all devices regression report - 2018-04-11-13:51:26 - 1da9cf6dd87287b952826e8606a26f97b474c556
[email protected] Wed, 11 Apr 2018 13:51:26 -0700 (PDT)
| Newsgroups | gmane.comp.printing.ghostscript.regression |
|---|---|
| Message-ID | <20180411205126.5E0ED20404B0@i7> |
Differences in all devices errors using gs/examples files Previous Revision: 36e7d38af2a11b5e4d7117ddd0ebf7326fad7c72 Current Revision: 1da9cf6dd87287b952826e8606a26f97b474c556 commit 1da9cf6dd87287b952826e8606a26f97b474c556 Author: Chris Liddell <[email protected]> AuthorDate: Tue Apr 10 16:01:44 2018 +0100 CommitDate: Wed Apr 11 08:34:19 2018 +0100 Multiple master font cached glyphs In the event that the WeightVector is changed (via setweightvector) on an already defined font object, we had no way to invalidate the glyph cache entries already created for that font, with the previous WeightVector. Given that MM fonts generally substitute for an entire use of a "real" font, and changing "mid stream" is rare (and almost entirely restricted to test files), this uses the sledgehammer approach of purging the existing glyphs from the cache. In real world use, this is going to have less effect on performance than adding a conditional check on a variable length array of floating point numbers is likely to have. psi/int.mak psi/zchar1.c new errors: errors that went away: errors that are not whitelisted: === ghostpdl/examples/text_graph_image_cmyk_rgb.pdf === pxlcolor === === ghostpdl/examples/text_graph_image_cmyk_rgb.pdf === pxlmono ===