Write and insert dataset metadata to Atlas

Writing EML metadata and inserting a new dataset to Atlas

Writing the EML

While the EML package provides certain functions to help writing certain fields, others need to be defined manually. The first step is to declare all the required and available fields at hand. This vignette only uses a subset of all the possible EML elements and tags; for a complete list, refer to the EML schema documentation.

Required fields

# Title of the dataset
title <- "Test dataset"

# Creator(s) of the dataset - at least one is required
creator <- list(
  EML::set_responsibleParty(
    givenName = "John",
    surName = "Doe",
    organizationName = "Biodiversité Québec",
    electronicMailAddress = "test@test.com"
  ),
  EML::set_responsibleParty(
    givenName = "Jane",
    surName = "Doe",
    organizationName = "Biodiversité Québec",
    electronicMailAddress = "test@test.com"
  )
)

# Contact(s) of the dataset - at least one is required
contact <- EML::set_responsibleParty(
  givenName = "John",
  surName = "Doe",
  organizationName = "Biodiversité Québec",
  electronicMailAddress = "test@test.com"
)

Optional fields

# Publication date of the dataset
pub_date <- "2025-09-16"

# Abstract of the dataset
abstract <- "This is a test dataset for demonstration purposes."

# Keywords associated to the dataset
keyword_set <- list(keyword = c("biodiversity", "ecology"))

# Urls related to information and download of the dataset
distribution <- list(
  list(
    online = list(
      url = list(
        `function` = "information",
        "https://test.com/info"
      )
    )
  ),
  list(
    online = list(
      url = list(
        `function` = "download",
        "https://test.com/download"
      )
    )
  )
)

# Alternate identifiers for the dataset
# DOI is provided here  (without the prefix of https://doi.org/)
alternate_identifier <- list("10.12309/asdqsd", "413")

# License information
licensed <- list(
  licenseName = "Creative Commons Attribution 4.0 International",
  identifier = "CC-BY-4.0",
  url = "http//:creativecommons.org/licenses/by/4.0/legalcode"
)

# Associated parties of the dataset
associated_party <- list(
  individualName = list(
    givenName = "Cookie",
    surName = "Doe"
  ),
  organizationName = "Culinary school",
  role = "Cookie provider"
)

# Provider of the metadata for the dataset
metadata_provider <- EML::set_responsibleParty(
  givenName = "Johnny",
  surName = "Bravo",
  organizationName = "Mirror Inc.",
  electronicMailAddress = "test@test.com"
)

# Any additional information about the dataset
additional_info <- list(markdown = EML::set_TextType(text = "Some more important information."))

# Language used in the dataset
language <- "English"

# Publisher of the dataset
publisher <- list(
  organizationName = "Biodiversité Québec"
)

# Coverage (taxonomic, temporal and geographic) of the dataset
geographic_description <- "Sherbrooke, Qc, Canada"
coverage <- EML::set_coverage(
  begin = "2025-09-01", end = "2025-09-24",
  sci_names = "Homo sapiens sapiens",
  geographicDescription = geographic_description,
  west = -122.44, east = -117.15,
  north = 37.38, south = 30.00,
  altitudeMin = 160, altitudeMaximum = 330,
  altitudeUnits = "meter"
)

# Citation, provided in additionalMetadata
additional_metadata <- list(
  metadata = list(
    citation = "Bravo, J. (2021). How to mirror. Journal of Bg, 48(7), 1552-1563."
  )
)

# The fields methods is EML compliant, but currently not implemented
# Methods used. Either provided as raw jsonb or as a markdown file with EML::set_methods()
#EML::set_methods(methods_file = "/dev_scratch/test.md")

Non-EML fields

Others fields that are non EML compliant are still required for tracking and managment of the information in Atlas.

# Data type (occurrence, time series, etc.)
data_type <- "occurrence"

# The url of the source EML
source_eml_url <- "https://url-of-the-source-eml.test"

# Source alias (custom alias to recognize the dataset in the database)
source_alias <- "TEST"

# Shareable data (TRUE/FALSE)
shareable_data <- FALSE

Validate and insert the dataset to Atlas

Once all the required and optional fields are defined, we can send it to the post_dataset function which will create, validate and insert the dataset to Atlas. Since this function is not exported, it needs to be used with the triple colon operator: ratlas:::post_dataset(). The function will return the UUID of the newly created dataset.

dataset_uuid <- ratlas:::post_dataset(
  title = title,
  creator = creator,
  contact = contact,
  pub_date = pub_date,
  abstract = abstract,
  keyword_set = keyword_set,
  distribution = distribution,
  alternate_identifier = alternate_identifier,
  licensed = licensed,
  associated_party = associated_party,
  metadata_provider = metadata_provider,
  additional_info = additional_info,
  language = language,
  publisher = publisher,
  coverage = coverage,
  methods = methods,
  additional_metadata = additional_metadata,
  data_type = data_type,
  source_eml_url = source_eml_url,
  source_alias = source_alias,
  shareable_data = shareable_data
)