add function for creating standard metadata

[collinschwantes]

Each dataset will require descriptive and structural metadata.

Descriptive metadata will likely be captured in the repository metadata. This can be created from the {deposits} package Who - creators and contributors What - description of the data Where - where were the data collected/processed Why - description of the project When - date created Funder - who supported this work

Structural metadata will need to be created for each dataset. This could likely be captured in a CSV

• [x] Field Name - name of field from column
• [x] Field description - free text description of a property of the data for human consumption
• [x] Field type - one of https://schema.org/DataType or a more specific R class (unlikely as all data products will be CSVs)
• [x] Field units - unit of measure for a field. For species count data (e.g. number of mosquitos) use https://dwc.tdwg.org/list/#dwc_individualCount if applicable.
• [x] Data standard - Name of data standard used for field * this is captured by term uri so maybe not needed?
• [x] Term URI - Uniform Resource Identifier for a term, should be provided by the data standard. e.g uri for individualCount provided above
• [x] Comments - Additional context for the field.

create_metadata_from_template <- function(data){

# create empty data frame 
metadata <- tibble(
    name  = character(),
    description = character(),
    type = character(),
    units = character(),
    term_uri = character(),
    comments = character()
    )

# get fields
metadata$name = names(data)
# get classes 
metadata$type <- purrr:map_chr(data,function(){
### map classes to types
})

return(metadata)

}

[collinschwantes]

• [x] Must be able to pull in fields from ODK template
  • [ ] Think about how to handle the binary versions of multiple select
• [x] See data agreements for descriptive metadata components

[collinschwantes]

Descriptive metadata components will be handled by metadata created for the {deposits} workflow.

Name and type components of the structural metadata are automatically captured in the frictionless workflow in {deposits}

[collinschwantes]

• [x] create vignette work updating frictionless datapackage.json files from this standard

[collinschwantes]

• [x] add function for building up frictionless metadata