fix #184 fix to europe pmc - changed /search route interface, bump de…

…v ver

DESCRIPTION

@@ -7,7 +7,7 @@ Description: Provides a single interface to many sources of full text
     included for searching for articles, downloading full or partial
     text, downloading supplementary materials, converting to various
     data formats.
-Version: 1.1.0.9211
+Version: 1.1.0.9231
 License: MIT + file LICENSE
 Authors@R: c(
     person("Scott", "Chamberlain",

R/europe_pmc_utils.R

@@ -3,29 +3,46 @@
 #' @export
 #' @keywords internal
 #' @name eupmc
-#' @param query (character) Search terms. Required.
-#' @param resulttype (character) The result type can either be idlist, core or lite. 
-#' This parameter determines the fields returned by XML and JSON formats, but it has 
-#' no effect on the DC format. See Details. 
-#' @param synonym (boolean) Synonym searches are not made by default (Default: \code{FALSE}), 
-#' however queries can be expanded using MeSH terminology and the UniProt synonyms list. 
-#' For example aspirin, a synonym would be acetylsalicylic acid;this could be included in 
+#' @param query (character) Search terms. Required. See Details
+#' @param resulttype (character) The result type can either be idlist, 
+#' core or lite. This parameter determines the fields returned by XML and 
+#' JSON formats, but it has no effect on the DC format. See Details. 
+#' @param synonym (boolean) Synonym searches are not made by default 
+#' (Default: \code{FALSE}), however queries can be expanded using MeSH 
+#' terminology and the UniProt synonyms list.  For example aspirin, a synonym 
+#' would be acetylsalicylic acid;this could be included in 
 #' the search by setting the parameter value to \code{TRUE}
-#' @param per_page (integer) Number of records to return. Max: 1000. Default: 25
-#' @param page Page (integer) Page numbers start at 1. Default: 1
+#' @param per_page (integer) Number of records to return. Max: 1000. 
+#' Default: 25
+#' @param cursorMark (character) cursor string, default: `*`
+#' @param sort (character) The default sort order is relevance. Specify the
+#' sort field and sort order. This parameter provides "asc" or "desc" order 
+#' for every single-valued field: P_PDATE_D, AUTH_FIRST, CITED etc. For 
+#' example sorting by CITED in ascending order: CITED asc
 #' @param ... curl options passed on to [crul::HttpClient]
-#' @param id A single Europe PMC article identifier, begins with "PMC", followed by 
-#' numbers, e.g.,  "PMC3257301"
+#' @param id A single Europe PMC article identifier, begins with "PMC", 
+#' followed by numbers, e.g.,  "PMC3257301"
 #' 
-#' @return `eupmc_search` returns a list with results. `eupmc_fields` returns 
+#' @return `eupmc_search` returns a list with results. `eupmc_fields` returns
 #' a data.frame. `eupmc_xml` returns an object of class `xml_document`
-#' @details resulttype parameter options:
+#' @section `query` parameter options:
+#' 
+#' - a keyword or combination of keywords (e.g. HPV virus).
+#' - a phrase with enclosing speech marks (e.g. "human malaria").
+#' - a fielded search (e.g. auth:stoehr). Available search fields are listed
+#' in the Appendix 1 of the Reference Guide or can be retrieved using the 
+#' fields module of the API.
+#' - a specific publication (e.g. ext_id:781840 src:med) Specify ext_id as the 
+#' article identifier, and src as the source database. List of the data sources
+#' can be found on the help pages or in section 3 of the Reference Guide.
+#' 
+#' @section `resulttype` parameter options:
 #' 
 #' - idlist - returns a list of IDs and sources for the given search terms
-#' - lite - returns key metadata for the given search terms; this is the default 
-#'  value if the parameter is unspecified.
-#' - core - returns full metadata for a given publication ID; including abstract, 
-#'  full text links, and MeSH terms.
+#' - lite - returns key metadata for the given search terms; this is the 
+#' default value if the parameter is unspecified.
+#' - core - returns full metadata for a given publication ID; including 
+#' abstract, full text links, and MeSH terms.
 #' 
 #' @references https://europepmc.org/RestfulWebService
 #' 
@@ -41,19 +58,21 @@
 #' eupmc_search(query = 'ext_id:IND43783977 src:agr')
 #' 
 #' # list indexed search fields
-#' head(eupmc_fields())
+#' x <- eupmc_fields()
+#' NROW(x)
+#' head(x)
 #' 
 #' # get full text XML
 #' eupmc_xml('PMC3257301')
 #' }
-eupmc_search <- function(query, resulttype = NULL, synonym = FALSE, per_page = 25, 
-  page = 1, ...) {
+eupmc_search <- function(query, resulttype = 'lite', synonym = FALSE, 
+  per_page = 25, cursorMark = '*', ...) {
 
   is_it(per_page, "numeric")
-  is_it(page, "numeric")
+  is_it(cursorMark, "character")
   is_it(synonym, "logical")
-  args <- ft_compact(list(query = query, resulttype = resulttype, 
-    synonym = ft_as_log(synonym), pageSize = per_page, page = page, 
+  args <- ft_compact(list(query = query, resultType = resulttype, 
+    synonym = ft_as_log(synonym), pageSize = per_page, cursorMark = cursorMark,
     format = "json"))
   newargs <- paste(names(args), args, sep = "=", collapse = "&")
   cli <- crul::HttpClient$new(
@@ -73,12 +92,10 @@ eupmc_fields <- function(...) {
   cli <- crul::HttpClient$new(
     url = utils::URLencode(paste0(eupmc_base(), "fields")),
     opts = list(...))
-  out <- cli$get()
+  out <- cli$get(query = list(format = "json"))
   eupmc_errors(out)
   tt <- out$parse("UTF-8")
-  tmp <- xml2::as_list(xml2::read_xml(tt))
-  data.frame(terms = unname(unlist(lapply(tmp$searchTermList, "[[", "term"))), 
-    stringsAsFactors=FALSE)
+  jsonlite::fromJSON(tt)$searchTermList$searchTerms
 }
 
 #' @export

R/ft_search.R

@@ -117,6 +117,9 @@
 #' # Europe PMC
 #' (res <- ft_search(query='ecology', from='europmc'))
 #' res$europmc
+#' ## get the next batch of results, using the cursorMark result
+#' ft_search(query='ecology', from='europmc', 
+#'   euroopts = list(cursorMark = res$europmc$cursorMark))
 #' 
 #' # Scopus
 #' (res <- ft_search(query = 'ecology', from = 'scopus', limit = 100,

R/plugins_search.R

@@ -92,7 +92,8 @@ plugin_search_europe_pmc <- function(sources, query, limit, start, opts){
     opts$query <- query
     out <- do.call(eupmc_search, opts)
     zz <- list(source = "europmc", found = out$hitCount, 
-               data = out$resultList$result, opts = opts)
+               data = out$resultList$result, opts = opts,
+               cursorMark = out$nextCursorMark)
     structure(zz, class = "ft_ind", query = query)
   } else {
     zz <- list(source = "europmc", found = NULL, data = NULL, opts = opts)