Is one.osf.ss01 a valid glyph name?

Jasper de Waard
Jasper de Waard Posts: 639
in Technique and Theory
In other words, is it okay to have two periods in a glyph name? Cheers!

Comments

  • James Puckett
    James Puckett Posts: 1,994
    Yes. 
  • Jasper de Waard
    Jasper de Waard Posts: 639
    Thanks guys!
  • Hin-Tak Leung
    Hin-Tak Leung Posts: 361
    edited September 2016
    You probably want to make sure you have no other one.* glyph name and keep glyph name "unique" though! Just to make the Adobe folks happy :smile:
  • Mark Simonson
    Mark Simonson Posts: 1,736
    edited September 2016
    You can have more than one glyph with the same thing on the left side of the dot (and different things on the right side). It's not a problem. In fact it's the whole point of the dot notation. (Or maybe I'm misunderstanding what you mean.)
  • Hin-Tak Leung
    Hin-Tak Leung Posts: 361
    If there is one well-known application which strips off things after the first dot, even if it is strictly speaking legal, may not be a good idea without a good reason.

    Just because something is strictly speaking legal, not necessarily means that it is a good thing to do :wink:
  • Mark Simonson
    Mark Simonson Posts: 1,736
    edited September 2016
    Dot notation is standard practice for anyone using the AFDKO (either directly, or indirectly in a font editor, which includes almost all of them) as a way of identifying unencoded glyph variants for the purposes of writing feature code. It is not a problem whatsoever to have a font containing glyphs named one, one.tnum, one.onum, one.pnum, etc. Most OpenType fonts (TTF and CFF) in existence are built this way.
  • Mark Simonson
    Mark Simonson Posts: 1,736
    edited September 2016
    Also, the dot notation is used by Acrobat to identify the base character of glyphs that would otherwise be unencoded in PDF documents when the original font is missing. It does this by stripping out the .* part. In other words, Acrobat relies on dot notation to do its job. Again, not a problem.
  • Paul van der Laan
    Paul van der Laan Posts: 242

    A glyph name may be up to 63 characters in length, must be entirely comprised of characters from the following set:

    A-Z a-z 0-9 . (period) _ (underscore)

    and must not start with a digit or period. The only exception is the special character ".notdef".


    Source: http://www.adobe.com/devnet/opentype/afdko/topic_feature_file_syntax.html#2.f.i
  • Thomas Phinney
    Thomas Phinney Posts: 2,888
    Yes, what Mark says: this is a feature and not a problem. It is the entire point of the naming structure!
  • Belleve Invis
    Belleve Invis Posts: 269
    In otfcc, yes. You can use any string as a glyph name.
  • John Hudson
    John Hudson Posts: 3,192
    If there is one well-known application which strips off things after the first dot, even if it is strictly speaking legal, may not be a good idea without a good reason.
    To my knowledge, there isn't such an application. There is one application — Acrobat — that ignores — not strips — everything after the first . when performing mapping from glyph names to Unicode characters. Support for the mechanism is why Adobe's glyph naming rules are defined as they are. Since multiple variant glyphs may map back to a single Unicode character, it stands to reason that the portion of names to the left of the . do not need to be unique.
  • [Deleted User]
    [Deleted User] Posts: 0
    The user and all related content has been deleted.
  • Mark Simonson
    Mark Simonson Posts: 1,736
    The whole point of a hammer is to drive nails into wood. Using a hammer in this way is not a problem.
  • Nick Shinn
    Nick Shinn Posts: 2,208
    edited September 2016
    In order to preserve text accurately in Acrobat PDFs, it is necessary to create two sets of small cap glyphs, one to map back to capitals, the other to map back to lower case, e.g:
    A.c2sc<br>a.smcp
    These have identical outlines and metrics.
    If you just use the lower case character as the base, e.g.
    #c2sc<br>sub A by a.smcp ;
    …then text that started off in the document as Titlecase, set in All Small Caps, will revert to titlecase.

    However, I must confess that I don’t bother with the “A.c2sc” set — did it once and decided it was too much effort for diminishing returns. 

    What is other people’s practice?


  • Chris Lozos
    Chris Lozos Posts: 1,458
    I always use the cap version as in "A.sc" and assume the PDF will reverse the feature code.  If it does not, having the text come back as caps indicates to the user that this text was singled out for a reason, most likely as small caps.
  • Thomas Phinney
    Thomas Phinney Posts: 2,888
    I do the dual sets of small caps, but partly because it's a habit I acquired from working at Adobe. I think it's a tradeoff each designer/foundry (or their production team) needs to determine on their own.
  • John Hudson
    John Hudson Posts: 3,192
    In order to preserve text accurately in Acrobat PDFs
    Note that this is only true for PDFs that have been distilled from a print stream. PDF export tools write the Unicode character strings direct to the PDF; the glyph name parsing mechanism is used only when the character strings are not available, which is less and less the case. I reckon even Adobe will cease caring sooner or later.

Categories