Merge pull request #10 from mrchypark/cpp11-package

change to cpp11 package

.Rbuildignore

@@ -11,3 +11,8 @@ auto*
 inst/ModelGenerator*
 ^\.github$
 ^_pkgdown\.yml$
+kiwilibtmp/*
+kiwilibs/*
+^model*
+^windows/*
+^tools/winbuild*

.github/workflows/check-full.yaml

@@ -25,16 +25,14 @@ jobs:
           - {os: macOS-latest,   r: 'release'}
 
           - {os: windows-latest, r: 'release'}
-          # Use 3.6 to trigger usage of RTools35
-          - {os: windows-latest, r: '3.6'}
+          - {os: windows-latest, r: 'oldrel-1'}
 
           # Use older ubuntu to maximise backward compatibility
           - {os: ubuntu-18.04,   r: 'devel', http-user-agent: 'release'}
           - {os: ubuntu-18.04,   r: 'release'}
           - {os: ubuntu-18.04,   r: 'oldrel-1'}
           - {os: ubuntu-18.04,   r: 'oldrel-2'}
           - {os: ubuntu-18.04,   r: 'oldrel-3'}
-          - {os: ubuntu-18.04,   r: 'oldrel-4'}
 
     env:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/pkgdown.yaml

@@ -27,7 +27,7 @@ jobs:
         run: |
           install.packages('remotes')
           remotes::install_deps(dependencies = TRUE)
-          install.packages("pkgdown", type = "binary")
+          install.packages("pkgdown")
           packageVersion("pkgdown")
           remotes::install_github("amirmasoudabdol/preferably", type = "source")
           packageVersion("preferably")

.gitignore

@@ -3,7 +3,12 @@
 .RData
 config.log
 config.status
-Kiwi*
+^Kiwi*
 src/Makevars
 inst/ModelGenerator*
-inst/*
+inst/model/*
+kiwilibs/*
+kiwilibtmp/*
+kiwi-model.tgz
+model*
+windows/*

DESCRIPTION

@@ -1,13 +1,15 @@
 Package: elbird
-Title: Blazing Fast Morphological Analyzer based on kiwi(Korean Intelligent Word Identifier)
+Title: Blazing Fast Morphological Analyzer Based on Kiwi(Korean Intelligent Word Identifier)
 Version: 0.1.0
 Authors@R: person(given = "Chanyub",
              family = "Park",
              role = c("aut","cre"),
              email = "mrchypark@gmail.com",
              comment = c(ORCID = "0000-0001-6474-2570"))
-Description: This is the R wrapper package Kiwi(Korean Intelligent Word Identifier), a blazing fast speed morphological analyzer for Korean. 
-             It supports configuration of user dictionary and detection of unregistered nouns based on frequency.
+Description: This is the R wrapper package Kiwi(Korean Intelligent Word Identifier), 
+             a blazing fast speed morphological analyzer for Korean. 
+             It supports configuration of user dictionary and detection of 
+             unregistered nouns based on frequency.
 License: LGPL (>= 3)
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
@@ -16,10 +18,17 @@ URL: https://github.com/mrchypark/elbird/
 BugReports: https://github.com/mrchypark/elbird/issues
 SystemRequirements: c("C++11", "git", "wget", "cmake")
 Depends: 
-    R (>= 3.3)
-LazyData: true
+    R (>= 3.5)
 Imports: 
     dplyr,
     purrr,
-    methods,
-    tibble
+    tibble,
+    R6 (>= 2.4.0),
+    vroom,
+    matchr
+LinkingTo: 
+    cpp11
+Suggests: 
+    testthat (>= 3.0.0)
+Config/testthat/edition: 3
+Config/testthat/parallel: true

NAMESPACE

@@ -1,25 +1,32 @@
 # Generated by roxygen2: do not edit by hand
 
+export(Kiwi)
+export(Match)
+export(Stopwords)
+export(Tags)
 export(analyze)
-export(kiwi_model_path)
-export(model_is_set)
+export(get_model)
+export(model_exists)
+export(model_home)
+export(model_works)
+export(split_into_sents)
 export(tokenize)
 export(tokenize_tbl)
 export(tokenize_tibble)
 export(tokenize_tidy)
 export(tokenize_tidytext)
 export(tokenize_tt)
+importFrom(R6,R6Class)
+importFrom(dplyr,anti_join)
 importFrom(dplyr,bind_rows)
-importFrom(methods,new)
+importFrom(dplyr,mutate)
+importFrom(matchr,Enum)
 importFrom(purrr,map)
 importFrom(purrr,map_chr)
 importFrom(purrr,map_int)
 importFrom(tibble,tibble)
 importFrom(utils,download.file)
 importFrom(utils,untar)
+importFrom(vroom,vroom)
+importFrom(vroom,vroom_write)
 useDynLib(elbird, .registration = TRUE)
-useDynLib(elbird,kiwi_analyze_)
-useDynLib(elbird,kiwi_clear_error_)
-useDynLib(elbird,kiwi_error_)
-useDynLib(elbird,kiwi_init_)
-useDynLib(elbird,kiwi_version_)

R/analyze.R

@@ -1,18 +1,31 @@
 #' analyze
 #'
-#' @param text target text.
-#' @param top_n Number of result. default is 3.
-#' @name analyze
+#' @param top_n  \code{integer}: Number of result. Default is 3.
+#' @inheritParams tokenize
+#' @examples
+#' \dontrun{
+#'   analyze("Test text.")
+#'   analyze("Please use Korean.", top_n = 1)
+#'   analyze("Test text.", 1, Match$ALL_WITH_NORMALIZING)
+#'   analyze("Test text.", stopwords = FALSE)
+#'   analyze("Test text.", stopwords = TRUE)
+#'   analyze("Test text.", stopwords = "user_dict.txt")
+#'   analyze("Test text.", stopwords = Stopwords$new(TRUE))
+#' }
 #' @export
-analyze <- function(text, top_n = 3) {
-  if (init_chk_not())
-    init()
+analyze <-
+  function(text,
+           top_n = 3,
+           match_option = Match$ALL,
+           stopwords = FALSE) {
+    if (init_chk_not())
+      init()
 
-  return(
-    kiwi_analyze(
+    kiwi_analyze_wrap(
       get("kb", envir = .el),
       text,
-      top_n, 1
+      top_n,
+      match_option,
+      stopwords
     )
-  )
-}
+  }

R/cpp11.R

@@ -0,0 +1,73 @@
+# Generated by cpp11: do not edit by hand
+
+kiwi_version_ <- function() {
+  .Call(`_elbird_kiwi_version_`)
+}
+
+kiwi_error_ <- function() {
+  .Call(`_elbird_kiwi_error_`)
+}
+
+kiwi_clear_error_ <- function() {
+  invisible(.Call(`_elbird_kiwi_clear_error_`))
+}
+
+kiwi_builder_close_ <- function(handle_ex) {
+  .Call(`_elbird_kiwi_builder_close_`, handle_ex)
+}
+
+kiwi_builder_init_ <- function(model_path, num_threads, options) {
+  .Call(`_elbird_kiwi_builder_init_`, model_path, num_threads, options)
+}
+
+kiwi_builder_add_word_ <- function(handle_ex, word, pos, score) {
+  .Call(`_elbird_kiwi_builder_add_word_`, handle_ex, word, pos, score)
+}
+
+kiwi_builder_add_alias_word_ <- function(handle_ex, alias, pos, score, orig_word) {
+  .Call(`_elbird_kiwi_builder_add_alias_word_`, handle_ex, alias, pos, score, orig_word)
+}
+
+kiwi_builder_add_pre_analyzed_word_ <- function(handle_ex, form, analyzed_r, score) {
+  .Call(`_elbird_kiwi_builder_add_pre_analyzed_word_`, handle_ex, form, analyzed_r, score)
+}
+
+kiwi_builder_load_dict_ <- function(handle_ex, dict_path) {
+  .Call(`_elbird_kiwi_builder_load_dict_`, handle_ex, dict_path)
+}
+
+kiwi_close_ <- function(handle_ex) {
+  .Call(`_elbird_kiwi_close_`, handle_ex)
+}
+
+kiwi_builder_extract_words_ <- function(handle_ex, input, min_cnt, max_word_len, min_score, pos_threshold) {
+  .Call(`_elbird_kiwi_builder_extract_words_`, handle_ex, input, min_cnt, max_word_len, min_score, pos_threshold)
+}
+
+kiwi_builder_extract_add_words_ <- function(handle_ex, input, min_cnt, max_word_len, min_score, pos_threshold) {
+  .Call(`_elbird_kiwi_builder_extract_add_words_`, handle_ex, input, min_cnt, max_word_len, min_score, pos_threshold)
+}
+
+kiwi_builder_build_ <- function(handle_ex) {
+  .Call(`_elbird_kiwi_builder_build_`, handle_ex)
+}
+
+kiwi_init_ <- function(model_path, num_threads, options) {
+  .Call(`_elbird_kiwi_init_`, model_path, num_threads, options)
+}
+
+kiwi_set_option_ <- function(handle_ex, option, value) {
+  invisible(.Call(`_elbird_kiwi_set_option_`, handle_ex, option, value))
+}
+
+kiwi_get_option_ <- function(handle_ex, option) {
+  .Call(`_elbird_kiwi_get_option_`, handle_ex, option)
+}
+
+kiwi_analyze_ <- function(handle_ex, text, top_n, match_options, stopwords_r) {
+  .Call(`_elbird_kiwi_analyze_`, handle_ex, text, top_n, match_options, stopwords_r)
+}
+
+kiwi_split_into_sents_ <- function(handle_ex, text, match_options, return_tokens) {
+  .Call(`_elbird_kiwi_split_into_sents_`, handle_ex, text, match_options, return_tokens)
+}

R/dictionary.R

@@ -0,0 +1,17 @@
+dict_path <- function() {
+  path <- Sys.getenv("ELBIRD_DICTIONARY_HOME")
+  if (nzchar(path)) {
+    normalizePath(path, mustWork = FALSE)
+  } else {
+    normalizePath(file.path(system.file("", package = "elbird"), "dicts"),
+                  mustWork = FALSE)
+  }
+}
+
+dict_stopwords_path <- function() {
+  normalizePath(file.path(dict_path(), "stopwords.txt"), mustWork = FALSE)
+}
+
+dict_user_path <- function() {
+  normalizePath(file.path(dict_path(), "stopwords.txt"), mustWork = FALSE)
+}

R/init.R

@@ -2,19 +2,18 @@ init_chk_not <- function() {
   length(ls(envir = .el)) != 1
 }
 
-#' @importFrom methods new
-init <- function() {
-  if (!model_exists())
-    get_model_file()
+init <- function(size = "small") {
+  if (!kiwi_model_exists(size))
+    get_kiwi_models(size)
 
-  kb <- kiwi_init(model_path_full(), 0, 0)
+  kb <- kiwi_init_(kiwi_model_path_full(size), 0, BuildOpt$DEFAULT)
+  err <- kiwi_error_wrap()
 
-  if (identical(kb, new("externalptr"))) {
-    tem <- kiwi_error()
-    kiwi_clear_error()
-    stop(tem)
-  }
+  if (!is.null(err))
+    stop(err)
 
   assign("kb", kb, envir = .el)
 }
 
+
+