Merge pull request #43 from viralemergence/feature/metadata_wizard

Feature/metadata wizard

Author: collinschwantes

.Rbuildignore

@@ -9,3 +9,4 @@
 ^doc$
 ^Meta$
 ^codemeta\.json$
+^deposit_files$

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: wddsWizard
 Title: Data Wizard for a Minimal Wildlife Disease Data Standard
-Version: 0.1.0
+Version: 0.2.0
 Authors@R: 
       person("Collin",
           "Schwantes",
@@ -20,6 +20,7 @@ Imports:
     here,
     jsonlite,
     purrr,
+    readr,
     rlang,
     snakecase,
     stringr,
@@ -32,7 +33,9 @@ Suggests:
     readxl,
     tibble,
     jsonvalidate,
-    testthat (>= 3.0.0)
+    testthat (>= 3.0.0),
+    taxize,
+    deposits (>= 0.2.1.63)
 Config/testthat/edition: 3
 VignetteBuilder: knitr
 Depends: 
@@ -41,3 +44,5 @@ LazyData: true
 License: MIT + file LICENSE
 URL: https://viralemergence.github.io/wddsWizard/
 BugReports: https://github.com/viralemergence/wddsWizard/issues
+Remotes: 
+    ropenscilabs/deposits

NAMESPACE

@@ -5,9 +5,16 @@ export(clean_field_names)
 export(create_schema_docs)
 export(create_schema_list)
 export(download_deposit_version)
+export(expand_tidy_dfs)
+export(extract_metadata_from_doi)
+export(extract_metadata_oa)
+export(generate_metadata_csv)
+export(generate_repeat_dfs)
 export(get_entity)
 export(get_required_fields)
 export(list_deposit_versions)
+export(make_simple_df)
+export(na_to_blank)
 export(paste_reduce)
 export(paste_reduce_ul)
 export(prep_affiliation)
@@ -31,8 +38,11 @@ export(prep_rights)
 export(prep_subjects)
 export(prep_titles)
 export(sanitize_version)
+export(translate_to_dcmi)
 export(use_wdds_template)
 export(wdds_data_templates)
 export(wdds_example_data)
 export(wdds_json)
+export(wdds_to_dcmi)
+export(wdds_to_pharos)
 importFrom(rlang,.data)

NEWS.md

@@ -1,3 +1,10 @@
+# wddsWizard 0.2.0
+
+* extract metadata from DOIs for projects
+* translate between data standards
+* generate metadata templates
+
+
 # wddsWizard 0.1.0
 
 # wddsWizard 0.0.3

R/extract_metadata_from_doi.R

@@ -0,0 +1,274 @@
+#' Extract Project Metadata from DOI
+#'
+#' Some works are explicitly connected to a publication and the metadata for that publication are fairly complete.
+#' Instead of re-writing the metadata, it would be better to extract it and transform it.
+#'
+#' @param doi String. DOI for a published work
+#' @param file_path String. Where should the output be written?
+#' @param write_output Logical. Should the output be written to a file?
+#' @family Project Metadata
+#'
+#' @returns data frame. A data frame structured in the same way as the metadata template csv.
+#' @export
+#'
+#' @examples
+#'
+#' doi <-"doi.org/10.1038/s41597-025-05332-x"
+#' extract_metadata_from_doi(doi = doi,write_output=FALSE)
+#'
+extract_metadata_from_doi <- function(doi, file_path, write_output = TRUE){
+  if(!curl::has_internet()){
+    rlang::abort("An internet connection is required to extract metadata")
+  }
+
+  assertthat::assert_that(assertthat::is.string(doi),msg = "doi must be a non-vector string")
+
+  assertthat::assert_that(assertthat::is.scalar(write_output)&is.logical(write_output),msg = "write_output must be logical and scalar")
+
+  # make sure DOI is properly formatted
+  doi <- trimws(doi,"both")
+
+  out <- extract_metadata_oa(doi = doi)
+  # extract_metadata_zenodo(doi = doi)
+
+  if(write_output){
+    assertthat::assert_that(assertthat::is.string(file_path),msg = "file_path must be a non-vector string")
+    readr::write_csv(x = out,file = file_path)
+  }
+
+  return(out)
+}
+
+#' Extract Metadata from Open Alex record
+#'
+#'  Uses the DOI for a work to extract metadata from OpenAlex - https://openalex.org/.
+#'  The OpenAlex data model does not included some fields that are part
+#'  of the wdds project metadata  related identifiers.
+#'
+#'  Carefully review and edit the metadata produced.
+#'
+#'  We recommend writing the metadata to a csv, editing the csv, then
+#'  processing it as demonstrated in the project metadata tutorial.
+#'
+#'
+#' @param doi Character. A digital object identifier for a published work.
+#'
+#' @returns data frame. A data frame structured in the same way as the metadata template CSV.
+#' @family Project Metadata
+#' @export
+#'
+#' @examples
+#'
+#' doi <- "doi.org/10.1038/s41597-025-05332-x"
+#' extract_metadata_oa(doi = doi)
+#'
+extract_metadata_oa<-function(doi){
+
+  assertthat::assert_that(assertthat::is.string(doi),msg = "doi must be a non-vector string")
+
+  # doi.org/10.1038/s41597-025-05332-x
+  oa_url <- sprintf("https://api.openalex.org/works/%s",doi)
+
+  oa_json <- jsonlite::fromJSON(txt = oa_url)
+  oa_json$authorships$affiliations
+
+  # Name	Jane Doe
+  # Given Name	Jane
+  # Family Name	Doe
+  # Name Identifier	https://orcid.org/0000-0003-fake-1111
+  # Affiliation	Department of Biology, State University
+  # Affiliation Identifier	https://ror.org/02aqsfake
+
+  ## get creators
+
+  creators <- oa_json$authorships
+
+  # get name
+
+  creator_df <- data.frame("Name" = creators$raw_author_name)
+  creator_df <- creator_df |>
+    dplyr::mutate("Family Name" = stringr::str_split_i(.data$Name, pattern = " ", -1)) |>
+    dplyr::mutate("Given Name" = stringr::str_remove(.data$Name, pattern = .data$`Family Name`))
+
+  # get identifier
+  creator_df$`Name Identifier` <- creators$author$orcid
+
+  # get affiliation string and identifier
+
+  aff_df <- creators$affiliations |>
+    purrr::map_df(function(x){
+      raw_affiliation <- x$raw_affiliation_string[[1]]
+
+      oa_inst_id <- x$institution_ids[[1]][1] |>
+        fs::path_file()
+
+      oa_inst_api  <- sprintf("https://api.openalex.org/institutions/%s", oa_inst_id)
+
+      oa_inst_list <- jsonlite::fromJSON(oa_inst_api)
+
+      oa_inst_list$ror
+
+      out <- data.frame("Affiliation" = raw_affiliation, "Affiliation Identifier" = oa_inst_list$ror)
+
+      return(out)
+    })
+
+  creator_df_tidy <- cbind(creator_df,aff_df) |>
+    dplyr::rename("Affiliation Identifier" = .data$Affiliation.Identifier)
+
+  # reshape to template form
+  creator_df_expanded <- expand_tidy_dfs(creator_df_tidy, group_prefix = "Creators")
+
+  # titles
+  title_df_expanded  <-  data.frame(title = oa_json$title) |>
+    expand_tidy_dfs(group_prefix = "Titles")
+
+  # publicationYear
+  publicationYear_df <- make_simple_df(property = "publicationYear",value=  oa_json$publication_year)
+  # language
+  language_df <- make_simple_df(property = "language", value = oa_json$language)
+  # description
+  description_df <- data.frame(Description = "FILL ME IN","Description Type" = "abstract",check.names = FALSE) |>
+    expand_tidy_dfs(group_prefix = "Descriptions")
+
+  # fundingReferences
+
+  "Funder Name	NSF
+Funder Identifier	http://dx.doi.org/10.13039/10000fake
+Award Number	DBI 2515340
+Award URI	https://www.viralemergence.org/grants
+Award Title	Verena Fellow-in-Residence Award"
+
+  funder_references_tidy <- oa_json$grants |>
+    dplyr::mutate(oa_funder_id = fs::path_file(.data$funder)) |>
+    dplyr::mutate(oa_funder_api = sprintf("https://api.openalex.org/funders/%s", .data$oa_funder_id)) |>
+    dplyr::mutate(funder_identifier = purrr::map_chr(.data$oa_funder_api, function(x){
+      funder_json <- jsonlite::fromJSON(x)
+      funder_ids <- funder_json$ids
+      #use one of ror, crossref doi, or openalex id
+      ids_ordered <- c("ror","doi","wikidata","openalex")
+      preferred_id <- which(ids_ordered %in% names(funder_ids))[1]
+      funder_ids[ids_ordered[preferred_id]][[1]]
+    })
+    ) |>
+    dplyr::select(dplyr::all_of(c("funder_display_name","funder_identifier","award_id"))) |>
+    dplyr::rename("Funder Name" = "funder_display_name",
+                  "Funder Identifier" = "funder_identifier",
+                  "Award Number" = "award_id"
+    ) |>
+    dplyr::mutate(
+      "Award URI" = "",
+      "Award ID" = ""
+    )
+
+  funder_references_df <- expand_tidy_dfs(funder_references_tidy, group_prefix = "Funding References")
+
+
+  # subjects
+
+  subjects_df <- data.frame(Subject = oa_json$keywords$display_name) |>
+    expand_tidy_dfs(group_prefix = 'Subjects')
+
+  # Related Identifiers
+
+  related_identifiers_tidy  <- data.frame("Related Identifier" = "A valid Identifier like a DOI",
+                                          "Related Identifier Type" = "see accepted values here https://datacite-metadata-schema.readthedocs.io/en/4.5/appendices/appendix-1/relatedIdentifierType/#relatedidentifiertype",
+                                          "Relation Type" = "see accepted values here: https://datacite-metadata-schema.readthedocs.io/en/4.5/appendices/appendix-1/relationType/#relationtype",check.names = FALSE )
+
+  related_identifiers_df <- related_identifiers_tidy |>
+    expand_tidy_dfs(group_prefix = "Related Identifiers")
+
+
+  # rbind data frames
+
+  out <- rbind(creator_df_expanded,
+               title_df_expanded,
+               publicationYear_df,
+               language_df,
+               description_df,
+               funder_references_df,
+               subjects_df,
+               related_identifiers_df
+  )
+
+  return(out)
+
+}
+
+
+#' Expand tidy dataframes to project metadata template format
+#'
+#' Creates a JSON-like structure in the csv that can be processed using
+#' established workflows in this package.
+#'
+#' @param tidy_df data frame. Each row corresponds to a complete entry.
+#' @param group_prefix character. A repeatable metadata property in the project
+#'  metadata section of WDDS. See https://viralemergence.github.io/wddsWizard/articles/schema_overview.html#project_metadata
+#'
+#' @family Project Metadata
+#' @returns Data frame. The data frame contains the fields Group, Variable, and Value.
+#' @export
+#'
+#' @examples
+#'
+#'# a nice tidy dataset
+#' creators_tidy <- data.frame("Name" = paste(letters[1:10],LETTERS[1:10]),
+#'          "Given Name" = letters[1:10],
+#'          "Family Name" = LETTERS[1:10],
+#'          "Name Identifier" = sample(1:100,10,FALSE),
+#'          "Affiliation" = letters[11:20],
+#'          "Affiliation Identifier" = 11:20,
+#'          check.names =FALSE)
+#'
+#'# an expanded dataset that matches the template format.
+#' creators_tidy |>
+#'  expand_tidy_dfs(group_prefix = "Creators")
+#'
+#'
+#'
+expand_tidy_dfs <- function(tidy_df,group_prefix){
+
+  assertthat::assert_that(assertthat::is.string(group_prefix),msg = "group_prefix must be character and length 1")
+  # number of groups
+  num_groups <- nrow(tidy_df)
+
+  # group of variables
+  group_variables <- names(tidy_df)
+
+  df_out <- df_out <- data.frame(Group = character(0),
+                                 Variable = character(0),
+                                 Value = character(0))
+
+  for(i in 1:num_groups){
+    values <- tidy_df[i,] |> unlist()
+    names(values) <- NULL
+    group_name <- sprintf("%s %s",group_prefix,i)
+    Group <- c(group_name, rep("",(length(group_variables)-1)))
+    df_i <- data.frame(Group = Group, Variable = group_variables, Value = values)
+    df_out <- rbind(df_out, df_i)
+  }
+
+  return(df_out)
+}
+
+#' A convenience function for making non-repeating items
+#'
+#' @param property string. Metadata group and variable name
+#' @param value A value for that property.
+#'
+#' @family Project Metadata
+#'
+#' @returns data frame. A data frame that conforms to non-repeatable structure in template.
+#' @export
+#'
+#' @examples
+#' language_df <- make_simple_df(property = "language", value = "fr")
+#'
+make_simple_df <- function(property,value){
+  assertthat::assert_that(assertthat::is.string(property),msg = "property must be length 1 and type character")
+
+  assertthat::assert_that(assertthat::is.scalar(value),msg = "value must be length 1")
+
+  out <- data.frame(Group = property,Variable = property, Value=  value)
+  return(out)
+}