From: godzilla AT gimail DOT af DOT mil (godzilla)
Newsgroups: comp.os.msdos.djgpp
Subject: Jpeg 6B library Green Rose error Solved
Date: 22 Feb 2003 19:06:01 -0800
Organization: http://groups.google.com/
Lines: 51
Message-ID: <4c9b0e92.0302221906.7c6c4c95@posting.google.com>
NNTP-Posting-Host: 210.134.90.146
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1045969561 30503 127.0.0.1 (23 Feb 2003 03:06:01 GMT)
X-Complaints-To: groups-abuse AT google DOT com
NNTP-Posting-Date: 23 Feb 2003 03:06:01 GMT
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

I had the same "problem."  After hours of search for this bug fix, I
finally took a closer look at testimg.ppm...the source image for
testout.jpg and testoutp.jpg (please note below).  Anway, the source
file is green...dohh!  Yep, it was apparently shipped that way in the
jpeg-6B archive jpegsrc.v6b.tar.gz.  I also note that it was red
(uncorrupted?) in in the jpeg-6A archive.  So, I think some fumbling
happened in the release, and that also might explain why the file
compare (fc /b ... ...) shows a mismatch.

Bottom line, I don't think it's a code bug at all.

djpeg -dct int -ppm -outfile testout.ppm  testorig.jpg
djpeg -dct int -bmp -colors 256 -outfile testout.bmp  testorig.jpg
cjpeg -dct int -outfile testout.jpg  testimg.ppm
djpeg -dct int -ppm -outfile testoutp.ppm testprog.jpg
cjpeg -dct int -progressive -opt -outfile testoutp.jpg testimg.ppm



>>Message 1 in thread
>>From: Clemens Valens (c DOT valensNOc DOT SPAM AT mindless DOT com DOT invalid)
>>Subject: independent jpeg group library problem
>>
>>
>>View this article only
>>Newsgroups: comp.os.msdos.djgpp
>>Date: 1999/09/30
>>
>>
>>Does anybody have any experience with the jpeg library from the
>>Independent JPEG Group and DJGPP?
>>
>>I have build the package with DJGPP and got the same warning with
two
>>files (jcmainct.c & jdmainct.c), something like "main is usually a
>>function". The rest compiles just fine, using the correct makefile
and
>>config file.
>>
>>When I run the self test (make test) only two tests are correct:
>>testorig.jpg is identical to testoutt.jpg and testimg.bmp is
identical
>>to testout.bmp. All other files are different. When I inspect the
files
>>with a viewer it turns out that when the input is a green rose, the
>>output is a red rose and vice versa. Recompiling doesn't help any.
>>
>>Does anybody recognize this problem and knows a solution?
>>
>>Thanks in advance,
>>Clemens