feat: translate long text

NAMESPACE

@@ -6,6 +6,7 @@ export(create_transliteration_table)
 export(google_get_supported_languages)
 export(google_is_valid_language_code)
 export(google_translate)
+export(google_translate_long_text)
 export(google_transliterate)
 export(language_detect)
 export(linguee_external_sources)

R/google_translate_long_text.R

@@ -0,0 +1,63 @@
+#' Translate long text using Google Translate
+#'
+#' This function translates long text from one language to another using Google Translate.
+#' It splits the text into smaller chunks if necessary to handle large inputs.
+#'
+#' @param text The long text to translate. Should be a single string.
+#' @param target_language The language to translate the text into. Default is "en" for English.
+#' @param source_language The language of the input text. Default is "auto" for automatic detection.
+#' @param chunk_size The maximum number of characters to send in a single translation request. Default is 1000.
+#'
+#' @return A single string containing the translated text.
+#' @export
+#'
+#' @examples
+#' \donttest{
+#' long_text <- paste(rep("This is a long text to translate.", 100), collapse = " ")
+#' google_translate_long_text(
+#' long_text, target_language = "de",
+#' source_language = "en",
+#' chunk_size = 500)
+#' }
+google_translate_long_text <- function(text, target_language = "en", source_language = "auto", chunk_size = 1000) {
+  if (!google_is_valid_language_code(target_language)) {
+    stop("Invalid target language code.")
+  }
+  if (!google_is_valid_language_code(source_language)) {
+    stop("Invalid source language code.")
+  }
+
+  # Function to split text into chunks
+  split_text <- function(text, chunk_size) {
+    split_indices <- seq(1, nchar(text), by = chunk_size)
+    sapply(split_indices, function(i) substr(text, i, i + chunk_size - 1))
+  }
+
+  # Split text into chunks if it's too long
+  if (nchar(text) > chunk_size) {
+    text_chunks <- split_text(text, chunk_size)
+  } else {
+    text_chunks <- list(text)
+  }
+
+  # Translate each chunk
+  translations <- sapply(text_chunks, function(chunk) {
+    formatted_text <- urltools::url_encode(chunk)
+    formatted_link <- paste0(
+      "https://translate.google.com/m?tl=",
+      target_language, "&sl=", source_language,
+      "&q=", formatted_text
+    )
+
+    response <- httr::GET(formatted_link)
+    translation <- httr::content(response) %>%
+      rvest::html_nodes("div.result-container") %>%
+      rvest::html_text()
+
+    translation <- urltools::url_decode(translation)
+    gsub("\n", "", translation)
+  })
+
+  # Combine translated chunks
+  paste(translations, collapse = " ")
+}

R/wmcloud_translate.R

@@ -51,7 +51,7 @@ wmcloud_translate <- function(content,
   valid_formats <- c("json", "markdown", "text", "webpage")
 
   # List of valid models
-  valid_models <- c("nllb200-600M") # Add more models here
+  valid_models <- c("nllb200-600M", "nllb-wikipedia", "opusmt-en-bi", "opusmt-en-bcl", "opusmt-en-to", "opusmt-en-chr", "opusmt-en-guw", "opusmt-en-srn", "opusmt-en-ty", "opusmt-en-ve", "opusmt-sv-fi", "softcatala", "indictrans2-indic-en", "indictrans2-en-indic", "indictrans2-indic-indic", "madlad-400")
 
   # Check if format and model are valid
   if (!format %in% valid_formats) {

_pkgdown.yml

@@ -13,6 +13,7 @@ reference:
       - google_is_valid_language_code
       - google_supported_languages
       - google_translate
+      - google_translate_long_text
       - google_transliterate
       - language_detect
       - translate_file

tests/testthat/test-google_translate.R

@@ -25,24 +25,7 @@ test_that("google_translate returns correct translation for unvectorized input",
   expect_equal(translation, expected_translation)
 })
 
-# Unit tests for special characters from French to English
-test_that("google translate returns translations with special characters", {
-  text_to_translate <- "La Saône prend sa source à Vioménil dans les pré-Vosges à 405 m d'altitude. La rivière conflue avec le Rhône 473,3 km plus loin."
-  translation <- google_translate(text_to_translate, target_language = "en")
 
-  expected_translation <- "The Saône has its source at Vioménil in the pre-Vosges at an altitude of 405 m. The river confluences with the Rhône 473.3 km further."
-  expect_equal(translation, expected_translation)
-})
-
-
-# Unit tests for special characters from Arabic to English
-test_that("google translate returns translations with special characters", {
-  text_to_translate <- "يتدفقُ النيل عبر الصحراء السودانية إلى مصر باتجاه الشمال ويمر في مدينةُ القاهرة الواقعة على دلتا النهر الكبيرة (دلتا النيل)، ثم يعبر النهر مدينتي دمياط ورشيد ويصب ..."
-  translation <- google_translate(text_to_translate, target_language = "en")
-
-  expected_translation <- "The Nile flows through the Sudanese desert to Egypt towards the north and passes through the city of Cairo, located on the large river delta (Nile Delta), then the river crosses the cities of Damietta and Rosetta and flows..."
-  expect_equal(translation, expected_translation)
-})
 
 # Unit tests for checking supported languages in Google Translate
 test_that("google_is_valid_language_code: valid codes", {