Adobe Acrobat & Postscript

picard · Post by **picard** » Mon Apr 28, 2003 11:47 pm

I have created several fonts with Fontcreator. None of them however will print through Adobe Acrobat (I cannot create the PDF's with these Fonts). Adobe has identified the problem as a font/postscript error.

Is there a way to add the postscript data to the fonts.

KL

Erwin Denissen · Post by **Erwin Denissen** » Mon Apr 28, 2003 11:59 pm

In the Postscript Names window (select Post command from the Format menu) you can edit the postscript names for all glyphs. Normally the postscript names are related to the mappings. The postscript name is not always included especially in large fonts the postscript names are left out.

Windows does not actually require Postscript name information for the glyphs. However it is recommend to include the postscript names because some word processors and DTP applications depend on this information. Also the printing behavior of fonts that have not included the postscript names is unspecified on PostScript printers, except that it should not result in a fatal or unrecoverable error. Some drivers may print nothing, other drivers may attempt to print using a default naming scheme.

Let us know your results.

picard · Post by **picard** » Tue Apr 29, 2003 3:06 am

Those names are already in the fonts. The error is exactly as follows:

%%[ ProductName: Distiller ]%%
%%[Page: 1]%%
%%[ Error: typecheck; OffendingCommand: rlineto ]%%

Stack:
/K

%%[ Flushing: rest of job (to end-of-file) will be ignored ]%%
%%[ Warning: PostScript error. No PDF file produced. ] %%

Any additional assistance would be appreciated. Thank you.

KL

Erwin Denissen · Post by **Erwin Denissen** » Tue Apr 29, 2003 5:23 am

Can you send a copy of your font directly to me? I'll see if I can identify your problem.

Chris Eilers · Post by **Chris Eilers** » Fri May 02, 2003 4:47 am

It may not be of much help to you, but I can confirm that the problem is not a global lack of PostScript data in Font Creator’s capacity to generate TrueType fonts. I recently, and successfully, created a PDF test file that embedded a number of fonts that I had created in Font Creator, and have received reports that the fonts have been successfully embedded in this PDF test file from email recipients of the test file who have different operating systems. (If you want, I can send you the PDF test file, which describes the kind of font-embedding errors that I have come to expect with PDF file production).

I must admit that I was a little surprised at quite how successfully my early Font Creator-generated TrueType fonts embedded themselves, especially in light of the fact that I:
1) Created a rather stringent PDF test file, combining different Font Creator-generated TrueType fonts in the same paragraph.
2) Included one Font Creator-generated font containing a single logo glyph that has 53 contours and 614 points.
3) Had little or no idea what I was doing in Font Creator when it came to specifying Unicode, etc. platforms, or font naming, etc.
4) Still experience all manner of PDF-font-embedding bugs, despite having years of experience in creating PDF files, having spent hours on the Internet trying to find solutions, and having exported all my PDFs via Adobe Acrobat Distiller from up-to-date versions of one of Adobe’s own programs, PageMaker.

Again, it may not be of any help to you, but it is clear from my Internet wanderings that even professional programmers experience all manner of PDF production bugs. All in all, I’ve formed the tentative hypothesis that the successful production of PDF files requires prayer as well as proficiency

(Sorry, it’s perhaps a little inconsiderate of me to express amusement at a problem that is no doubt extremely annoying for you. As you will no doubt realise, my humour derives from years of extreme frustration.)

Incidentally, I will be very interested in anything that anyone can contribute on this topic. Despite my successful experience in embedding Font Creator-generated TrueType fonts thus far, I don’t trust PDF file generation as far as I can kick it, and my personal reasons for creating TrueType fonts rely completely on their successful embedding in PDF files.

picard · Post by **picard** » Fri May 02, 2003 12:25 pm

Chris,

I solved my problem by using VOLT to add postscript names. Problem solved across all fonts. I have also taken a new path that seems to eliminate the issue.

I start with the cmap from a standard windows font (tahoma) and delete all of the glyphs (leaving the names and mappings). This first step takes me about 10 minutes longer than creating from scratch but it seems to be worth it.

KL

Erwin Denissen · Post by **Erwin Denissen** » Fri May 02, 2003 11:19 pm

I noticed your fonts use 'Segment mapping to delta values encoding' for the Macintosh Roman platform. Maybe this is the cause of the problem.

Could you send a copy of the fixed font? I want to locate the exact differences in the file data and file structure.

Chris Eilers · Post by **Chris Eilers** » Sat May 03, 2003 2:28 am

Thanks for the info, KL. Am tiptoeing tenderly into the techno-voodoo of typographic tables, and have applied for VOLT. Hope it won’t be too much of a shock to the system.

Erwin Denissen · Post by **Erwin Denissen** » Sun May 04, 2003 8:19 am

There are three glyphs with different postscript names:

Glyf 105
'K - Bullet' -> 'Dagger'
Glyf 106
'M - Bullet' -> 'Degree'
Glyf 107
'L- Bullet' -> 'oSlash'

Maybe the spaces were causing the problem.

picard · Post by **picard** » Mon May 05, 2003 1:25 pm

Perhaps, I did correct that as well but those glyphs did not exist in the other fonts that had the same problem.

Thanks for the assist. The FC3 is a good program and easy to use. It appears that fonts created from scratch have this problem with Adobe. I have taken to starting with Times, deleting all of the glyphs (leaving the mappings) and then going from there. I have saved this "blank" font and am starting all from that point.

KL

Erwin Denissen · Post by **Erwin Denissen** » Mon May 05, 2003 2:10 pm

Because there aren't any other differences, I'm positive this web page describes the cause of the problem:

typecheck
An operator was expecting objects of one type, but got a different type. Like finding a name instead of a number.
Usually the cause is corruption.

Because you've been able to produce fonts with FCP that do print, I guess the fonts are correct but the PDF's were corrupted.