3 Replies Latest reply on Aug 13, 2012 5:20 PM by MP989

    Missing Unicode Encodings

    MP989

      After running a full check on a MSWord 2010 guide converted to PDF using Acrobat X Pro, the accessibility report brings up an error "No Unicode for characters" for several pages where List bulletstyle is used in the MSWord document. The suggestions on "How to add character encodings" are not helpful/or do not apply. I am already using the latest Adobe Acrobat Distiller, already have the latest Adobe PSDriver, and am not allowed to use any Open Font as the fonts are prescribed. Has anyone encountered this issue and knows a solution?

        • 1. Re: Missing Unicode Encodings
          CtDave Level 5

          The problem is with the list bullet style selected in Word.

          Correct this in the Word file.

           

          There is nothing that Distiller, PDFMaker, or Acrobat can do about input characters that do not map to Unicode.

           

           

          If your prescribed template prohibits this then you'll not be having an accessible PDF.

           

          ISO 14289-1:2012, Section 7.2 states

          "Character codes shall map to Unicode as described in ISO 32000-1:2008, 14.8.2.4.2. Characters not included in the Unicode specification may use the Unicode private use area or declare another character encoding."

          Note that the second sentence is not something you'd readily accomplish during manual post-processing of a tagged PDF with Acrobat Pro.

           

          (ISO 14289-1:2012 is the ISO standard for PDF/UA (PDF that is accessible)

           

          Ideally you'd master content in an authoring file that robustly supports up-front support for accessible (tagged) PDF output.

          Applications such as FrameMaker, InDesign, Word 2007, Word 2010 are examples.

          Word 2010 in particular provides good upfront coaching / constraints when authoring for accessible output PDF as does the MS Office suite CommonLook for Office ver. 1.2 .

          At the enterprise level there is Netcentrics full CommonLook product.

          There are some basic "validators" such as Acrobat's Full Check and some others.

           

          But the final step is always the 'walk the structure tree'.

           

          Unfortunately, one often works with a minimal toolkit. Such a work environment means you need to become one with ISO 32000-1 (Section 14 in particular) and ISO 14289-1.

          While I rather enjoy both I suspect that most others "in the trenches" do not (neither is like a good read of a J.D. Robb novel).

          You have to become the theory / theory to practice resource that the software tools otherwise take care of.

           

           

          ~~~~~~~~~~~~~~
          For those who need to or desire to provide accessible PDF the ISO standard PDF/UA-1is a "must have" reference.

          Hold onto to them until the software implementers get up to speed.


          --| The ISO Standard for PDF/UA (ISO 14289-1:2012).
          --| http://www.iso.org/iso/catalogue_detail.htm?csnumber=54564


          Other documents that are pending release are:

          --| PDF/UA-1 Technical Implementation Guide

           

          --| PDF/UA-1 Technical Implementation Guide: 32000-1
          (ISO 32000-1 is the current ISO Standard for PDF)

           

          --| Achieving WCAG 2.0 with PDF/UA (WCAG 2.0 for PDF is PDF/UA)

          The latter three documents are expected to be made available on AIIM's web site.
          http://www.aiim.org/Research-and-Publications/standards/committees/PDFUA/App-notes


          Be well...

           

          Message was edited by: CtDave

          • 2. Re: Missing Unicode Encodings
            raeben3 Level 3

            Try selecting a bullet from the base font, i.e., the font used for the body text in your document.

             

            Unicode mapping has to do with the font used. I can't say I understand it that well, but I've encountered this problem numerous times and know how to fix it.  It seems a lot of the people who understand unicode font mapping find it difficult to give practical solutions for end users. And, there is no list of fonts that will map well in all circumstances that I can find.

             

            I've solved the problem in the past by using the base font or another TrueType font. The problem will frequently show up with the Postscript Symbol font. There is no reason to use a "fancy" font if your bullet is the standard round kind. 

             

            Can't find a font with the right bullet that works? Say your bullet is a decorative glyph  -- you can create a graphic for the bullet and insert that with alternative text in the final PDF.  Not an automatic or easy solution, but it works and if you only have a few bullets it will give you an accessible PDF at the end of the day.

            • 3. Re: Missing Unicode Encodings
              MP989 Level 1

              Hi Guys,

               

              Thanks for the suggestions. Here are the steps I took to fix the problem:

               

              1. Verified that though the entire guide was in Arial font, the bullet itself was in "Symbol" font.

              2. Accessed my Fonts folder in the Control Panel and located the Arial Unicode MS Regular Font. It was hidden and so I had to turn on 'Show' font in that folder. It then appeared in the Font drop down list in MSWord.

              3. Next I clicked the Home>Bullets icon in MS Word and selected "Define a new Bullet" choosing the Arial Unicode font for the bullet.

              4. I modified the list bullet style in the style pallet which updated all the bullets in the entire guide.

              5. Upon running the full check in the converted PDF, there were no unicode character issues.