perf: remove reset_encoding

Author: paithiov909

.Rbuildignore

@@ -1,12 +1,14 @@
 ^.*\.Rproj$
 ^\.Rproj\.user$
 ^\.github$
+^\.vscode$
 ^\.Rprofile$
 ^\.covrignore$
 ^doc$
 ^Meta$
 ^docs$
 ^codecov\.yml$
+^renovate\.json$
 ^README\.Rmd$
 ^crfpp$
 ^cabocha$

.gitignore

@@ -42,3 +42,4 @@ Meta
 *.log
 *.pdf
 
+.vscode

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: pipian
 Title: Tiny Interface to CaboCha for R
-Version: 0.3.7
+Version: 0.3.8
 Authors@R: c(
     person("Akiru", "Kato", , "paithiov909@gmail.com", role = c("aut", "cre")),
     person("Marcin", "Kalicinski", role = "aut",
@@ -15,9 +15,9 @@ License: MIT + file LICENSE
 URL: https://github.com/paithiov909/pipian,
     https://paithiov909.github.io/pipian/
 BugReports: https://github.com/paithiov909/pipian/issues
-Depends: 
+Depends:
     R (>= 4.0.0)
-Imports: 
+Imports:
     dplyr,
     igraph,
     purrr,
@@ -32,9 +32,8 @@ Suggests:
     roxygen2,
     spelling,
     testthat (>= 3.0.0)
-LinkingTo: 
-    Rcpp,
-    RcppThread
+LinkingTo:
+    Rcpp
 Config/testthat/edition: 3
 Encoding: UTF-8
 Language: en-US

NEWS.md

@@ -1,3 +1,7 @@
+# pipian 0.3.8
+
+* For performance, now skips resetting the output encodings to UTF-8.
+
 # pipian 0.3.4
 
 * Dropped `processx` dependency.

R/RcppExports.R

@@ -7,5 +7,5 @@ parse_xml <- function(path) {
 
 # Register entry points for exported C++ functions
 methods::setLoadAction(function(ns) {
-    .Call('_pipian_RcppExport_registerCCallable', PACKAGE = 'pipian')
+    .Call(`_pipian_RcppExport_registerCCallable`)
 })

R/cabocha-parser.R

@@ -58,12 +58,12 @@ ppn_parse_xml <- function(path,
             sentence_id = as.integer(.data$sentence_id) + 1L,
             chunk_id = as.integer(.data$chunk_id) + 1L,
             token_id = as.integer(.data$token_id),
-            token = reset_encoding(.data$token),
+            token = .data$token,
             chunk_link = as.integer(.data$chunk_link) + 1L,
             chunk_score = as.numeric(.data$chunk_score),
             chunk_head = as.integer(.data$chunk_head) + 1L,
             chunk_func = as.integer(.data$chunk_func),
-            token_entity = dplyr::na_if(reset_encoding(.data$token_entity), "O")
+            token_entity = dplyr::na_if(.data$token_entity, "O")
           ) %>%
           dplyr::select(!"token_feature") %>%
           dplyr::bind_cols(features) %>%

R/utils.R

@@ -1,15 +1,3 @@
-#' Format character vector
-#' @param vec Character vector.
-#' @param enc Encoding of `vec`.
-#' @returns Formatted character vector is returned.
-#' @keywords internal
-reset_encoding <- function(vec, enc = "UTF-8") {
-  unlist(lapply(vec, function(elem) {
-    Encoding(elem) <- enc
-    return(elem)
-  }))
-}
-
 #' Check if CaboCha command is available
 #' @returns Logical.
 #' @keywords internal

inst/include/pipian_RcppExports.h

@@ -4,7 +4,6 @@
 #ifndef RCPP_pipian_RCPPEXPORTS_H_GEN_
 #define RCPP_pipian_RCPPEXPORTS_H_GEN_
 
-#include <RcppThread.h>
 #include <Rcpp.h>
 
 namespace pipian {

man/reset_encoding.Rd

@@ -2,7 +2,6 @@
 // Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
 
 #include "../inst/include/pipian.h"
-#include <RcppThread.h>
 #include <Rcpp.h>
 #include <string>
 #include <set>

src/RcppExports.cpp

@@ -1,6 +1,3 @@
-// [[Rcpp::depends(RcppThread)]]
-#define RCPPTHREAD_OVERRIDE_THREAD 1
-
 #include <Rcpp.h>
 #include <fstream>
 #include "./rapidxml-1.13/rapidxml.hpp"

src/parsexml.cpp

@@ -1,5 +1,4 @@
-skip_if_not(is_cabocha_available())
-
 test_that("can execute cabocha", {
+  skip_if_not(is_cabocha_available())
   expect_type(ppn_cabocha(enc2utf8("\u3053\u3093\u306b\u3061\u306f")), "character")
 })