added emoji extraction and emoji categorize

Author: Youzhi Yu

DESCRIPTION

@@ -17,4 +17,5 @@ Imports:
     purrr,
     stringr,
     tibble,
+    tidyr,
     utils

NAMESPACE

@@ -1,5 +1,8 @@
 # Generated by roxygen2: do not edit by hand
 
+export(emoji_categorize)
+export(emoji_extract_nest)
+export(emoji_extract_unnest)
 export(emoji_summary)
 export(emoji_tweets)
 export(top_n_emojis)
@@ -8,4 +11,5 @@ import(emoji)
 import(purrr)
 import(stringr)
 import(tibble)
+import(tidyr)
 import(utils)

R/emoji-categorize.R

@@ -0,0 +1,43 @@
+emoji_category_add <- function(emoji_unicodes, emoji_category, tweet_tbl, tweet_text){
+
+  tweet_tbl %>%
+    dplyr::filter(str_detect({{ tweet_text }}, emoji_unicodes)) %>%
+    dplyr::mutate(.emoji_category = emoji_category)
+
+}
+
+
+
+#' Categorize Emoji Tweets/text based on Emoji category
+#'
+#' Users can use \code{emoji_categorize} to see the all the categories each
+#' Emoji Tweet has. The function preserves the input data structure, and the
+#' only change is it adds an extra column with information about Emoji
+#' category separated by \code{|} if there is more than one category.
+#'
+#' @inheritParams emoji_summary
+#' @import purrr
+#' @import tidyr
+#' @import dplyr
+#' @return A filtered dataframe with the presence of Emoji only, and with an
+#' extra column \code{.emoji_category}.
+#' @export
+#'
+
+
+emoji_categorize <- function(tweet_tbl, tweet_text) {
+
+  purrr::map2_dfr(tidyEmoji::category_unicode_crosswalk$unicodes,
+           tidyEmoji::category_unicode_crosswalk$category,
+           emoji_category_add,
+           tweet_tbl,
+           {{ tweet_text }}) %>%
+    tidyr::pivot_wider(names_from = .emoji_category,
+                       values_from = .emoji_category) %>%
+    tidyr::unite(".emoji_category", c("Smileys & Emotion": "Flags"), sep = "|", na.rm = T)
+
+
+}
+
+
+

R/emoji-extraction.R

@@ -0,0 +1,55 @@
+#' Emoji extraction unnested summary
+#'
+#' If users would like to know how many Emojis and what kinds of Emojis each
+#' Tweet has, \code{emoji_extract} is a useful function to output a global
+#' summary with the row number of each Tweet containing Emoji and the Unicodes
+#' associated with each Tweet.
+#'
+#' @inheritParams emoji_summary
+#'
+#' @import dplyr
+#' @import stringr
+#' @import tidyr
+#' @return A summary tibble with the original row number and Emoji count.
+#' @export
+#'
+emoji_extract_unnest <- function(tweet_tbl, tweet_text){
+  tweet_tbl %>%
+    tidyEmoji::emoji_extract_nest({{ tweet_text }}) %>%
+    dplyr::select({{ tweet_text }}, .emoji_unicode) %>%
+    dplyr::mutate(row_number = dplyr::row_number()) %>%
+    tidyr::unnest(.emoji_unicode) %>%
+    dplyr::group_by(row_number, .emoji_unicode) %>%
+    dplyr::summarize(emoji_count = dplyr::n()) %>%
+    dplyr::ungroup()
+
+}
+
+
+
+
+
+
+
+#' Emoji extraction nested summary
+#'
+#' This function adds an extra list column called \code{.emoji_unicode} to the
+#' original data, with all Emojis included.
+#'
+#' @inheritParams emoji_summary
+#'
+#' @import dplyr
+#' @import stringr
+#' @import emoji
+#' @return The original dataframe/tibble with an extra column collumn called
+#' \code{.emoji_unicode}.
+#' @export
+#'
+emoji_extract_nest <- function(tweet_tbl, tweet_text){
+  tweet_tbl %>%
+    dplyr::mutate(.emoji_unicode = stringr::str_extract_all({{ tweet_text }}, emoji::emojis %>%
+                                                             dplyr::filter(!str_detect(name, "keycap: \\*")) %>%
+                                                             dplyr::pull(emoji) %>%
+                                                             paste(., collapse = "|")))
+
+}

R/tidyEmoji.R

@@ -13,4 +13,6 @@ if(getRversion() >= "2.15.1")  utils::globalVariables(c(".",
                                                         "emoji_name",
                                                         "unicode",
                                                         "emoji_category",
-                                                        "emoji_unicode_crosswalk"))
+                                                        ".emoji_category",
+                                                        ".emoji_unicode",
+                                                        ".emoji_unicode"))

R/top-n-emojis.R

@@ -24,16 +24,16 @@
 
 top_n_emojis <- function(tweet_tbl, tweet_text, n = 20, duplicated_unicode = "no"){
 
-  emoji_tbl <- emoji_tweets(tweet_tbl, {{ tweet_text }})
+  emoji_tbl <- tidyEmoji::emoji_tweets(tweet_tbl, {{ tweet_text }})
 
-  emoji_count_list <- purrr::map(emoji_unicode_crosswalk$unicode,
+  emoji_count_list <- purrr::map(tidyEmoji::emoji_unicode_crosswalk$unicode,
                                  .f = count_each_emoji,
                                  emoji_tbl,
                                  {{ tweet_text }})
 
-  tbl <- tibble::tibble(unicode = emoji_unicode_crosswalk$unicode,
+  tbl <- tibble::tibble(unicode = tidyEmoji::emoji_unicode_crosswalk$unicode,
                         emoji_count = unlist(emoji_count_list)) %>%
-    dplyr::inner_join(emoji_unicode_crosswalk, by = "unicode") %>%
+    dplyr::inner_join(tidyEmoji::emoji_unicode_crosswalk, by = "unicode") %>%
     dplyr::distinct() %>%
     dplyr::count(emoji_name, unicode, emoji_category, wt = emoji_count, sort = T)