diff options
author | Jörg Frings-Fürst <debian@jff-webhosting.net> | 2015-11-06 07:14:47 +0100 |
---|---|---|
committer | Jörg Frings-Fürst <debian@jff-webhosting.net> | 2015-11-06 07:14:47 +0100 |
commit | d479dd1aab1c1cb907932c6595b0ef33523fc797 (patch) | |
tree | ad7d454b9edaae3d8892d84cd8f8ef5c2697b79b /tiff/contrib/win_dib/README.Tiffile | |
parent | 9491825ddff7a294d1f49061bae7044e426aeb2e (diff) |
Imported Upstream version 1.8.3upstream/1.8.3
Diffstat (limited to 'tiff/contrib/win_dib/README.Tiffile')
-rwxr-xr-x | tiff/contrib/win_dib/README.Tiffile | 31 |
1 files changed, 0 insertions, 31 deletions
diff --git a/tiff/contrib/win_dib/README.Tiffile b/tiff/contrib/win_dib/README.Tiffile deleted file mode 100755 index 82c6e5c..0000000 --- a/tiff/contrib/win_dib/README.Tiffile +++ /dev/null @@ -1,31 +0,0 @@ -Frank, - -I attached a file that uses RGBA interface (tif_getimage.c) to read a tiff -file and convert to a DIB. It's advantage is that it is easy to read *any* -tiff file suported by libtiff and easily convert it to a DIB. The disadvantage -is that bilevel (B&W) bitmaps (and all other non-rgba images) are also -converted to RGBA, thus taking up 32x as much memory as needed (4 bytes per -pixel, rather than 1 bit). I read tiff files, but don't need to -write them. And my files are typically small, so the overhead is mostly -inconsequential. But for bilevel images, I overrode the get() and put() -routines of the RGBA interface to just copy the data from the input raster -to the output raster, rather than expanding out to full 32 bit format. It -would be nice if there were a simple way to handle all palletized images, -but that would take more custom routines, and it's not that important to me. - -Usage: - - m_pDIB = (PBITMAPINFOHEADER)::ReadTIFF(pathName); - if (m_pDIB != 0) { - m_format = IMAGETYPE_TIF; - } - -This is intended as Win32, but the modifications for new get() and put() -routines may be independent of platform. - -Thanks for your work supporting the forum and the library! - -Regards, - -Mark James -mark@james.net |