Create EML from Template • EMLaide

This article demonstrates how to create an EML document for a data package containing multiple data entities. To follow along with this example, please download the Banet-Example.

We will create a nested list from our metadata templates and then use the EML R package’s write_eml function to convert our list into a valid EML document.

Our EML file will contain the following elements:

- eml 
  -access
  -dataset
    - creator
    - contact
    - associated parties
    - title  
    - abstract 
    - keyword set  
    - license 
    - methods 
    - maintenance 
    - project
    - coverage 
    - data table

We have two main sections within our EML document:

Access - The Access Section defines the access for the entire resource (entire EML document)
Dataset - The Dataset Resource contains the descriptive metadata information on a data set. A data set can contain one or more data entities. A data entity is a tabular data table or a vector or raster image.

For a more in depth description of EML please see the EML Specification.

Required R Packages

The following libraries are needed to create a working EML document.

library(EMLaide)
library(tidyverse)
library(readxl)
library(EML)

Datatable(s) and associated metadata

To follow along with this example please download the files from the Banet-Example. Within the directory, there are two subdirectories (“data” and “metadata”) which contain all the necessary data and metadata to create a valid EML document using EMLaide.

At minimum, four files are needed to use our tools:

Dataset text file containing machine readable tabular data
Metadata Template An excel workbook containing the majority of the metadata. Full documentation on the template materials are here.
Abstract Document A word or markdown document containing the abstract text
Methods Document A word or markdown document containing the methods text

Because our data set is composed of multiple data entities, we are creating a dataframe with each row representing a different data entity with the following information:

filepath - filepath of the data entity (datatables)
attribute_info - filepath to the metadata describing the specific data entity (should be a metadata.xlsx)
datatable_description - brief description of data entity
datatable_url - (Optional) url to a publicly accessible file hosted online. Required if using EMLaide::evaluate_edi_package() or EMLaide::upload_edi_package() to evaluate or upload your EML document to EDI from R.
datatable_methods - (Optional) filepath data entity specific methods document
additional_information - (Optional) brief text description containing additional information that is pertinent to a specific data entity.

This dataframe is the input to the add_datatable function which function generates attribute metadata from the attribute_info files and physical information describing the datatable from the filepath and datatable_url information.

Example Dataframe Structure for datatable_metadata

datatable_metadata <- 
  dplyr::tibble(filepath = c("data/enclosure-study-growth-rate-data.csv",
                             "data/enclosure-study-gut-contents-data.csv",
                             "data/microhabitat-use-data-2018-2020.csv",
                             "data/seining-weight-lengths-2018-2020.csv",
                             "data/snorkel-index-data-2015-2020.csv"),  
                attribute_info = c("metadata/enclosure-study-growth-rates-metadata.xlsx",
                                   "metadata/enclosure-study-gut-contents-metadata.xlsx",
                                   "metadata/microhabitat-use-metadata.xlsx",
                                   "metadata/seining-weight-length-metadata.xlsx",
                                   "metadata/snorkel-index-metadata.xlsx"),
                datatable_description = c("Growth Rates - Enclosure Study",
                                          "Gut Contents - Enclosure Study",
                                          "Microhabitat Data",
                                          "Seining Weight Lengths Data",
                                          "Snorkel Survey Data"),
                datatable_url = paste0("https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/",
                                       c("enclosure-study-growth-rate-data.csv",
                                         "enclosure-study-gut-contents-data.csv",
                                         "microhabitat-use-data-2018-2020.csv",
                                         "seining-weight-lengths-2018-2020.csv",
                                         "snorkel-index-data-2015-2020.csv")))

Each row contains all information needed for a data entity to be added to the dataset element of a data package. If you only have one datatable keep this structure or use a named list with the same information.

filepath	attribute_info	datatable_description	datatable_url
data/enclosure-study-growth-rate-data.csv	metadata/enclosure-study-growth-rates-metadata.xlsx	Growth Rates - Enclosure Study	https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/enclosure-study-growth-rate-data.csv
data/enclosure-study-gut-contents-data.csv	metadata/enclosure-study-gut-contents-metadata.xlsx	Gut Contents - Enclosure Study	https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/enclosure-study-gut-contents-data.csv
data/microhabitat-use-data-2018-2020.csv	metadata/microhabitat-use-metadata.xlsx	Microhabitat Data	https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/microhabitat-use-data-2018-2020.csv
data/seining-weight-lengths-2018-2020.csv	metadata/seining-weight-length-metadata.xlsx	Seining Weight Lengths Data	https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/seining-weight-lengths-2018-2020.csv
data/snorkel-index-data-2015-2020.csv	metadata/snorkel-index-metadata.xlsx	Snorkel Survey Data	https://raw.githubusercontent.com/FlowWest/edi.749.1/main/data/snorkel-index-data-2015-2020.csv

Data Package Metadata, Methods, and Abstract

The following code loads the “data-package-metadata.xlsx”, “abstract.docx”, and “methods.docx”. Each sheet of the excel workbook pertains to a different metadata element and will be the input to the add_[blank] functions used throughout this example.

excel_path <- "Banet-Example/metadata/data-package-metadata.xlsx"
sheets <- readxl::excel_sheets(excel_path)
metadata <- lapply(sheets, function(x) readxl::read_excel(excel_path, sheet = x))
names(metadata) <- sheets

abstract_docx <- "metadata/abstract.docx"
methods_docx <- "metadata/methods.docx"

EDI Identifier

In addition to these files, we will need a unique EDI data package identifier. We use the function reserve_edi_id to generate a EDI id. You must already have an account associated with EDI to do this.

edi_number <- reserve_edi_id(user_id = "your user id", password = "your user password ")

You can also reserve this data package identifier on the EDI data repository under tools.

For this example, we will use the following identifier.

edi_number <- "edi.750.1"

Create Dataset Element

We will use magrittr::%>% with our add_[blank] functions to append each EML element to a list. The %>% is a pipe like operator which takes the left-hand side as the first argument of the function appearing on the right-hand side.

For details on appropriate inputs to the functions see documentation at ?add_[blank].

The add_methods() and add_abstract() functions take in the methods_docx and the abstract_docx. The add_datatable() function takes in the datatable_metadata defined and described above. Every other function takes in one or more sheets from the metadata object. For template items with multiple rows, the add_[blank] functions map through each row and adds a named nested list for each row to the dataset element.

The code below adds all dataset elements.

dataset <- list() %>%
  add_pub_date() %>%
  add_title(metadata$title) %>%
  add_personnel(metadata$personnel) %>%
  add_keyword_set(metadata$keyword_set) %>%
  add_abstract(abstract_docx) %>%
  add_license(metadata$license) %>%
  add_method(methods_docx) %>%
  add_maintenance(metadata$maintenance) %>%
  add_project(metadata$funding) %>%
  add_coverage(metadata$coverage, metadata$taxonomic_coverage) %>%
  add_datatable(datatable_metadata)

Handle Custom Units

When units aren’t standard add_datatable()will give a message like the following: "We identified the following custom unit: fishPerSchool , please make sure to add information on this custom unit in additional metadata information:". We must formally define each of these custom units and add them to the EML document as an additional metadata section.

The code below defines 4 custom units and uses the EML::set_unitList() function to format them into a unitList that can be added to our EML document.

custom_units <- data.frame(id = c("fishPerEnclosure", "thermal unit", "day", "fishPerSchool"),
                           unitType = c("density", "temperature", "dimensionless", "density"),
                           parentSI = c(NA, NA, NA, NA),
                           multiplierToSI = c(NA, NA, NA, NA),
                           description = c("Fish density in the enclosure, number of fish in total enclosure space",
                                           "thermal unit of energy given off of fish",
                                           "count of number of days that go by",
                                           "Number of fish counted per school"))

unitList <- EML::set_unitList(custom_units)

Combine EML Elements and Build Document

The code below adds all of the elements we generated above and an access element into an eml list.

The add_access adds an access section at the beginning of our EML document. The add_access default is public principal with a read permission.
The dataset list from above contains all elements of the dataset section of the EML. This includes the datatables, abstract, methods, and all the other metadata sections appended above.
The addtionalMetadata contains the unitList that we generated to hold our custom units.

eml <- list(packageId = edi_number,
            system = "EDI",
            access = add_access(),
            dataset = dataset,
            additionalMetadata = list(metadata = list(unitList = unitList)))

Once all of our information is appended to our eml list we can use the write_eml and eml_validate functions from the EML package to convert our list to EML and check validity.

EML::write_eml(eml, "edi.750.1.xml")
EML::eml_validate("edi.750.1.xml")

Evaluation using the EDI’s EML Congruency Checker

To evaluate your document in R using EDI’s EML Congruence Checker you can use evaluate_edi_package(). To use this function you must have the data entities text files publicly accessible by a URL. This URL must be added in the datatable_metadata section above. If you do not have a URL available then you can upload the EML document and the dataset on the EDI data portal.

evaluate_edi_package(user_id = "Your User Id", password = "Your password", eml_file_path = "edi.750.1.xml")