Validate pdf with and without pdfx

Question

The following document, which uses hyperref, illustrates two ways of embedding metadata into the output from pdflatex:

Using the pdfx package with option a-2u — this happens when the flag validate is set to true, as it is below.
Using instead the hyperxmp package with \hyperrefset including options pdfapart=2, pdfaconformance=u — this happens if the flag validate is set to 'false`.

Both methods seem to embed essentially the same metadata about title, author, etc., into the pdf.

The pdf output file passes PDF/A-2U validation (with the veraPDF app, e.g.) with method 1. However, it fails validation with method 2. (See an extract from the veraPDF report, at the end.)

Question: What can be done to allow validation to succeed with method 2?

Source:

\RequirePackage{filecontents}
\begin{filecontents}{\jobname metadata.xmp}
\hyxmp@at@end{%
  % Create XMP code and write it to macro \hyxmp@xml
  % (cf. hyperxmp.sty, \hyxmp@construct@packet (ll. 847-868))
  \gdef\hyxmp@xml{}%
  \hyxmp@add@to@xml{%
<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="3.1-702">^^J%
___<rdf:RDF
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns\hyxmp@hash">^^J%
  }%
  \hyxmp@pdf@schema
  \hyxmp@xmpRights@schema
  \hyxmp@dc@schema
  \hyxmp@photoshop@schema
  \hyxmp@photometa@schema
  \hyxmp@xmp@basic@schema
  \hyxmp@mm@schema
  \hyxmp@add@to@xml{%
___</rdf:RDF>^^J%
</x:xmpmeta>^^J%
  }%
}
\end{filecontents}
\begin{filecontents}{\jobname.xmpdata}
\Title{A Book}
\Author{Anonymous}
\Language{en-US}
\Keywords{things\sep stuff}
\Subject{matters}
\Publisher{Anonymous}
\Copyright{\copyright 2020 The Company CC-BY-NC-ND}
\CopyrightURL{https://creativecommons.org/licenses/by-nc-nd/4.0/}
\PublicationType{book}
\Lastpage{3}
\Date{2020-06-26}
\CoverDisplayDate{June\ 26,\ 2020}
\CoverDate{2020-06-26}
\end{filecontents}
\documentclass{book}
\usepackage{ifthen}
\newboolean{validate}
\setboolean{validate}{true}
\ifthenelse{\boolean{validate}}%
  {\RequirePackage[a-2u]{pdfx}%
    \pdfglyphtounicode{EM}{0058 0058 0058 0058 0058 0058 0058 0058}%
    \pdfglyphtounicode{NUL}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \RequirePackage[type={CC},modifier={by-nc-nd},version={4.0},lang={english}]{doclicense}
    \usepackage{hyperref}
    \hypersetup{
       pdfa,
       bookmarksnumbered,
       pdftitle={A Book}, pdfauthor={Anonymous}, pdfcreator={somebody},
       pdfsubject={A general introducton to things}, pdfkeywords={things, stuff},
   }  %

}%
  {\RequirePackage{hyperxmp} % to add CC info into pdf
   \RequirePackage[type={CC},modifier={by-nc-nd},version={4.0},lang={english}]{doclicense}
   \usepackage[pdfa]{hyperref}
   \hypersetup{
      pdfapart=2, pdfaconformance=u,
      bookmarksnumbered,
      pdftitle={A Book}, pdfauthor={Anonymous}, pdfcreator={somebody},
      pdfsubject={A general introducton to things}, pdfkeywords={things, stuff},
      pdflicenseurl={http://creativecommons.org/licenses/by-nc-nd/4.0/}
    }%
    \immediate\pdfobj stream attr{/N 3} file{sRGB.icc}
    \pdfcatalog{%
        /OutputIntents [
           <<
             /Type /OutputIntent
             /S /GTS_PDFA2
             /DestOutputProfile \the\pdflastobj\space 0 R
             /OutputConditionIdentifier (sRGB)
             /Info (sRGB)
          >>
      ]
    }
}
\newcommand\mytitle{A Book}
\newcommand\myauthor{Anonymous}
\newcommand\myabstract{An introduction to things in general.}
\newcommand\mydate{\today}
\title{\mytitle}
\author{\myauthor}
\date{\mydate}
\usepackage{newtxtext,newtxmath}
\usepackage[french,ngerman,russian,main=english]{babel}
\usepackage{blindtext}
\begin{document}
\maketitle
\blindmathpaper
\end{document}

Validation failure report for method 2:

<rule specification="ISO 19005-2:2011" clause="6.2.4.3" testNumber="4" status="failed" passedChecks="0" failedChecks="230">
   <description>DeviceGray shall only be used if a device independent DefaultGray colour space has been set when the DeviceGray colour space is used,
    or if a PDF/A OutputIntent is present.</description>
   <object>PDDeviceGray</object>
   <test>gOutputCS != null</test>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[170]/fillCS[0]</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[168]/fillCS[0]</context>
   </check>
   .... [many more similar failures] 
</rule>
<rule specification="ISO 19005-2:2011" clause="6.2.11.7" testNumber="1" status="failed" passedChecks="0" failedChecks="27">
   <description>The Font dictionary of all fonts shall define the map of all used character codes to Unicode values, either via a ToUnicode entry,
    or other mechanisms as defined in ISO 19005-2, 6.2.11.7.2.</description>
   <object>Glyph</object>
   <test>toUnicode != null</test>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[150]/usedGlyphs[1](YKMMCJ+NewTXMI 67 0  0)</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[150]/usedGlyphs[0](YKMMCJ+NewTXMI 109 0  0)</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[135]/usedGlyphs[0](PGXZDZ+txmiaX 8 0  0)</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[2](23 0 obj PDPage)/contentStream[0](24 0 obj PDContentStream)/operators[123]/usedGlyphs[0](YKMMCJ+NewTXMI 50 0  0)</context>
   </check>
   .... [more like the above] 
   <check status="failed">
     <context>root/document[0]/pages[1](12 0 obj PDPage)/contentStream[0](13 0 obj PDContentStream)/operators[601]/usedGlyphs[0](GEXQPI+txsys 0 0  0)</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[1](12 0 obj PDPage)/contentStream[0](13 0 obj PDContentStream)/operators[595]/usedGlyphs[0](YKMMCJ+NewTXMI 63 0  0)</context>
   </check>
   <check status="failed">
     <context>root/document[0]/pages[1](12 0 obj PDPage)/contentStream[0](13 0 obj PDContentStream)/operators[580]/usedGlyphs[0](UOCZFW+txexs 112 0  0)</context>
   </check>
   .... [more like the above]
</rule>

Related:

pdfx + hyperref prevents setting PDF metadata [incompatible packages]

Is it possible to use both hyperxmp and xmpincl in LaTeX?

Added 2020-06-29:

I've continued in How to find glyphs in pdf requiring \pdfglyphtounicode to allow validation? the discussion begun in the comments to the answer https://tex.stackexchange.com/a/551291/13492.

score 3 · Accepted Answer · answered Jun 26 '20 at 22:45

3

The following validates. The main error was that the subtype of the outputintent should be /GTS_PDFA1, /GTS_PDFA2 doesn't exist according the pdf reference.

Beside this some glyph of your fonts hadn't and unicode representation, I added dummy meanings. I didn't try to include the additional xmp data.

\documentclass{article}
\usepackage{hyperxmp}
\RequirePackage[type={CC},modifier={by-nc-nd},version={4.0},lang={english}]{doclicense}
 \usepackage[pdfa]{hyperref}
   \hypersetup{
      pdfapart=2, pdfaconformance=u,
      bookmarksnumbered,
      pdftitle={A Book}, pdfauthor={Anonymous}, pdfcreator={somebody},
      pdfsubject={A general introducton to things}, pdfkeywords={things, stuff},
      pdflicenseurl={http://creativecommons.org/licenses/by-nc-nd/4.0/}
    }%
    \input{glyphtounicode}
    \pdfgentounicode=1
    \pdfglyphtounicode{EM}{0058 0058 0058 0058 0058 0058 0058 0058}%
    \pdfglyphtounicode{NUL}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \pdfglyphtounicode{uni222B.dsp}{222B}%
    \pdfglyphtounicode{summationdisplay.1}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \pdfglyphtounicode{summationdisplay}{0060 0060 0060 0060 0060 0060 0060 0060}%

    \pdfglyphtounicode{radicalBigg}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \pdfglyphtounicode{radicalbig}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \pdfglyphtounicode{radicalbigg}{0060 0060 0060 0060 0060 0060 0060 0060}%
    \immediate\pdfobj stream attr{/N 3} file{sRGB.icc}
    \pdfcatalog{%
        /OutputIntents [
         <<
             /Type /OutputIntent
             /S /GTS_PDFA1
             /DestOutputProfile \the\pdflastobj\space 0 R
             /OutputConditionIdentifier (sRGB)
             /Info (sRGB)
          >>
      ]
    }
\newcommand\mytitle{A Book}
\newcommand\myauthor{Anonymous}
\newcommand\myabstract{An introduction to things in general.}
\newcommand\mydate{\today}
\title{\mytitle}
\author{\myauthor}
\date{\mydate}
\usepackage{newtxtext,newtxmath}
\usepackage[french,ngerman,russian,main=english]{babel}
\usepackage{blindtext}
\begin{document}
abc abc
\maketitle
\blindmathpaper
\end{document}

answered Jun 26 '20 at 22:45

Ulrike Fischer

327,261

What is the "pdf reference" to which you refer? – murray Jun 27 '20 at 20:59
https://www.iso.org/standard/63534.html – Ulrike Fischer Jun 27 '20 at 21:00
ouch: the iso standards doc for PDF costs more than USD 200! How else might I determine which additional \pdfglyphtounicode{...}{...} commands I need to satisfy PDF validation. (In my real, book-length, document, I'm still getting many "The Font dictionary of all fonts shall define the map of all used character codes to Unicode values, either via a ToUnicode entry: rule failures referrring to glyphs in the MathTime Pro 2 fonts, e.g., in MT2MIT, MT2SYT, MT2SYS, MT2EXA, MTBMIT, MT2EXA, MT2SYF, MT2BYST fonts. – murray Jun 27 '20 at 21:50
The glyphs are not in the reference. Their names I found by looking into the uncompressed pdf. – Ulrike Fischer Jun 27 '20 at 23:08
What is meant by "the uncompressed pdf"? Is that the pdf that results from running pdflatex, or is it something else? – murray Jun 28 '20 at 14:29
you can set \pdfcompresslevel=0 and \pdfobjcompresslevel=0. Then the pdf is ascii and you can open it in some editor and look at the content. (I normally use the expl3 variant \RequirePackage{l3pdf}\ExplSyntaxOn\pdf_uncompress:\ExplSyntaxOff as it works with other engines too). – Ulrike Fischer Jun 28 '20 at 14:54
there is also an older version of the pdf reference around (for pdf until 1.7) which is free. Google for it (but be warned it has also around 1000 pages). – Ulrike Fischer Jun 28 '20 at 14:55
OK, now I can examine the uncompressed, ascii, pdf output. But what do I need to look for in order to determine what additional \pdfglyphtounicode commands I need for the file to pass pdf validation? (I need some hint in order to get started!) – murray Jun 29 '20 at 14:37
well look at the example above and search e.g. for summationdisplay.1. Then you can try to guess how entries for other fonts look like. – Ulrike Fischer Jun 29 '20 at 14:39
I've continued this discussion at: https://tex.stackexchange.com/questions/551625/how-to-find-glyphs-in-pdf-requiring-pdfglyphtounicode-to-allow-validation – murray Jun 29 '20 at 20:37

Validate pdf with and without pdfx

1 Answers1

Linked

Related