CE Font Character Substitution Map (a little OT)

[Kenneth C. Benson]

Sorry, this is a little OT.

I'm converting a book in Polish from Ventura (before Unicode) to Indesign (Unicode). They used Bookman CE (Central European) to get characters like a ogonek (ą). Bookman CE uses the ordinal one position for a ogonek (the position that normally holds ¹ instead holds ą in Bookman CE). Likewise n accent acute is found in the n tilde position (ñ position has ń instead). Since I'm using the full Unicode version of Bookman, I have to Find/Replace each substitution. For instance, I have to Find all instances of ñ and replace them with ń).

Just by skimming the text, I've found and replaced 12 of these substitutions. What I'm worried about is the substitutions I haven't noticed. For instance, I found the substitution for lower-case e ogonek (ê position has ę instead), but I haven't noticed whatever character is substituted for capital E ogonek. It could be that this book has no E ogonek. Or it could be that the character is something relatively innocuous and I just haven't noticed it. The problem is that I don't know what I'm looking for. Capital E ogonek could be any upper ANSI character.

Is there a resource somewhere that lists CE font character substitutions? What I need is a list of substitutions, like "CE ñ = Unicode ń".

[MW Design]

My experience in the same (VP into whatever) using old fonts is there is not enough consistency as to where the characters are located in a given non-Unicode font. I would suggest using a font manager or other system utility to print a character map. You could also just make one yourself using a document with the 255 characters in any font, then switch the paragraph style to the non-Unicode font, print that.

Then it makes it reasonably easy to search the document and do the requisite replacements.

[Kenneth C. Benson]

Right, but if I had the font I wouldn't need to do any of this.

I've actually had to do this before, and I remember some of the substitutions (like a ordinal is always substituted for a ogonek). Feels like there's at least some consistency. Someone must have made a table at least of the most common substitutions.

[MW Design]

If you have any such font then perhaps it would suffice to make a map.

How about the name(s) of the fonts? I may have them. And I do have VP installed so could always get to the if they were bundled with VP.

Just trying to help.

[Kenneth C. Benson]

Ventura shows the document as using

Bookman Old Style

Bookman Old Style Baltic

Bookman Old Style CE

But I can't actually find any usages of Baltic or CE in the Ventura document (not even in the style definitions). Everything appears to use plain old Bookman Old Style. But my Bookman Old Style is Unicode, and the Bookman Old Style that was used in the Ventura document was clearly not. I think I remember that (pre-Unicode) in some applications the CE version of a font would be used silently if a CE language was specified. So if the style said font:Bookman Old Style and language:Polish, the document would appear to use Bookman Old Style but quietly use Bookman Old Style CE.

I don't think this was a Corel font. Corel fonts were Bitstream, and had BT in their names. This likely came with an old version of Windows or MS Office. This would explain also why I don't have the font. I've always kept fonts that I acquire, but fonts that come with MS products install with the application. I've never bothered to transfer these from computer to computer because I already have them with the new OS.

[MW Design · Accepted solution]

Does this help? Calibri on the left, ITC Bookman CE on the right in each column.

Dropbox - Publication1.pdf

[Kenneth C. Benson]

Yes, this fills in some of the holes, although I wonder about Z accent acute (Ź). If Ÿ replaces ź, what would replace Ź in a less extended font? Calibri already has CE, Greek, Cyrillic, Hebrew, Arabic. Most fonts don't have a Ź, though. Any of Adobe's "Std" fonts have a basic Latin set that does not include a Ź. What shows up in that space if you used, for instance, Bembo Std, or Goudy Oldstyle Std (or, really, any Adobe Std font)?

[MW Design]

As you noted, as the STD fonts do not have that character, at best the .notdef character would show up, but it may simply appear blank unless ID will substitute with a font that has that character...in which case whatever is in that spot in the sub'd font has.

In the case of the ITC Bookman I used, it is multi-mapped. It has both a $8F control code and $0179, which is a superscript 3 character. So if I just type alt+0179 I get ³. (Which is a different code than the 3 superior used in Bembo.)

It's just a right pita. The last time I did this with a non-English book, remapping characters took longer than re-layout.

I can poke around inside the ITC Bookman and Adobe's Bembo STD et al later. Got some uploads I need to do...and a lunch date. I'll pop back in a couple hours from now.

Mike