Summary: | regression: subtle logic change in fz_paintlinear | ||
---|---|---|---|
Product: | MuPDF | Reporter: | zeniko |
Component: | fitz | Assignee: | Tor Andersson <tor.andersson> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | sebastian.rasmussen |
Priority: | P4 | ||
Version: | unspecified | ||
Hardware: | PC | ||
OS: | Windows 7 | ||
URL: | http://code.google.com/p/sumatrapdf/issues/detail?id=1157 | ||
Customer: | Word Size: | --- | |
Bug Depends on: | 691855 | ||
Bug Blocks: | |||
Attachments: | micron.pdf.bz2 |
Description
zeniko
2011-01-05 12:46:04 UTC
Created attachment 7087 [details]
micron.pdf.bz2
Local copy of the sample file. The file works fine on gs.
Turns out that this doesn't have anything to do with radial shades at all. Rather the logic in fz_paintlinear has been subtly changed, both by fz_paintquad not always passing the same three vertices to fz_painttriangle as fz_paintlinear originally did and by changing the additions in the shade->extend[1] case to subtractions. |