version 0.0.2

cran · Jun 8, 2023 · 6ac0d7b · 6ac0d7b
1 parent 78d295a
commit 6ac0d7b
Show file tree

Hide file tree

Showing 18 changed files with 133 additions and 106 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: RcppJagger
 Title: An R Wrapper for Jagger
-Version: 0.0.1
+Version: 0.0.2
 Authors@R: c(
         person("Shusei", "Eshima", email = "shuseieshima@gmail.com", role = c("aut","cre"), comment = c(ORCID = "0000-0003-3613-4046")),
         person("Naoki", "Yoshinaga", role = c("ctb"))
@@ -21,9 +21,9 @@ Suggests: dplyr (>= 1.1.0), testthat (>= 3.1.5), tibble
 Config/testthat/edition: 3
 LazyData: TRUE
 NeedsCompilation: yes
-Packaged: 2023-06-04 13:28:58 UTC; shusei
+Packaged: 2023-06-08 19:03:16 UTC; shusei
 Author: Shusei Eshima [aut, cre] (<https://orcid.org/0000-0003-3613-4046>),
   Naoki Yoshinaga [ctb]
 Maintainer: Shusei Eshima <shuseieshima@gmail.com>
 Repository: CRAN
-Date/Publication: 2023-06-06 07:00:06 UTC
+Date/Publication: 2023-06-08 22:22:56 UTC
diff --git a/MD5 b/MD5
@@ -1,38 +1,38 @@
-e4ab7751d95cc499025f796dd5835db0 *DESCRIPTION
+a6bb715255a378bab04efe6c7071c538 *DESCRIPTION
 8bb48ef08475ae0cb3e0852924bd54ee *NAMESPACE
-72068cd54fcf95d6eae221b46120eb8c *NEWS.md
-f9f7af660472a402911b86dab482e9b2 *R/RcppExports.R
+40b0fd51a66ff0f4fe0e07c08e997208 *NEWS.md
+1eab0d1046382d4e44cb01d8bce7ec97 *R/RcppExports.R
 dc6f159d77512c9b39bb9df2a6211966 *R/RcppJagger.R
 49ab70a4e1432ae366ac3f1b3ed573a8 *R/data.R
-a1123c56bfe1d024bd9b4177d2a03e0f *R/lemmatizer.R
-9666151174d17bfd01a2463aa09644d6 *R/pos.R
-00ca98c3b4ce1f0136935b29699e3d79 *R/tokenizer.R
+02a4b2b92d572134845d668ae12aaa69 *R/lemmatizer.R
+ba70802ce80891d3a8977e3cb718257a *R/pos.R
+3b4efa985838a5b4e3718edfe556766c *R/tokenizer.R
 80fcba092843dccd645184befeb5c35f *R/util.R
 fa7b592dc30adfd4d804f6b365c728a3 *data/sentence_example.rda
 6429b333f6879da564935edf00802f7c *inst/include/ccedar_core.h
 83aa264c4503d476805cae347dfcf226 *inst/include/jagger.cc
 1ff928aad2d761f6580473fda8bba1d2 *inst/include/jagger.h
 bb5e4ec24ecaed6c71be0d76836eedba *inst/include/mman.h
-b296159a609862232c8fc2136140e08d *man/lemmatize.Rd
+a5abcde7cded3808464f4e3c118fc7ef *man/lemmatize.Rd
 9cfd165a7b6b42bc8928c7bc49a51ec8 *man/lemmatize_cpp_vec.Rd
-329a9e0c25db5fc2b250ef78661736a4 *man/lemmatize_tbl.Rd
+7c59709815d427b02ffaf6247cb6198c *man/lemmatize_tbl.Rd
 9cfc71ce88c41c1194b2426fe7e0fef3 *man/pos.Rd
-d41d1ec436d7777c0dd5477a6a26982f *man/pos_cpp_vec.Rd
+b6d398dc9ed399dd1cf16d4eab24c1b6 *man/pos_cpp_vec.Rd
 910b500dd766df96908f2908c035fa8f *man/pos_simple.Rd
-698357db8eb2731dabf6f6578dfe4e88 *man/pos_simple_cpp_vec.Rd
+859e55c9f42c04894903b45d5315074a *man/pos_simple_cpp_vec.Rd
 f026c0964eac7216fbe1650737573bae *man/sentence_example.Rd
-7abe7d690f6471024dea9ee17a898ee1 *man/tokenize.Rd
+1c20a20a1ac4b8a95febac0e75fc6dbe *man/tokenize.Rd
 a3f87da0a42e58fe8b2bb48ce563913d *man/tokenize_cpp_vec.Rd
 5ad3d74d8e56c53ad324184d55cf28bf *man/tokenize_tbl.Rd
-c1b4a373e6630576f922d4391312cb5b *src/RcppExports.cpp
-b020b7503215ce3884f7d19e4516ab5f *src/lemmatizer.h
-8fd27335b165bca5f2a1aa7f2ca99863 *src/main.cpp
-a8ed2fb3d8b00a7c1c70298d904b662e *src/pos.h
+01a357bad718092e0beb2a561afebdc4 *src/RcppExports.cpp
+192d9de22d0dd43048cf3f60ed2e3323 *src/lemmatizer.h
+f6ee5d871a44fc6fa1e007bbc7c244fb *src/main.cpp
+9272de4b3ee203551d47a5cfea916a71 *src/pos.h
 a34ca3124494098634b0a434998a8106 *src/pos_simple.h
 dd9a9a7825a8e4764ea26282793a1616 *src/reader.h
 2294072a6814593ae4b60c6f4d948f65 *src/tokenizer.h
 1199cbae84522095d5b3c49bebbadfb9 *tests/testthat.R
 e77d910f914dea04b374c5759f2c9b0f *tests/testthat/test-lemmatizer.R
-e5dd6c1eacfb9c5f52fd05ad18429538 *tests/testthat/test-pos.R
-1ccbc80765a129be582e2d3db1e677cc *tests/testthat/test-sanity.R
+9dd4dd39f70331fa5a933d328af45328 *tests/testthat/test-pos.R
+7416e5a9070a9c6bfa6a59fc713cc411 *tests/testthat/test-sanity.R
 91b0056a14954bcfb9d4be16bc1fa48a *tests/testthat/test-tokenizer.R
diff --git a/NEWS.md b/NEWS.md
@@ -1,2 +1,5 @@
+# RcppJagger 0.0.2
+* Improvement of performance and stability.
+
 # RcppJagger 0.0.1
-* First submission to CRAN
+* First submission to CRAN.
diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -4,15 +4,15 @@
 #' POS tagging in C++
 #'
 #' @keywords internal
-pos_cpp_vec <- function(inputs, model_path) {
-    .Call('_RcppJagger_pos_cpp_vec', PACKAGE = 'RcppJagger', inputs, model_path)
+pos_cpp_vec <- function(inputs, model_path, keep_vec, keep_all) {
+    .Call('_RcppJagger_pos_cpp_vec', PACKAGE = 'RcppJagger', inputs, model_path, keep_vec, keep_all)
 }
 
 #' POS tagging in C++ (only token and pos)
 #'
 #' @keywords internal
-pos_simple_cpp_vec <- function(inputs, model_path) {
-    .Call('_RcppJagger_pos_simple_cpp_vec', PACKAGE = 'RcppJagger', inputs, model_path)
+pos_simple_cpp_vec <- function(inputs, model_path, keep_vec, keep_all) {
+    .Call('_RcppJagger_pos_simple_cpp_vec', PACKAGE = 'RcppJagger', inputs, model_path, keep_vec, keep_all)
 }
 
 #' Tokenizer (a vector input)

diff --git a/R/lemmatizer.R b/R/lemmatizer.R
@@ -4,7 +4,7 @@
 #' @param model_path  a path to the model.
 #' @param keep a vector of POS(s) to keep. Default is `NULL`.
 #' @param concat logical. If TRUE, the function returns a concatenated string. Default is `TRUE`.
-#' @return a list.
+#' @return a vector (if `concat = TRUE`) or a list (if `concat = FALSE`).
 #' @examples
 #'  data(sentence_example)
 #'  res_lemmatize <- lemmatize(sentence_example$text)
@@ -40,7 +40,7 @@ lemmatize <- function(input, model_path = NULL, keep = NULL, concat = TRUE) {
 #' @param column a column name of the tibble to tokenize.
 #' @param model_path  a path to the model.
 #' @param keep a vector of POS(s) to keep. Default is `NULL`.
-#' @return a vector.
+#' @return a tibble.
 #' @examples
 #'  data(sentence_example)
 #'  res_lemmatize <- lemmatize_tbl(tibble::as_tibble(sentence_example), "text")

diff --git a/R/pos.R b/R/pos.R
@@ -21,19 +21,15 @@ pos <- function(input, model_path = NULL, keep = NULL, format = c("list", "data.
   }
   format <- rlang::arg_match(format)
 
-  result <- pos_cpp_vec(input, model_path)
-
-  if (!is.null(keep)) {
-    result <- purrr::map(result, function(x) {
-      idx <- x$pos %in% keep
-      x$token <- x$token[idx]
-      x$lemma <- x$lemma[idx]
-      x$subtype <- x$subtype[idx]
-      x$pos <- x$pos[idx]
-      return(x)
-    })
+  if (is.null(keep)) {
+    keep_all <- TRUE
+    keep <- c("")
+  } else {
+    keep_all <- FALSE
   }
 
+  result <- pos_cpp_vec(input, model_path, keep, keep_all)
+
   if (format == "data.frame") {
     result <- purrr::map(result, function(x) {
       return(data.frame(
@@ -70,17 +66,16 @@ pos_simple <- function(input, model_path = NULL, keep = NULL, format = c("list",
   }
 
   format <- rlang::arg_match(format)
-  result <- pos_simple_cpp_vec(input, model_path)
 
-  if (!is.null(keep)) {
-    result <- purrr::map(result, function(x) {
-      idx <- x$pos %in% keep
-      x$token <- x$token[idx]
-      x$pos <- x$pos[idx]
-      return(x)
-    })
+  if (is.null(keep)) {
+    keep_all <- TRUE
+    keep <- c("")
+  } else {
+    keep_all <- FALSE
   }
 
+  result <- pos_simple_cpp_vec(input, model_path, keep, keep_all)
+
   if (format == "data.frame") {
     result <- purrr::map(result, function(x) {
       return(data.frame(

diff --git a/R/tokenizer.R b/R/tokenizer.R
@@ -4,7 +4,7 @@
 #' @param model_path  a path to the model.
 #' @param keep a vector of POS(s) to keep. Default is `NULL`.
 #' @param concat logical. If TRUE, the function returns a concatenated string. Default is `TRUE`.
-#' @return a list.
+#' @return a vector (if `concat = TRUE`) or a list (if `concat = FALSE`).
 #' @examples
 #'  data(sentence_example)
 #'  res_tokenize <- tokenize(sentence_example$text)

diff --git a/man/lemmatize.Rd b/man/lemmatize.Rd
diff --git a/man/lemmatize_tbl.Rd b/man/lemmatize_tbl.Rd
diff --git a/man/pos_cpp_vec.Rd b/man/pos_cpp_vec.Rd
diff --git a/man/pos_simple_cpp_vec.Rd b/man/pos_simple_cpp_vec.Rd
diff --git a/man/tokenize.Rd b/man/tokenize.Rd
diff --git a/src/RcppExports.cpp b/src/RcppExports.cpp
@@ -11,26 +11,30 @@ Rcpp::Rostream<false>& Rcpp::Rcerr = Rcpp::Rcpp_cerr_get();
 #endif
 
 // pos_cpp_vec
-List pos_cpp_vec(StringVector& inputs, std::string model_path);
-RcppExport SEXP _RcppJagger_pos_cpp_vec(SEXP inputsSEXP, SEXP model_pathSEXP) {
+List pos_cpp_vec(StringVector& inputs, std::string model_path, StringVector& keep_vec, bool keep_all);
+RcppExport SEXP _RcppJagger_pos_cpp_vec(SEXP inputsSEXP, SEXP model_pathSEXP, SEXP keep_vecSEXP, SEXP keep_allSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;
     Rcpp::RNGScope rcpp_rngScope_gen;
     Rcpp::traits::input_parameter< StringVector& >::type inputs(inputsSEXP);
     Rcpp::traits::input_parameter< std::string >::type model_path(model_pathSEXP);
-    rcpp_result_gen = Rcpp::wrap(pos_cpp_vec(inputs, model_path));
+    Rcpp::traits::input_parameter< StringVector& >::type keep_vec(keep_vecSEXP);
+    Rcpp::traits::input_parameter< bool >::type keep_all(keep_allSEXP);
+    rcpp_result_gen = Rcpp::wrap(pos_cpp_vec(inputs, model_path, keep_vec, keep_all));
     return rcpp_result_gen;
 END_RCPP
 }
 // pos_simple_cpp_vec
-List pos_simple_cpp_vec(StringVector& inputs, std::string model_path);
-RcppExport SEXP _RcppJagger_pos_simple_cpp_vec(SEXP inputsSEXP, SEXP model_pathSEXP) {
+List pos_simple_cpp_vec(StringVector& inputs, std::string model_path, StringVector& keep_vec, bool keep_all);
+RcppExport SEXP _RcppJagger_pos_simple_cpp_vec(SEXP inputsSEXP, SEXP model_pathSEXP, SEXP keep_vecSEXP, SEXP keep_allSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;
     Rcpp::RNGScope rcpp_rngScope_gen;
     Rcpp::traits::input_parameter< StringVector& >::type inputs(inputsSEXP);
     Rcpp::traits::input_parameter< std::string >::type model_path(model_pathSEXP);
-    rcpp_result_gen = Rcpp::wrap(pos_simple_cpp_vec(inputs, model_path));
+    Rcpp::traits::input_parameter< StringVector& >::type keep_vec(keep_vecSEXP);
+    Rcpp::traits::input_parameter< bool >::type keep_all(keep_allSEXP);
+    rcpp_result_gen = Rcpp::wrap(pos_simple_cpp_vec(inputs, model_path, keep_vec, keep_all));
     return rcpp_result_gen;
 END_RCPP
 }
@@ -64,8 +68,8 @@ END_RCPP
 }
 
 static const R_CallMethodDef CallEntries[] = {
-    {"_RcppJagger_pos_cpp_vec", (DL_FUNC) &_RcppJagger_pos_cpp_vec, 2},
-    {"_RcppJagger_pos_simple_cpp_vec", (DL_FUNC) &_RcppJagger_pos_simple_cpp_vec, 2},
+    {"_RcppJagger_pos_cpp_vec", (DL_FUNC) &_RcppJagger_pos_cpp_vec, 4},
+    {"_RcppJagger_pos_simple_cpp_vec", (DL_FUNC) &_RcppJagger_pos_simple_cpp_vec, 4},
     {"_RcppJagger_tokenize_cpp_vec", (DL_FUNC) &_RcppJagger_tokenize_cpp_vec, 4},
     {"_RcppJagger_lemmatize_cpp_vec", (DL_FUNC) &_RcppJagger_lemmatize_cpp_vec, 4},
     {NULL, NULL, 0}

diff --git a/src/lemmatizer.h b/src/lemmatizer.h
@@ -81,7 +81,7 @@ class RcppJaggerLemmatize : public jagger::tagger {
     // Add the final part after the last comma to `parts`.
     parts.emplace_back(pos_info.substr(start));
 
-    if (parts[0] != "*" && parts.size() >= 7) {  // first appearance of the token (i.e. not a concatenation)
+    if (parts[0] != "*" && parts.size() >= 6) {  // first appearance of the token (i.e. not a concatenation)
       pos_vec.emplace_back(parts[0]);
       lemma_vec.emplace_back(parts[parts.size() - 3]);
     } else if (parts[0] != "*" && parts.size() == 4) {  // concatenation