[
https://issues.apache.org/jira/browse/PDFBOX-1880?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13889729#comment-13889729
]
Tilman Hausherr commented on PDFBOX-1880:
-----------------------------------------
The three inversions are done only once in the constructor, not for every
coordinate.
The code does whats explained on page 308 of the 1.7 pdf spec. I need the "rat"
matrix variable that is created by the 3 inversions in the constructor to
transform the physical coordinates to the internal shading coordinate space.
> [PATCH] Type 1 Shading must not ignore current transformation matrix
> --------------------------------------------------------------------
>
> Key: PDFBOX-1880
> URL: https://issues.apache.org/jira/browse/PDFBOX-1880
> Project: PDFBox
> Issue Type: Bug
> Affects Versions: 2.0.0
> Reporter: Tilman Hausherr
> Attachments: RGBCUBE.pdf, Type1ShadingContext.patch,
> rgbcube.pdf-1-bad.png, rgbcube.pdf-1-good.png
>
>
> My implementation for Type 1 Shading (PDFBOX-1869) ignored the current
> transformation matrix, because I thought that the shading matrix would be
> relevant only. That is the cause for the wrong rendering of the rgbcube file.
> The attached patch fixes this.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
