Overview
EMLeditor is designed not only to generate metadata that interacts
with DataStore but to also allow users to ‘surgically’ edit EML fields
without having to edit any of the .txt templates or run the
EML::make_eml() command (which can be time consuming,
especially if there is substantial taxonomic information). Using
EMLeditor also means you don’t need to edit the .xml file by hand. We
strongly discourage editing the .xml file by hand!
If you find typos or mistakes in your EML or if after your data package is reviewed changes to the EML are requested, it is our hope that you can use the EMLeditor functions to make changes to your EML rather than re-running the entire EML creation script.
When you run the DPchecker::run_congruence_checks()
function on your data package, each warning and error returned should
include an indication of how to address the problem. If the problem is
with the metadata (as opposed to the data), DPchecker should give you a
specific function you can use to fix the metadata using the EMLeditor
package.
To edit your EML in R using EMLeditor, you will first need to load the *_metadata.xml file in to R, edit the R object from within R, and then write the edited R object back to .xml.
Function list
The References page has a comprehensive list of all the available functions and links to documentation on how to use each function along with examples.
Loading .xml into R
There are two main methods of loading your *_metadata.xml file in to
R to work on it. You can either use the DPchecker
load_metadata() function or you can use the EML
read_eml() function. The main benefit of
load_metadata() is that you don’t need to specify the file
name or type, so there is less typing. The downside to
load_metadata() is that it is less flexible: if there are
multiple .xml files in the directory or the metadata file doesn’t end in
*_metadata.xml it just won’t work.
Using load_metadata():
#assuming your metadata file is in your current working directory:
metadata <- load_metadata()
#assuming your metadata file is in sub-directory of your current directory, "../other/directory":
dir <- here::here("other", "directory")
metadata <- load_metadata(directory = dir)Using read_eml():
#this assumes your metadata file is in your current working directory and is called 'filename_metadata.xml'
metadata <- read_eml("filename_metadata.xml", from = "xml")
#or if your metadata is in a different directory, change to that directory:
setwd("C:/Users/username/Documents/data_package_directory")
metadata <- read_eml("filename_metadata.xml", from = "xml")Examples of editing metadata
Edit the title
First, you might want to view the current title:
get_title(metadata)Then you can consider editing the title:
metadata <- set_title(metadata, "My New and Improved Data Package Title")Edit the abstract
View your current abstract:
get_abstract(metadata)Supply a new abstract. This function is relatively simple and does not support multiple parts or paragraphs, it’s probably best to keep the text relatively short and straight forward.
new_abstract <- "Put your new abstract here. It should be more than 20 words and shoudl be sufficient for a knowledgeable person to understand what whas done, when, and where including both data collection and data QA/QC but does nto need to be as detailed as a methods statement"
metadata <- set_abstract(metadata, new_abstract)Add ORCIDs
Do this BEFORE adding any organizations as creators. If you already
have organizations as creators, you may want to temporarily remove them
while you add ORCIDs for individuals (see
set_creator_order()). If you did not know that you the
authors/creators of your data package had ORCIDs (or you didn’t register
one until after making the initial metadata), you can add ORCIDs for
individual creators. Make sure to list the ORCIDs in the same order as
the authors/creators are listed. If a creator does not have an ORCID,
list it as NA:
#single author publications:
metadata <- set_creator_orcids(metadata, "1234-1234-1234-1234")
#multiple author publications where the middle author does not have an ORCID:
creator_orcids <- c("1234-1234-1234-1234", NA, "4321-4321-4321-4321")
metadata <- set_creator_orcids(metadata, creator_orcids)Add an Organization as an author
You can add an organization as an author (“creator”) and then re-order the authors to reflect whatever order you would like them to appear in the citation.
First, view the current authors (“creators”). Note that this function will only return a list of individual creators and will not return any organizations that may already be listed as creators/authors:
Add an organization as an author:
# Set an outside organization as an author:
metadata <- set_creator_orgs(metadata, creator_orgs = "Broadleaf, Inc.")
# Set a park unit as a creator
metadata <- set_creator_orgs(metadata, park_units = "YELL")
# You can also add a list of organizations as authors for collaborative works:
networks <- c("ROMN", "MOJN")
metadata <- set_creator_orgs(metadata, park_units = networks)Re-order or remove authors
You may notice that any author or creator you add will be added to
the end of the list of authors/creators. You may then wish to re-order
the authors/creators. The set_creator_order() function’s
default interactive mode lists all the authors in order (surName only
for individuals) and asks you to supply an order. You can also use this
function to remove authors.
metadata <- set_creator_order(metadata)
#Your current creators are in the following order:
# Order Creator
# 1 Smith
# 2 Broadleaf, Inc.
# 3 Yellowstone National Park
# 4 Mojave Desert Network
# 5 Rocky Mountain Network
#Please enter comma-separated numbers for the new creator order.
#Example: put 5 creators in reverse order, enter: 5, 4, 3, 2, 1
#Example: remove the 3rd item (out of 5) enter: 1, 2, 4, 5
5, 4, 3, 2, 1
#Your new creators order is:
# Order Creator
# 1 Rocky Mountain Network
# 2 Mojave Desert Network
# 3 Yellowstone National Park
# 4 Broadleaf, Inc.
# 5 SmithWrite your edits to .xml
Before writing your edits to .xml, make sure to validate your EML:
test_validate_schema(metadata)Write your R object to .xml. Don’t forget that the filename must end in *_metadata.xml:
write_eml(metadata, "filename_metadata.xml")At this point, you should re-run the congruence checks to make sure your data package is still passes them all (or passes any previous warnings/errors you got):
#Assuming you data package is in the working directory and there are no extra .csv or .xml files in that directory:
run_congruence_checks()