I am having some problems with resolving some character entities when saving as HTML. I have looking in the archive and help but can not figure this out myself.
I am using Epic 4.1F with SGML documents that conform to our custom DTD for Technical Manuals and the ati-default SGML declaration (ati-dflt.dcl). The DTD references the ArborText ISO Character Sets (PUBLIC "-//ArborText::dist//ENTITIES ISO Character Entity Sets::19970805//EN" "../entities/all-iso.gml"). We have a mature FOSI for formatting this information. I can Compose PDF without error and all char entities are represented in the PDF with the appropriate glyph.
However, when using the Save-As-HTML, certain char entities, such as "square" (#x25A1), are not converted to a numeric reference in the HTML output. That is, for a document that uses both "infin" (the infinity symbol) and "square", the "infin" entity is changed appropriately to ∞ but "square" is just □ in the HTML output. And my HTML browsers don't know how to represent □. Not surprisingly, the set of char entities that do not convert are the same set that Epic, using my FOSI, does not represent on the screen nor in the Epic Editor Insert Symbol window as the glyph.
I am hoping that there is some magic pref setting that I use to overcome this or failing that a better understanding of how Epic resolves char entities.
Regards, David "david" dot "g" dot "maltby" at "lmco" dot "com" LMSSC-MSO, Cocoa Beach, FL
This will give you an understanding of how Epic resolves character entities. You might consider customizing this file--see this Help topic:
Customizing and configuring Epic => Using characters from non-default character sets => Changing character entity definitions
If you can find a unicode glyph that you want mapped to &square, you should be able to make that happen in a custom charent.cf. &square is by default mapped to an AMS character set that doesn't have unicode mappings.