Bug 692264

Summary: GhostScript 9.01 output PDF Font printed with squares
Product: Ghostscript Reporter: Nilamadhaba <nilamadhabt>
Component: Font APIAssignee: Chris Liddell (chrisl) <chris.liddell>
Status: RESOLVED INVALID    
Severity: normal CC: chris.liddell, nilamadhabt
Priority: P4    
Version: 9.01   
Hardware: PC   
OS: Linux   
Customer: Word Size: ---

Description Nilamadhaba 2011-06-08 17:27:37 UTC 
When we print one of the output PDF created from GS9.01 we get Font printed with squares. Here ghostScript9.01 doesn't produce any error or warning.
If we print the same title created from GS8.71 the square issue doesn't happen. However, GS8.71 produce the following error or warning on that PDF.

GPL Ghostscript 8.71: Failed to interpret TT instructions for glyph index 12 of font FBAIEI+BellMTBold. Continue ignoring instructions of the font.
GPL Ghostscript 8.71: Failed to interpret TT instructions for glyph index 18 of font FBAINL+BellMTBold. Continue ignoring instructions of the font.
GPL Ghostscript 8.71: Failed to interpret TT instructions for glyph index 33 of font FBAJFM+BellMTBold. Continue ignoring instructions of the font.
GPL Ghostscript 8.71: Failed to interpret TT instructions for glyph index 30 of font FBAJNP+BellMTBold. Continue ignoring instructions of the font.


We want either GS9.01 to give the same error or GS9.01 should successfully generate the output without any print issue.
Comment 1 Chris Liddell (chrisl) 2011-06-08 18:01:48 UTC 
Without a test file to illustrate the problem, there's not a lot we can do to help. Please supply a PDF which illustrates the problem.

This is more than likely a change in the font code than the PDF writing code, so I will look into it first.

FWIW, the squares are very probably Truetype notdef glyphs, so the 9.xx output may well be *strictly* correct - but I realise it is not likely as intended.
Comment 2 Chris Liddell (chrisl) 2011-07-06 12:44:44 UTC 
This may have been fixed by:
http://bugs.ghostscript.com/show_bug.cgi?id=692310

But without a test job, it's impossible to tell.

With no test job forthcoming, I'm closing this bug.