Hello Jaye, 

I had a few quick questions that might help me reproduce this over here:

What typeface are you using with the character?

Does the typeface you're using with the character have a customized cmap / PTS spec?

Are you producing the PDF using divpdf/pdflib or psfmtdrv with some sort of distiller?

Kind regards,
Thomas

The typeface is Symbol. I included it in the subject but I didn't put it in the description. The primary font is Guardian Text Egyptian Regular and (as you probably know) the secondary fast calls in a series of pi fonts, one of them being Symbol which is delivered with the system so no, there are no customized cmap or PTS spec. 

We are using pstfmtdrv and use both Adobe Experience Manager and Ghostscript to distill files. Both of these seem irrelevant as it doesn't display in the xyview and we can generate at least one other character in the font.

Seems you thought of everything already - best of luck.

Thanks. I'm sure I didn't think of everything but something somewhere must be wrong with that particular character in that particular font--at least in my install. If someone else can reproduce the character from that font, I think it indicates it's my install. 

Hi Jaye

When I type that character it goes to MathematicalPiThree - may just be the order of the secondary fonts we have. and displays perfectly - not a lot of help!

Chris

Thanks for that, Chris. I did not have that character mapped in MathematicalPiThree but I plugged in the unicode value and ran genfast and it solved the problem as MathematicalPiThree precedes Symbol in the pts/psf spec field list in the fgs. Problem solved.  

Hi Jaye,

I was asked to look at your font setup to diagnose the problem and since you posted publicly in this forum, I thought others might be interested in what was wrong.

First, I want to add a bit more explanation as to the "unspecified typesetter character" composition message which results in the reverse video question mark characters when viewing the division in the XyView.

You only get this message (and xyview display behavior) if a character (Unicode value) does not map to any glyph in any font based on the FASTs referenced in the FV spec by the FF#/FV# pair (including all the FGS/PTS/etc. font specs used to create such FASTs). Ultimately, there's no font called out for such a character, so the Status window in the xyview will not give you any "font" information that's relevant; all you can do in the xyview is to determine the Unicode value of the character (or by the Unicode value given in the composition message when you compose). Since such a character is not mapped to any font glyph, it's an "unspecified typesetter character" with no specified glyph and so you get that message (when composing) and the reverse video question mark to visually show there's a font setup problem.

As you said, that was not the case in your scenario. The Status window reported Font: Symbol (29).

The first thing I did was to open the TSF spec and look at the Rule for the Symbol font (Font Map No: 29).

Right away I noticed that the Encoding field was set incorrectly to "extended" as the Symbol font is a Pi font (and in this case a Type 1 font). The XPP documentation indicates that for a (Type 1) Pi font that the Extended field should be set to "none", rather than "extended".

As soon as I changed "extended" to "none" and saved the TSF spec, when I opened the xyview again the "logical and" (x2227) character displayed just fine from the Symbol font.

I think that's probably all you and others need to know, but if anyone wants a further explanation of why using the wrong Encoding causes this problem just let me know here on this post and I'd be glad to explain further.

Happy font-ing.