[
https://issues.apache.org/jira/browse/PDFBOX-6260?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18114831#comment-18114831
]
valery commented on PDFBOX-6260:
--------------------------------
Claude:
This whitepoint is physically nonsensical. No real CIE standard illuminant (A,
D50–D75, E, F1–F12, etc.) has an X component anywhere near 0.55 — real
illuminants cluster around X≈0.93–1.10, Z≈0.35–1.30. This one's X=0.55 is far
outside that range. But it's not random corruption either — I checked, and the
matrix rows sum exactly to the whitepoint components
(0.0124+0.3576+0.1805=0.5505, etc.), so it's an internally self-consistent,
deliberately-constructed edge case, matching its "unrealistic" framing and its
origin in a Ghostscript bug report about extreme CalRGB values.
Can it be fixed? Yes, narrowly: add a plausibility guard alongside the D65
check — if the whitepoint's X/Z fall well outside any real illuminant's range
(e.g. X∉[0.8,1.2] or Z∉[0.3,1.5]), also fall back to the skip-calibration hack.
That would restore the old/majority-renderer-matching output for this
torture-test file while leaving genuinely realistic non-D65 whitepoints (like
PDFBOX-5079's own D50 file, comfortably inside that range) still correctly
calibrated.
> CalRGB color regression
> -----------------------
>
> Key: PDFBOX-6260
> URL: https://issues.apache.org/jira/browse/PDFBOX-6260
> Project: PDFBox
> Issue Type: Bug
> Components: Rendering
> Affects Versions: 2.0.37, 3.0.8 PDFBox
> Reporter: Tilman Hausherr
> Priority: Minor
> Labels: CalRGB
> Attachments: gs-bugzilla686749-CalRGB-reduced.pdf
>
>
> This is about this "unrealistic" CalRGB file of this GS issue
> https://bugs.ghostscript.com/show_bug.cgi?id=686749
> Rendering has different colors since PDFBOX-5079.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
