% \iffalse meta-comment % %% File: tagpdf-roles.dtx % % Copyright (C) 2019-2021 Ulrike Fischer % % It may be distributed and/or modified under the conditions of the % LaTeX Project Public License (LPPL), either version 1.3c of this % license or (at your option) any later version. The latest version % of this license is in the file % % https://www.latex-project.org/lppl.txt % % This file is part of the "tagpdf bundle" (The Work in LPPL) % and all files in that bundle must be distributed together. % % ----------------------------------------------------------------------- % % The development version of the bundle can be found at % % https://github.com/u-fischer/tagpdf % % for those people who are interested. %<*driver> \RequirePackage{pdfmanagement-testphase} \DeclareDocumentMetadata{} \makeatletter \declare@file@substitution{doc.sty}{doc-v3beta.sty} \makeatother \documentclass{l3doc} \usepackage{array,booktabs,caption} \hypersetup{pdfauthor=Ulrike Fischer, pdftitle=tagpdf-checks module (tagpdf)} \begin{document} \DocInput{\jobname.dtx} \end{document} % % \fi % \title{^^A % The \pkg{tagpdf-roles} module\\ Tags, roles and namesspace code ^^A % \\ Part of the tagpdf package % } % % \author{^^A % Ulrike Fischer\thanks % {^^A % E-mail: % \href{mailto:fischer@troubleshooting-tex.de} % {fischer@troubleshooting-tex.de}^^A % }^^A % } % % \date{Version 0.92, released 2021-08-27} % \maketitle % \begin{implementation} % \begin{macrocode} %<@@=tag> %<*header> \ProvidesExplPackage {tagpdf-roles-code} {2021-08-27} {0.92} {part of tagpdf - code related to roles and structure names} % % \end{macrocode} % \section{Code related to roles and structure names} % \subsection{Variables} % Tags have both a name (a string) and a number (for the lua attribute). % Testing a name is easier with a prop, while accessing with a number is % better done with a seq. So both are used and must be kept in sync if a new % tag is added. The number is only relevant for the MC type, tags with the same % name from different names spaces can have the same number. % \begin{variable}{\g_@@_role_tags_seq,\g_@@_role_tags_prop} % \begin{macrocode} %<*package> \@@_seq_new:N \g_@@_role_tags_seq %to get names (type/NS) from numbers \@@_prop_new:N \g_@@_role_tags_prop %to get numbers from names (type/NS) % \end{macrocode} % \end{variable} % \begin{variable}{\g_@@_role_tags_NS_prop} % in pdf 2.0 tags belong to a name space. For every tag we store a default % name space. % The keys are the tags, the value shorthands like pdf2, or mathml. % There is no need to access this from lua, so we use the standard prop commands. % \begin{macrocode} \prop_new:N \g_@@_role_tags_NS_prop %to namespace info % \end{macrocode} % \end{variable} % \begin{variable}{\g_@@_role_NS_prop} % The standard names spaces are the following. The keys are the name % tagpdf will use, the urls are the identifier in the namespace object. % \begin{description} % \item[mathml] http://www.w3.org/1998/Math/MathML % \item[pdf2] http://iso.org/pdf2/ssn % \item[pdf] http://iso.org/pdf/ssn (default) % \item[user] \cs{c_@@_role_userNS_id_str} (random id, for user tags) % \end{description} % More namespaces are possible and % their objects references and the ones of the namespaces must be collected % so that an array can be written to the StructTreeRoot at the end (see tagpdf-tree). % We use a prop to store also the object reference as it will be needed rather % often. % \begin{macrocode} \prop_new:N \g_@@_role_NS_prop % collect namespaces % \end{macrocode} % \end{variable} % % We need also a bunch of temporary variables: % \begin{variable} % { % \l_@@_role_tag_tmpa_tl % ,\l_@@_role_tag_namespace_tmpa_tl % ,\l_@@_role_role_tmpa_tl % ,\l_@@_role_role_namespace_tmpa_tl % } % \begin{macrocode} \tl_new:N \l_@@_role_tag_tmpa_tl \tl_new:N \l_@@_role_tag_namespace_tmpa_tl \tl_new:N \l_@@_role_role_tmpa_tl \tl_new:N \l_@@_role_role_namespace_tmpa_tl % \end{macrocode} % \end{variable} % \subsection{Namesspaces} % The following commands setups a names space. Namespace dictionaries can % contain an optional |/Schema| and |/RoleMapNS| entry. We only reserve the % objects but delay the writing to the finish code, where we can test if the % keys and the name spaces are actually needed % This commands setups objects for the name space and its rolemap. It also % initialize a prop to collect the rolemaps if needed. % \begin{function}{\@@_role_NS_new:nnn} % \begin{syntax} % \cs{@@_role_NS_new:nnn}\Arg{shorthand}\Arg{URI-ID}{Schema} % \end{syntax} % \end{function} % \begin{macro}{\@@_role_NS_new:nnn} % \begin{macrocode} \cs_new_protected:Npn \@@_role_NS_new:nnn #1 #2 #3 { \pdf_object_new:nn {tag/NS/#1}{dict} \pdfdict_new:n {g_@@_role/Namespace_#1_dict} \pdf_object_new:nn {@@/RoleMapNS/#1}{dict} \pdfdict_new:n {g_@@_role/RoleMapNS_#1_dict} \pdfdict_gput:nnn {g_@@_role/Namespace_#1_dict} {Type} {/Namespace} \pdf_string_from_unicode:nnN{utf8/string}{#2}\l_tmpa_str \tl_if_empty:NF \l_tmpa_str { \pdfdict_gput:nnx {g_@@_role/Namespace_#1_dict} {NS} {\l_tmpa_str} } %RoleMapNS is added in tree \tl_if_empty:nF {#3} { \pdfdict_gput:nnx{g_@@_role/Namespace_#1_dict} {Schema}{#3} } \prop_gput:Nnx \g_@@_role_NS_prop {#1}{\pdf_object_ref:n{tag/NS/#1}~} } % \end{macrocode} % \end{macro} % We need an id for the user space. For the tests it should be possible % to set it to a fix value. So we use random numbers which can % be fixed by setting a seed. We fake a sort of % GUID but not try to be really exact as it doesn't matter ... % % \begin{variable}{\c_@@_role_userNS_id_str} % \begin{macrocode} \str_const:Nx \c_@@_role_userNS_id_str { data:, \int_to_Hex:n{\int_rand:n {65535}} \int_to_Hex:n{\int_rand:n {65535}} - \int_to_Hex:n{\int_rand:n {65535}} - \int_to_Hex:n{\int_rand:n {65535}} - \int_to_Hex:n{\int_rand:n {65535}} - \int_to_Hex:n{\int_rand:n {16777215}} \int_to_Hex:n{\int_rand:n {16777215}} } % \end{macrocode} % \end{variable} % Now we setup the standard names spaces. Currently only if we detect pdf2.0 but % this will perhaps have to change if the structure code gets to messy. % \begin{macrocode} \pdf_version_compare:NnT > {1.9} { \@@_role_NS_new:nnn {pdf} {http://iso.org/pdf/ssn}{} \@@_role_NS_new:nnn {pdf2} {http://iso.org/pdf2/ssn}{} \@@_role_NS_new:nnn {mathml}{http://www.w3.org/1998/Math/MathML}{} \exp_args:Nnx \@@_role_NS_new:nnn {user}{\c_@@_role_userNS_id_str}{} } % \end{macrocode} % % \subsection{Data} % In this section we setup the standard data. % At first the list of structure types. We split them in three lists, % the tags with which are both in the pdf and pdf2 namespace, % the one only in pdf and the one with the tags only in pdf2. % We also define a rolemap for the pdfII only type to pdf so that % they can always be used. % \begin{variable} % { % \c_@@_role_sttags_pdf_pdfII_clist, % \c_@@_role_sttags_only_pdf_clist, % \c_@@_role_sttags_only_pdfII_clist, % \c_@@_role_sttags_mathml_clist, % \c_@@_role_sttags_pdfII_to_pdf_prop % } % \begin{macrocode} % \clist_const:Nn \c_@@_role_sttags_pdf_pdfII_clist { Document, %A complete document. This is the root element %of any structure tree containing %multiple parts or multiple articles. Part, %A large-scale division of a document. Sect, %A container for grouping related content elements. Div, %A generic block-level element or group of elements Caption, %A brief portion of text describing a table or figure. Index, NonStruct, %probably not needed H, H1, H2, H3, H4, H5, H6, P, L, %list LI, %list item (around label and list item body) Lbl, %list label LBody, %list item body Table, TR, %table row TH, %table header cell TD, %table data cell THead, %table header (n rows) TBody, %table rows TFoot, %table footer Span, %generic inline marker Link, % Annot, Figure, Formula, Form, % ruby warichu etc .. Ruby, RB, RT, Warichu, WT, WP, Artifact % only MC-tag ?... } \clist_const:Nn \c_@@_role_sttags_only_pdf_clist { Art, %A relatively self-contained body of text %constituting a single narrative or exposition BlockQuote, %A portion of text consisting of one or more paragraphs %attributed to someone other than the author of the %surrounding text. TOC, %A list made up of table of contents item entries %(structure tag TOCI; see below) and/or other %nested table of contents entries TOCI, %An individual member of a table of contents. %This entry's children can be any of the following structure tags: %Lbl,Reference,NonStruct,P,TOC Index, Private, Quote, %inline quote Note, %footnote, endnote. Lbl can be child Reference, %A citation to content elsewhere in the document. BibEntry, %bibentry Code } \clist_const:Nn \c_@@_role_sttags_only_pdfII_clist { DocumentFragment ,Aside ,H7 ,H8 ,H9 ,H10 ,Title ,FENote ,Sub ,Em ,Strong ,Artifact } \clist_const:Nn \c_@@_role_sttags_mathml_clist { abs ,and ,annotation ,apply ,approx ,arccos ,arccosh ,arccot ,arccoth ,arccsc ,arccsch ,arcsec ,arcsech ,arcsin ,arcsinh ,arctan ,arctanh ,arg ,bind ,bvar ,card ,cartesianproduct ,cbytes ,ceiling ,cerror ,ci ,cn ,codomain ,complexes ,compose ,condition ,conjugate ,cos ,cosh ,cot ,coth ,cs ,csc ,csch ,csymbol ,curl ,declare ,degree ,determinant ,diff ,divergence ,divide ,domain ,domainofapplication ,emptyset ,eq ,equivalent ,eulergamma ,exists ,exp ,exponentiale ,factorial ,factorof ,false ,floor ,fn ,forall ,gcd ,geq ,grad ,gt ,ident ,image ,imaginary ,imaginaryi ,implies ,in ,infinity ,int ,integers ,intersect ,interval ,inverse ,lambda ,laplacian ,lcm ,leq ,limit ,ln ,log ,logbase ,lowlimit ,lt ,maction ,maligngroup ,malignmark ,math ,matrix ,matrixrow ,max ,mean ,median ,menclose ,merror ,mfenced ,mfrac ,mglyph ,mi ,min ,minus ,mlabeledtr ,mlongdiv ,mmultiscripts ,mn ,mo ,mode ,moment ,momentabout ,mover ,mpadded ,mphantom ,mprescripts ,mroot ,mrow ,ms ,mscarries ,mscarry ,msgroup ,msline ,mspace ,msqrt ,msrow ,mstack ,mstyle ,msub ,msubsup ,msup ,mtable ,mtd ,mtext ,mtr ,munder ,munderover ,naturalnumbers ,neq ,none ,not ,notanumber ,notin ,notprsubset ,notsubset ,or ,otherwise ,outerproduct ,partialdiff ,pi ,piece ,piecewise ,plus ,power ,primes ,product ,prsubset ,quotient ,rationals ,real ,reals ,reln ,rem ,root ,scalarproduct ,sdev ,sec ,sech ,selector ,semantics ,sep ,set ,setdiff ,share ,sin ,sinh ,subset ,sum ,tan ,tanh ,tendsto ,times ,transpose ,true ,union ,uplimit ,variance ,vector ,vectorproduct ,xor } \prop_const_from_keyval:Nn \c_@@_role_sttags_pdfII_to_pdf_prop { DocumentFragment = Art, Aside = Note, Title = H1, Sub = Span, H7 = H6 , H8 = H6 , H9 = H6 , H10 = H6, FENote= Note, Em = Span, Strong= Span, } % \end{macrocode} % \end{variable} % We fill the structure tags in to the seq. We allow all pdf1.7 and pdf2.0, % and role map if needed the 2.0 tags. % \begin{macrocode} % get tag name from number: \seq_item:Nn \g_@@_role_tags_seq { n } % get tag number from name: \prop_item:Nn \g_@@_role_tags_prop { name } \clist_map_inline:Nn \c_@@_role_sttags_pdf_pdfII_clist { \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ pdf2 } } \clist_map_inline:Nn \c_@@_role_sttags_only_pdf_clist { \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ pdf } } \clist_map_inline:Nn \c_@@_role_sttags_only_pdfII_clist { \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ pdf2 } } \pdf_version_compare:NnT > {1.9} { \clist_map_inline:Nn \c_@@_role_sttags_mathml_clist { \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ mathml } } } % \end{macrocode} % For luatex and the MC we need a name/number relation. % The name space is not relevant. % \begin{macrocode} \int_step_inline:nnnn { 1 }{ 1 }{ \seq_count:N \g_@@_role_tags_seq } { \@@_prop_gput:Nxn \g_@@_role_tags_prop { \seq_item:Nn \g_@@_role_tags_seq { #1 } } { #1 } } % \end{macrocode} % % \subsection{Adding new tags and rolemapping} % % \subsubsection{pdf 1.7 and earlier} % With this versions only RoleMap is filled. % At first the dictionary: % \begin{variable}{g_@@_role/RoleMap_dict} % \begin{macrocode} \pdfdict_new:n {g_@@_role/RoleMap_dict} % \end{macrocode} % \end{variable} % \begin{macro}{\@@_role_add_tag:nn} % The pdf 1.7 version has only two arguments: new and rolemap name. % To make pdf 2.0 types usable we directly define a rolemapping for them. % % \begin{macrocode} \cs_new_protected:Nn \@@_role_add_tag:nn %(new) name, reference to old { \prop_if_in:NnF \g_@@_role_tags_prop {#1} { \int_compare:nNnT {\l_@@_loglevel_int} > { 0 } { \msg_info:nnn { tag }{new-tag}{#1} } \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \@@_prop_gput:Nnx \g_@@_role_tags_prop { #1 } { \seq_count:N \g_@@_role_tags_seq } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ user } } \@@_check_add_tag_role:nn {#1}{#2} \tl_if_empty:nF { #2 } { \pdfdict_gput:nnx {g_@@_role/RoleMap_dict} {#1} {\pdf_name_from_unicode_e:n{#2}} } } \cs_generate_variant:Nn \@@_role_add_tag:nn {VV} \pdf_version_compare:NnT < {2.0} { \prop_map_inline:Nn \c_@@_role_sttags_pdfII_to_pdf_prop { \@@_role_add_tag:nn {#1}{#2} } } % \end{macrocode} % \end{macro} % \subsubsection{The pdf 2.0 version} % \begin{macro}{\@@_role_add_tag:nnnn} % The pdf 2.0 version takes four arguments: % tag/namespace/role/namespace % \begin{macrocode} \cs_new_protected:Nn \@@_role_add_tag:nnnn %tag/namespace/role/namespace { \int_compare:nNnT {\l_@@_loglevel_int} > { 0 } { \msg_info:nnn { tag }{new-tag}{#1} } \@@_seq_gput_right:Nn \g_@@_role_tags_seq { #1 } \@@_prop_gput:Nnx \g_@@_role_tags_prop { #1 } { \seq_count:N \g_@@_role_tags_seq } \prop_gput:Nnn \g_@@_role_tags_NS_prop { #1 }{ #2 } \@@_check_add_tag_role:nn {#1}{#3} \pdfdict_gput:nnx {g_@@_role/RoleMapNS_#2_dict}{#1} { [ \pdf_name_from_unicode_e:n{#3} \c_space_tl \pdf_object_ref:n {tag/NS/#4} ] } } \cs_generate_variant:Nn \@@_role_add_tag:nnnn {VVVV} % \end{macrocode} % \end{macro} % \subsection{Key-val user interface} % The user interface use the key |add-new-tag|, which takes either a % keyval list as argument, or a tag/role. % \begin{macro}{tag,tag-namespace,role,role-namespace,add-new-tag} % \begin{macrocode} \keys_define:nn { @@ / tag-role } { ,tag .tl_set:N = \l_@@_role_tag_tmpa_tl ,tag-namespace .tl_set:N = \l_@@_role_tag_namespace_tmpa_tl ,role .tl_set:N = \l_@@_role_role_tmpa_tl ,role-namespace .tl_set:N = \l_@@_role_role_namespace_tmpa_tl } \keys_define:nn { @@ / setup } { add-new-tag .code:n = { \keys_set_known:nnnN {@@/tag-role} { tag-namespace=user, role-namespace=, %so that we can test for it. #1 }{@@/tag-role}\l_tmpa_tl \tl_if_empty:NF \l_tmpa_tl { \exp_args:NNno \seq_set_split:Nnn \l_tmpa_seq { / } {\l_tmpa_tl/} \tl_set:Nx \l_@@_role_tag_tmpa_tl { \seq_item:Nn \l_tmpa_seq {1} } \tl_set:Nx \l_@@_role_role_tmpa_tl { \seq_item:Nn \l_tmpa_seq {2} } } \tl_if_empty:NT \l_@@_role_role_namespace_tmpa_tl { \prop_get:NVNTF \g_@@_role_tags_NS_prop \l_@@_role_role_tmpa_tl \l_@@_role_role_namespace_tmpa_tl { \prop_if_in:NVF\g_@@_role_NS_prop \l_@@_role_role_namespace_tmpa_tl { \tl_set:Nn \l_@@_role_role_namespace_tmpa_tl {user} } } { \tl_set:Nn \l_@@_role_role_namespace_tmpa_tl {user} } } \pdf_version_compare:NnTF < {2.0} { %TODO add check for emptyness? \@@_role_add_tag:VV \l_@@_role_tag_tmpa_tl \l_@@_role_role_tmpa_tl } { \@@_role_add_tag:VVVV \l_@@_role_tag_tmpa_tl \l_@@_role_tag_namespace_tmpa_tl \l_@@_role_role_tmpa_tl \l_@@_role_role_namespace_tmpa_tl } } } % % \end{macrocode} % \end{macro} % \end{implementation} % \PrintIndex