✨ full text search

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: GISAIDR
 Type: Package
 Title: R wrapper for GISAID "API"
-Version: 0.9.5
+Version: 0.9.6
 Author: Wytamma Wirth
 Maintainer: Wytamma Wirth <[email protected]>
 Description: programmatically interact with the GISAID database query.

R/internal_query.R

@@ -1,6 +1,7 @@
 #' Query GISAID Database
 #'
 #' @param credentials GISAID credentials.
+#' @param text full text search.
 #' @param location search for entries based on geographic location.
 #' @param lineage search for entries based on pango lineage designations.
 #' @param variant search for entries based on variant designation
@@ -23,6 +24,7 @@
 #' @return Dataframe.
 internal_query <-
   function(credentials,
+           text = NULL,
            location = NULL,
            lineage = NULL,
            variant = NULL,
@@ -45,6 +47,18 @@ internal_query <-
     df <- tryCatch({
       # search
       queue = list()
+      if (!is.null(text)) {
+        queue <-
+          append(
+            queue,
+            create_search_queue(
+              credentials,
+              credentials$text_ceid,
+              text,
+              'DoSimpleSearch'
+            )
+          )
+      }
       if (!is.null(location)) {
         queue <-
           append(
@@ -320,6 +334,7 @@ internal_query <-
         return(
           query(
             credentials = credentials,
+            text = text,
             location = location,
             lineage = lineage,
             variant = variant,
@@ -328,8 +343,8 @@ internal_query <-
             to = to,
             to_subm = to_subm,
             nrows = j$totalRecords,
-            load_all = FALSE,
             # set load_all to false to break the recursion
+            load_all = FALSE,
             low_coverage_excl = low_coverage_excl,
             complete = complete,
             high_coverage = high_coverage,

R/login.R

@@ -199,10 +199,10 @@ login <- function(username, password, database="EpiCoV") {
   } else {
     linage_ceid <- NULL
   }
-  
+
   # Virus Name
   virus_name_ceid <- extract_search_ceid('covv_virus_name', customSearch_page_text)
-  
+
   # From
   from_ceid <- extract_search_ceid('covv_collection_date_from', customSearch_page_text)
 
@@ -222,6 +222,9 @@ login <- function(username, password, database="EpiCoV") {
     low_coverage_excl_ceid <- NULL
   }
   if (database == 'EpiCoV'){
+    # full text search
+    text_ceid <-
+      extract_search_ceid("fts", customSearch_page_text)
     # Highq
     highq_ceid <-
       extract_search_ceid("highq", customSearch_page_text)
@@ -237,6 +240,7 @@ login <- function(username, password, database="EpiCoV") {
     # quality not used by EpiCov
     quality_ceid <- NULL
   } else {
+    text_ceid <- NULL
     variant_ceid <- NULL
     complete_ceid <- NULL
     highq_ceid <- NULL
@@ -274,6 +278,7 @@ login <- function(username, password, database="EpiCoV") {
       selection_panel_cid = selection_panel_cid,
       selection_ceid = selection_ceid,
       download_panel_cid = query_cid,
+      text_ceid = text_ceid,
       location_ceid = location_ceid,
       search_cid = search_cid,
       linage_ceid = linage_ceid,

R/query.R

@@ -2,6 +2,7 @@
 #' Query GISAID Database
 #'
 #' @param credentials GISAID credentials.
+#' @param text full text search.
 #' @param location search for entries based on geographic location.
 #' @param lineage search for entries based on pango lineage designations.
 #' @param variant search for entries based on variant designation
@@ -24,6 +25,7 @@
 #' @return data.frame
 query <-
   function(credentials,
+           text = NULL,
            location = NULL,
            lineage = NULL,
            variant = NULL,
@@ -51,6 +53,7 @@ query <-
       for (i in 1:nrow(batches)) {
         results <- rbind(results, internal_query(
           credentials = credentials,
+          text = text,
           location = location,
           lineage = lineage,
           variant = variant,
@@ -74,6 +77,7 @@ query <-
       return(
         internal_query(
           credentials = credentials,
+          text = text,
           location = location,
           lineage = lineage,
           variant = variant,
@@ -97,6 +101,7 @@ query <-
       return(
         internal_query(
           credentials = credentials,
+          text = text,
           location = location,
           lineage = lineage,
           variant = variant,

README.md

@@ -105,6 +105,16 @@ df$submission_date
 [1] "2020-01-10" "2020-01-10" "2020-01-11" "2020-01-11" "2020-01-11" "2020-01-12" "2020-01-14"  
 [8] "2020-01-14" "2020-01-14" "2020-01-14" "2020-01-16" "2020-01-17" "2020-01-17" ...
 
+### Full text search
+
+Use `text` for full text search.
+
+``` r
+accession_ids = c("EPI_ISL_17398411", "EPI_ISL_17199001", "EPI_ISL_17409201", "EPI_ISL_17243716")
+df <- query(credentials = credentials, text = paste(accession_ids, collapse = "\n"))
+> df$accession_id
+```
+[1] "EPI_ISL_17199001" "EPI_ISL_17243716" "EPI_ISL_17398411" "EPI_ISL_17409201"
 
 ### Search by location
 

tests/testthat/test-query.R

@@ -9,7 +9,7 @@ test_that("basic query works", {
 })
 
 test_that("can change number of rows", {
-  df <- query(credentials = credentials, nrows = 100, location = "Africa / ...")
+  df <- query(credentials = credentials, nrows = 1000, location = "Africa / ...")
   expect_equal(nrow(df), 100)
 })
 
@@ -144,3 +144,9 @@ test_that("order_by works", {
   df <- query(credentials = credentials, order_by = 'submission_date')
   expect_true(df$submission_date[1] == "2020-01-10")
 })
+
+test_that("text search works", {
+  accession_ids = c("EPI_ISL_17398411", "EPI_ISL_17199001", "EPI_ISL_17409201", "EPI_ISL_17243716")
+  df <- query(credentials = credentials, text = paste(accession_ids, collapse = "\n"))
+  expect_true(nrow(df) == 4)
+})