour only
- works on the source bitmap. If you remap you can┤t gurantee that
- there┤s only the pens set with the remapped transparency colour.
- And if i do a 256->truecolour remap there┤s no pen information at all.
-
- > PDTA_MaskPlane, &tst,
- >
- > .. tst is filled with a value.
-
- That┤s correct.
-
- >
- >
- > How do I use this? I'm guessing that the above reference to Blt..
- > is a typo and should be a reference to BltMaskBitMapRastPort, and that
- > PDTA_MaskPlane (tst) should be used as the mask.
- >
-
- Correct.
-
- > This does, however, not work.
- >
-
- Mhhh..are you sure the picture contains a transparency pen ?
- Try it with Multiview...there it does it automaticly.
- Perhaps you blit the bitmap with a SourceBitmap Offset !=0..then the
- mask bitmap can┤t work anymore.
-
- > passing tst as the mask results in a blank picture, passing &tst, results
- > in colors other than the transparency color dissapearing (several colors
- > are missing).
- >
- > Should I fill a 1 plane bitmap with this value and use Plane[0] as the
- > mask? PDTA_MaskPlane is defined as an ULONG, but a mask is supposed
- to
- > consist of bytes, the same width and height as the original bitmap.
- >
-
- Tags have no other type as ULONG.
-
-
- >
- >3. The new picture datatype does not support TAG_IGNORE, instead it gives
- > a couple of Enforcer hits.
- >
- > 4. How do I get the correct transparency pen when loading a picture
- through
- > datatypes with remapping, and how do I generate a mask for use with
- > BltMaskBitMapRastPort()? I need to support both the 24bit picture datatype
- > and the original OS3.0/3.1 one.
- >
- >
-
- I simply pass TAG_IGNORE to the datatypes.class. I also don┤t handle
- TAG_MORE afaik.
-
-
- --
- Ralph Schmidt,laire@popmail.owl.de,NextMail welcome
-