047 handling http failures

Author: michalovadek

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: eurlex
 Type: Package
 Title: Retrieve Data on European Union Law
-Version: 0.4.6
+Version: 0.4.7
 Authors@R: c(person(given = "Michal",
              family = "Ovadek",
              role = c("aut", "cre", "cph"),
@@ -12,7 +12,7 @@ License: GPL-3
 Encoding: UTF-8
 Language: en-US
 Depends:
-    R (>= 3.4.0)
+    R (>= 3.5.0)
 Imports:
     magrittr,
     dplyr,
@@ -36,7 +36,7 @@ Suggests:
     ggiraph,
     testthat (>= 3.0.0)
 URL: https://michalovadek.github.io/eurlex/
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.1
 Roxygen: list(markdown = TRUE)
 VignetteBuilder: knitr
 Config/testthat/edition: 3

NEWS.md

@@ -1,3 +1,9 @@
+# eurlex 0.4.7
+
+## Minor changes
+- some http calls were still not failing gracefully
+- some leftover .data in tidyselect
+
 # eurlex 0.4.6
 
 ## Minor changes

R/elx_curia_list.R

@@ -33,7 +33,6 @@ elx_curia_list <- function(data = c("all","ecj_old","ecj_new","gc_all","cst_all"
 
   else {data <- match.arg(data)}
 
-
   if (data == "all"){
 
     res_c1 <- elx_curia_scraper(url_c1)
@@ -89,7 +88,6 @@ elx_curia_list <- function(data = c("all","ecj_old","ecj_new","gc_all","cst_all"
 
   }
 
-
 }
 
 #' Curia scraper function
@@ -118,16 +116,16 @@ elx_curia_scraper <- function(url, ...){
     dplyr::mutate(dplyr::across(.cols = dplyr::everything(),
                                 .fns = ~dplyr::na_if(., ""))) %>%
     dplyr::filter(!is.na(.data$X1) & !is.na(.data$X2)) %>%
-    dplyr::rename(case_id = .data$X1,
-                  case_info = .data$X2) %>%
+    dplyr::rename(case_id = "X1",
+                  case_info = "X2") %>%
     dplyr::group_by(.data$case_id) %>%
     dplyr::mutate(n_id = dplyr::row_number()) %>%
     dplyr::ungroup()
 
   hrefs <- page %>%
     xml2::xml_find_all('//a[contains(@href, "numdoc")]')
 
-  linked_id <- rvest::html_text(hrefs, "href")
+  linked_id <- rvest::html_text(hrefs, trim = TRUE)
 
   linked_celex <- rvest::html_attr(hrefs, "href") %>%
     stringr::str_extract("numdoc=.*") %>%
@@ -141,7 +139,7 @@ elx_curia_scraper <- function(url, ...){
 
   out <- dplyr::left_join(tab, linked, by = c("case_id"="linked_id","n_id"="n_id")) %>%
     dplyr::select("case_id", "linked_celex", "case_info") %>%
-    dplyr::rename(case_id_celex = linked_celex)
+    dplyr::rename(case_id_celex = "linked_celex")
 
   return(out)
 

R/elx_run_query.R

@@ -59,6 +59,11 @@ graceful_http <- function(remote_file, headers = NULL, body = NULL,
     return(invisible(NULL))
   }
 
+  # if missing verb pick GET
+  if (missing(verb)){
+    verb <- "GET"
+  }
+  
   # Make the HTTP request based on the verb
   make_request <- function(verb) {
     tryCatch({
@@ -72,25 +77,21 @@ graceful_http <- function(remote_file, headers = NULL, body = NULL,
       }
     },
     error = function(e) {
-      message("Error: ", conditionMessage(e))
+      message(conditionMessage(e), " (Error)")
       return(invisible(NULL))
     },
     warning = function(w) {
-      message("Warning: ", conditionMessage(w))
+      message(conditionMessage(w), " (Warning)")
       return(invisible(NULL))
     })
   }
 
   # Execute the request
   resp <- make_request(verb)
 
-  # Check for HTTP errors
-  if (httr::http_error(resp)) {
-    httr::message_for_status(resp)
-    return(invisible(NULL))
-  }
-  
+  # return
   return(resp)
+  
 }
 
 #' Parse RDF/XML triplets to data frame

docs/404.html

@@ -0,0 +1,5 @@
+HTMLWidgets.widget({
+  name: 'girafe',
+  type: 'output',
+  factory: ggiraphjs.factory(HTMLWidgets.shinyMode)
+});