Bug 691183 - Regression: another problem with -dDOINTERPOLATE
Summary: Regression: another problem with -dDOINTERPOLATE
Status: RESOLVED INVALID
Alias: None
Product: Ghostscript
Classification: Unclassified
Component: Graphics Library (show other bugs)
Version: master
Hardware: PC Linux
: P1 normal
Assignee: Michael Vrhel
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-03-17 17:26 UTC by Marcos H. Woehrmann
Modified: 2011-08-17 21:16 UTC (History)
1 user (show)

See Also:
Customer:
Word Size: ---


Attachments
screenshot.png (14.75 KB, image/png)
2010-03-17 17:27 UTC, Marcos H. Woehrmann
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Marcos H. Woehrmann 2010-03-17 17:26:49 UTC
Created attachment 6074 [details]
vtm2k.pdf

This problem may be the same as Bug 691182 but the symptoms are different enough that it likely is a different issue.

The attached file renders incorrectly when converted by Ghostscript head (r10928) and -dDOINTERPOLATE is specified on the command line.  This behaviour is a regression, r7646 renders the file correctly, however later revisions are different yet again:

r7646 - correct
r7647 - wrong
r8515 - wrong in a different way
r8664 - wrong in another different way

See the attached screenshot.png for output from these revisions.

The command line I'm using:

  bin/gs -sDEVICE=ppmraw -o test.ppm -dDOINTERPOLATE ./vtm2k.pdf
Comment 1 Marcos H. Woehrmann 2010-03-17 17:27:17 UTC
Created attachment 6075 [details]
screenshot.png
Comment 2 Ray Johnston 2010-03-18 16:54:09 UTC
Also assigning to Michael. Note that rev 7647 is the one that effectively enables -dDOINTERPOLATE for PDF files. Before that it was ignored, so the problem was probably there before.
Comment 3 Michael Vrhel 2011-08-15 03:12:13 UTC
One of the unique things about this file is that the source image is in the CIELAB color space.  Likely there is a decoding issue.
Comment 4 Michael Vrhel 2011-08-17 21:16:40 UTC
I don't believe this is incorrect with head.  The source file is a tiny image (like 3x5) and doing our interpolation method on this is going to give an odd looking result.  My belief is that r7646 was not doing the interpolation.