From 583c00902c89d5c6fa142bdf6f13b7db7b4aa6f0 Mon Sep 17 00:00:00 2001
From: JBGruber <j.gruber.1@research.gla.ac.uk>
Date: Wed, 18 Dec 2019 20:43:57 +0100
Subject: [PATCH] Added default encoding and changed to readLines instead
 stringi

---
 R/rwhatsapp.R   | 13 ++++++++-----
 README.md       |  2 +-
 inst/WORDLIST   |  1 +
 man/rwa_read.Rd |  7 +++++--
 4 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/R/rwhatsapp.R b/R/rwhatsapp.R
index ab18207..572cc2e 100755
--- a/R/rwhatsapp.R
+++ b/R/rwhatsapp.R
@@ -16,7 +16,9 @@
 #'   \link[stringi]{stri_datetime_parse} for guidance.
 #' @param verbose A logical flag indicating whether information should be
 #'   printed to the screen.
-#' @param ... Further arguments passed to \link[stringi]{stri_read_lines}.
+#' @param encoding Input encoding. Should usually be "UTF-8" if files haven't
+#'   changed since export from WhatsApp.
+#' @param ... Further arguments passed to \link[base]{readLines}.
 #'
 #' @return A tibble with the information parsed from the history file.
 #' @export
@@ -31,6 +33,7 @@ rwa_read <- function(x,
                      tz = NULL,
                      format = NULL,
                      verbose = FALSE,
+                     encoding = "UTF-8",
                      ...) {
 
   if (verbose) {
@@ -40,7 +43,7 @@ rwa_read <- function(x,
     start_time <- NULL
   }
 
-  chat_raw <- rwa_read_lines(x, verbose, start_time, ...)
+  chat_raw <- rwa_read_lines(x, verbose, start_time, encoding, ...)
 
   chat_raw <- chat_raw[!chat_raw == ""]
   time <- stri_extract_first_regex(
@@ -126,7 +129,7 @@ rwa_read <- function(x,
 #' @inherit rwa_read
 #' @import stringi
 #' @noRd
-rwa_read_lines <- function(x, verbose, start_time = NULL, ...) {
+rwa_read_lines <- function(x, verbose, start_time = NULL, encoding, ...) {
   # get files
   zps <- grep(".zip$", x, ignore.case = TRUE)
   temp <- NULL
@@ -144,7 +147,7 @@ rwa_read_lines <- function(x, verbose, start_time = NULL, ...) {
 
   if (f_exist_s(x)) {
     if (length(x) == 1) {
-      chat_raw <- stri_read_lines(x, ...)
+      chat_raw <- readLines(x, encoding = encoding, ...)
       names(chat_raw) <- rep(x, length(chat_raw))
       if (verbose) {
         message(" one log file...")
@@ -152,7 +155,7 @@ rwa_read_lines <- function(x, verbose, start_time = NULL, ...) {
       }
     } else {
       chat_raw <- unlist(lapply(x, function(t) {
-        cr <- stri_read_lines(t)#, ...)
+        cr <- readLines(t, encoding = encoding, ...)
         names(cr) <- rep(t, length(cr))
         return(cr)
       }))
diff --git a/README.md b/README.md
index 2373de6..73da2c1 100755
--- a/README.md
+++ b/README.md
@@ -11,7 +11,7 @@ developed.](https://www.repostatus.org/badges/latest/active.svg)](https://www.re
 ![Downloads](https://cranlogs.r-pkg.org/badges/grand-total/rwhatsapp)
 [![Travis-CI Build
 Status](https://travis-ci.org/JBGruber/rwhatsapp.svg?branch=master)](https://travis-ci.org/JBGruber/rwhatsapp)
-[![Say
+
 
 ## Motivation
 
diff --git a/inst/WORDLIST b/inst/WORDLIST
index d86c505..c9be653 100644
--- a/inst/WORDLIST
+++ b/inst/WORDLIST
@@ -19,6 +19,7 @@ Macbook
 ne
 oman
 prozessor
+readLines
 stopwords
 stri
 tf
diff --git a/man/rwa_read.Rd b/man/rwa_read.Rd
index b0f5779..1b4a0d8 100644
--- a/man/rwa_read.Rd
+++ b/man/rwa_read.Rd
@@ -4,7 +4,7 @@
 \alias{rwa_read}
 \title{Read WhatsApp history into R}
 \usage{
-rwa_read(x, tz = NULL, format = NULL, verbose = FALSE, ...)
+rwa_read(x, tz = NULL, format = NULL, verbose = FALSE, encoding = "UTF-8", ...)
 }
 \arguments{
 \item{x}{Path to a txt or zip file of a WhatsApp history or the history
@@ -21,7 +21,10 @@ problems you can provide a custom format here. Refer to
 \item{verbose}{A logical flag indicating whether information should be
 printed to the screen.}
 
-\item{...}{Further arguments passed to \link[stringi]{stri_read_lines}.}
+\item{encoding}{Input encoding. Should usually be "UTF-8" if files haven't
+changed since export from WhatsApp.}
+
+\item{...}{Further arguments passed to \link[base]{readLines}.}
 }
 \value{
 A tibble with the information parsed from the history file.