Bug 692605

Summary:	Regression: font issues starting with 0d472d and UFST
Product:	Ghostscript	Reporter:	Marcos H. Woehrmann <marcos.woehrmann>
Component:	Regression	Assignee:	Chris Liddell (chrisl) <chris.liddell>
Status:	RESOLVED FIXED
Severity:	normal
Priority:	P2
Version:	master
Hardware:	PC
OS:	Linux
Customer:		Word Size:	---
Attachments:	tests__pdf__zero-length-font.pdf.pgmraw.300.1.page_1.png

Description Marcos H. Woehrmann 2011-10-17 04:31:28 UTC

Starting with 0d472d36958ba10ce3e5747f91776ad2d53530c7 several regressions have been introduced when using UFST.

I've attached a couple of examples, the image on the left is with 0d472d36958ba10ce3e5747f91776ad2d53530c7, the middle is before that commit, and the right is the difference.

Here is a partial list of files that have changed:

tests__pdf__zero-length-font.pdf
tests_private__ps__ps3cet__13-01.PS
tests_private__ps__ps3cet__13-20.PS
tests_private__ps__ps3cet__13-22.PS
tests_private__ps__ps3cet__13-27.PS
tests_private__ps__ps3cet__14-12.PS
tests_private__ps__ps3cet__14-14.PS

Comment 1 Marcos H. Woehrmann 2011-10-17 04:34:53 UTC

Created attachment 8013 [details]
tests__pdf__zero-length-font.pdf.pgmraw.300.1.page_1.png

Comment 3 Chris Liddell (chrisl) 2011-10-26 15:56:51 UTC

Fixed (I hope!) with:

http://git.ghostscript.com/?p=ghostpdl.git;a=commit;h=cf8815

We had some conflicting scaling requirements between what FAPI needs to work correctly, and what was require in the PS world for it to "accept" the Microtype fonts as analogues of the standard Type 1 fonts.