A custom reader for Pandoc to turn SDD 1.1 datasets into documents (PDF, LaTeX, HTML, Markdown, etc.). Requires Pandoc v2.16.2 or later.

Currently it only supports checklists and dichotomous keys (see Unsupported features).

For more information, see:

Willighagen, L. G. (2023). Ingesting Structured Descriptive Data into Pandoc. Syntaxus baccata. https://doi.org/10.59350/yg9hm-f1x47

pandoc -f path/to/sdd.lua [...]

pandoc -f sdd.lua -t html sdd.xml > sdd.html

pandoc -f sdd.lua -t pdf --pdf-engine=xelatex -V mainfont="Times New Roman" sdd.xml > sdd.pdf

• Metadata (authors, publication date) is read from <RevisionData> (respectively <Creators> and <DateCreated>).
• The first <TaxonHierarchy> is used to structure the document, and if there is no hierarchy specified, the <TaxonNames are displayed in order.
• <IdentifcationKey>s are displayed before the taxonomy or, if a taxonomic <Scope> is specified, under the heading belonging to the first <TaxonName> in the <Scope>.
• The plain text and title belonging to <NaturalLanguageDescription>s are displayed under the headings of all the <TaxonName>s in the <Scope>.
• <MediaObject>s are displayed the first time they are referenced, in a <TaxonName> or <Lead>. Every <MediaObject> is expected to have a caption in the first <Label>.
• Taxon names are displayed in short in keys (no authorship, abbreviated generic epithet for species); in full in headings (with authorship); and if different the vernacular name is listed below the heading. This uses <Representation>/<Label> for the vernacular/fallback name, <CanonicalName> (<Simple>) and <CanonicalAuthorship> for the scientific name, and <Rank> for determining when to italicize.

Valid extensions, according to the XSD.

• <MediaObject>s can have an element <exif:PixelXDimension> to specify the image width.

Invalid extensions, according to the XSD.

• <Lead> can have both a <TaxonName> and <Subkey>, in which case only is the former is displayed, under the assumption that the subkey is listed in the heading belonging to the <TaxonName>.

• Only supports one <Dataset> per file, as document-level metadata is defined in <Dataset> and not <Datasets>.
• As xml:lang is mandatory on <Dataset> in SDD 1.1, making multi-language <Dataset>s difficult, xml:lang on sub-elements is not supported and the first label is used.
• Species and sample descriptions (<CodedDescriptions>, <Specimens>, and <Characters>) are not yet supported.
• Identifcation keys with <Question> are not yet supported.
• Publications (<Publications>) are not yet supported.
• The role of <Label> elements in <MediaObject> elements is not yet taken into account.
• The more detailed information that can be entered in <CanonicalName>, such as <Genus> and <SpecificEpithet>, is not yet handled.