Update function names, tests, documentation

NAMESPACE

@@ -3,27 +3,20 @@
 export(ena_query)
 export(extract_accn)
 export(flag_files)
-export(get_meta)
 export(get_uid)
 export(link_uid)
 export(mgnify_endpoints)
 export(mgnify_instance)
 export(mgnify_list)
 export(ncbi_download_genome)
+export(ncbi_parse)
 export(ncbi_parse_assembly_xml)
 export(ncbi_parse_biosample_txt)
 export(parse_gb_header)
 export(parse_report)
-importFrom(XML,xmlParse)
-importFrom(XML,xmlRoot)
-importFrom(XML,xmlToList)
 importFrom(curl,curl_download)
 importFrom(curl,handle_setopt)
 importFrom(curl,new_handle)
-importFrom(dplyr,bind_cols)
-importFrom(dplyr,bind_rows)
 importFrom(httr,RETRY)
 importFrom(httr,content)
-importFrom(rentrez,entrez_fetch)
 importFrom(stringr,str_locate)
-importFrom(tibble,as_tibble)

R/ncbi_meta.R → R/ncbi_parse.R

@@ -1,8 +1,10 @@
 #' Parse NCBI sequence metadata
 #' 
 #' This function can be used to parse various non-sequence data sets from NCBI
-#' into a tibble. The function currently supports parsing NCBI BioSample data
-#' from XML format.
+#' into a tibble. These data sets usually accompany the biological sequences and
+#' contain additional information e.g. identifiers, information about the
+#' sample, the sequencing platform, etc. The function currently supports parsing
+#' NCBI BioSample data from XML format.
 #' @param meta character; either a character vector containing a data set that
 #' was retrieved through \code{rentrez::entrez_fetch()} or a path to an file
 #' that was downloaded from NCBI.
@@ -27,24 +29,24 @@
 #'   retmode = "xml"
 #' )
 #' # Parse XML
-#' ncbi_meta(meta = meta_xml, db = "biosample", format = "xml")
+#' ncbi_parse(meta = meta_xml, db = "biosample", format = "xml")
 #' 
 #' # NCBI BioSample, download XML file from NCBI and parse
 #' 
 #' # Manually download the XML file
 #' # https://www.ncbi.nlm.nih.gov/biosample/?term=SAMN02714232
 #' # upper right corner -> send to -> file -> format = full (xml) -> create file
 #' # Parse XML
-#' ncbi_meta(meta = "biosample_result.xml", db = "biosample", format = "xml")
+#' ncbi_parse(meta = "biosample_result.xml", db = "biosample", format = "xml")
 #' }
 #' @export
-ncbi_meta <- function(
+ncbi_parse <- function(
   meta,
   db,
   format = "xml",
   verbose = getOption("verbose")
 ) {
-  f <- get(paste("ncbi_meta", db, format, sep = "_"))
+  f <- get(paste("ncbi_parse", db, format, sep = "_"))
   if (db == "biosample" & format == "xml") {
     out <- f(meta, verbose)
   } else {

R/ncbi_meta_biosample.R → R/ncbi_parse_biosample_xml.R

@@ -6,13 +6,13 @@
 #' that was retrieved through \code{rentrez::entrez_fetch()} or a path to an xml
 #' file that was downloaded from NCBI BioSample.
 #' @param verbose logical; Should verbose messages be printed to console?
-ncbi_meta_biosample_xml <- function(
+ncbi_parse_biosample_xml <- function(
     biosample_xml,
     verbose = getOption("verbose")
     ) {
   parsed_xml <- xml2::as_list(xml2::read_xml(biosample_xml))[[1]]
   names(parsed_xml) <- sapply(parsed_xml, function(x) attributes(x)$accession)
-  out <- lapply(parsed_xml, ncbi_meta_biosample_xml_entry)
+  out <- lapply(parsed_xml, ncbi_parse_biosample_xml_entry)
   out <- dplyr::bind_rows(out)
   out <- out[, c(
     "biosample_uid",
@@ -23,7 +23,7 @@ ncbi_meta_biosample_xml <- function(
   return(out)
 }
 
-ncbi_meta_biosample_xml_entry <- function(x, verbose = getOption("verbose")) {
+ncbi_parse_biosample_xml_entry <- function(x, verbose = getOption("verbose")) {
   # attributes(x)$names contains all fields! use for validation
   main_attrs <- attributes(x)
   expected_names <- c(

tests/testthat/test-get_uid.R

@@ -10,4 +10,4 @@ test_that("get_uid works with a complex term", {
 
   expect_s3_class(res, c("tbl_df", "tbl", "data.frame"))
   expect_true(nrow(res) > 3000)
-})
+})

tests/testthat/test-ncbi_parse.R

@@ -0,0 +1,13 @@
+test_that("ncbi_parse() works with BioSamples", {
+  data(examples)
+  biosample_uid <- get_uid(examples$biosample, db = "biosample")
+  meta_xml <- rentrez::entrez_fetch(
+    db = "biosample",
+    id = biosample_uid$uid,
+    rettype = "full",
+    retmode = "xml"
+  )
+  res <- ncbi_parse(meta = meta_xml, db = "biosample", format = "xml")
+  expect_s3_class(res, c("tbl_df", "tbl", "data.frame"))
+  expect_equal(res$biosample, c("SAMN02714232", "SAMD00057211", "SAMN32745369"))
+})