R/extractors.R

In pensoft/ropenbio: R Package of the Open Biodiveristy Knowledge Management System

  #' Node Extractor
  #'
  #' Given an XML2 object, and an \code{ResourceDescriptionFramework} object, which
  #' are both mutable, the \code{ResourceDescriptionFramework} object is
  #' mutated into containing the triples extracted from the XML2 object,
  #' while the XML2 object is possibly mutated into containing some annotations
  #' allowing the faster reprocessing.
  #'
  #' @inheritParams xml2rdf
  #' @param triples the RDF object where the newly created triples will be
  #'   stored (note this is a side-effect/pass by reference)
  #'
  #' @return \code{ResourceDescriptionFramework} containing all the triples from
  #'   the article
  #'
  #' @export
  node_extractor = function (node, xml_schema, xml, reprocess, triples, prefix, new_taxons,
                                dry = FALSE, filename, root_id, publisher_id = publisher_id,
                                journal_id = journal_id, plazi_doc = plazi_doc, plazi_treatment_id=plazi_treatment_id, doi = doi,
                                article_id = article_id)
  {

    #  print(!is.null(xml_schema$injector))
    if (processing_status(node) == FALSE || reprocess == TRUE && !(is.null(triples)) || xml2::xml_name(node)=="article" || xml2::xml_name(node)=="document") {
      #  if (!is.null(xml_schema$injector)) {
      #    xml_schema$injector(obkms_id = rdf4r::last_token(rdf4r::strip_filename_extension(filename),
      #                                                    split = "/"), node)
      #  }s

      atoms = find_literals(node, xml_schema)
      #TODO fix parent_id prefix
      #if (plazi_doc == TRUE && xml_schema$schema_name %in% c("nomenclature_section", "materials_examined", "diagnosis_section", "distribution_section", "discussion_section", "taxonomic_key", "figure", "taxonomic_name_usage", "reference")){
    #    pid = plazi_treatment_id
    #  }else{
        pid = identifier(parent_id(node), prefix)
     # }
        if (is.null(pid)){
          pid = root_id
        }

      new_triples = xml_schema$constructor(atoms, identifiers = list(nid = identifier_new(node, xml, mongo_key = xml_schema$mongo_key,prefix = prefix, blank = FALSE, publisher_id =publisher_id, journal_id=journal_id, doi=doi, article_id = article_id),
                                                                     pid = pid,
                                                                     root_id = root_id),
                                           prefix = xml_schema$prefix, new_taxons = new_taxons, mongo_key = xml_schema$mongo_key, publisher_id = publisher_id,
                                           journal_id = journal_id, plazi_doc = plazi_doc, doi = doi,
                                           article_id = article_id)

      new_triples$set_context(triples$context)

      #  serialization = new_triples$serialize()
      # if (dry == FALSE) {
      #    add_data(serialization, access_options = access_options)
      #  }

      xml2::xml_attr(node, "obkms_process") = "TRUE"

      triples$add_triples(new_triples)

    }
    for (c in xml_schema$components) {
      nodel = xml2::xml_find_all(node, c$xpath)
      for (n in nodel) {

        node_extractor(n, c, xml = xml, reprocess = reprocess, triples = triples,
                          prefix = prefix, new_taxons = new_taxons, dry = dry, filename = filename,
                          root_id = root_id, publisher_id = publisher_id,
                          journal_id = journal_id, plazi_doc = plazi_doc, plazi_treatment_id = plazi_treatment_id, doi = doi,
                          article_id = article_id)

      }
    }
    return(triples)
  }