Merge pull request #40 from andrechalom/multilingual

Multilingual

README.md

@@ -34,10 +34,16 @@ runApp("<path to Rsampling-shiny>")
 ```
 Remember that you need to install the shiny/shinyBS packages while you have internet access!
 
+## Languages
+The app is currently available in English (default) and Portuguese.
+To use the version in Portuguese, run the following command 
+before running the app (runGitHub(...) or runApp(...))
+```R
+language <- "pt"
+```
 ## Current features
 * Access the Rsampling sample datafiles, or upload your own data in csv format
 * Choose between a range of statistics, or write your own R function
 * Select your randomization scheme, with or without replacement
 * Optional stratified randomization
 * Reactive graph showing the distribuion of the statistic of interest and estimated p-value
-* Interactive tutorials on resampling statistics

global.R

@@ -0,0 +1,30 @@
+##################################################
+###       MULTILINGUAL VARIABLES AND FUNCTION  ###
+#################################################
+# The default language is "en". To use the translated version, set the language such as
+# > language <- "pt" 
+# before calling the app
+language <- ifelse(exists("language"), language, "en")
+dictionary <- read.csv('lang.csv', sep = ';', stringsAsFactors = FALSE, fileEncoding="UTF-8", header=TRUE)
+# If the language is not recognized, does not attempt to run the app. Stop immediately:
+if (! language %in% colnames(dictionary)[-1])
+  stop(paste0("Language is not recognized: ", language, "\n"))
+## Main translation function. Uses "language" and "dictionary" as globals, defined above
+# in this file. 
+tr <- function(text) {
+    for (i in 1:length(text)) {
+      id <- which(dictionary$value %in% text[i])
+      # Trying again: 
+      if (length(id) == 0) 
+        id <- which(dictionary$en %in% text[i])
+      # If still no matches, prints a warning:
+      if (length(id) == 0) 
+        cat(paste('No matches for text:', text[i], '\n'))
+      if (length(id) == 1) {
+        text[i] <- ifelse(language == "pt", dictionary$pt[id], dictionary$en[id])
+      } else if (length(id) > 1) {
+        cat(paste('More than one match for text:', text[i], '\n'))
+      }
+    }
+    return(text)
+}

help.html

@@ -18,7 +18,7 @@
 </ul>
 <p>The idea behind this package is to help students to understand the logic behind these tests,
 	 with lots of flexibility on what is the statistic that will be calculated.</p>
-<p>Rsampling and and this web interface were written by <a href="mailto:[email protected]">Andre Chalom</a>,  
+<p>Rsampling and this web interface were written by <a href="mailto:[email protected]">Andre Chalom</a>,  
    <a href="mailto:[email protected]">Alexandre Adalardo</a>, <a href="mailto:[email protected]">Ayana Martins</a>
    and <a href="mailto:[email protected]">Paulo Prado</a>.</p>
 <p>Please visit our <a href="https://github.com/andrechalom/Rsampling-shiny">project webpage</a>

help_pt.html

@@ -0,0 +1,31 @@
+<div style="float: right; border:1px dashed #ccc; width:22%; text-align: center;">
+<img src="chicken.png" align="none">
+<p style="font-size: 75%;">A Galinha Esférica foi criada por 
+<a href="[email protected]">Luisa Novara</a> e pode ser utilizada sob a licença
+<a href="https://creativecommons.org/licenses/by/3.0/">CC3-by</a>.</p>
+</div>
+<div style="float: left; width: 73%">
+<p>Rsampling é uma pacote gratuito e de código aberto desenvolvido
+	com o objetivo de ajudar no aprendizado de métodos estatísticos
+	baseados em permutação.
+	O pacote foi baseado no programa 
+	<a href="http://www.resample.com">Resampling Stats</a>.</p>
+<p>Podemos pensar na lógica por trás de um teste de significância da seguinte forma:
+<ul><li>Escolha a sua estatística de interesse</li>
+		 <li>Determine qual é a sua hipótese nula</li>
+		 <li>Obtenha a distribuição da estatísica de interesse sob a hipótese nula</li>
+		 <li>Rejeite a hipótese nula se a probabilidade da estatística de interesse observada for
+		 menor do que um certo limiar</li>
+</ul>
+<p>A ideia por trás deste pacote é auxiliar na compreensão desta lógica, com grande flexibilidade de estatísticas
+que podem ser calculadas.</p>
+<p>Rsampling e sua interface gráfica foram escritos por <a href="mailto:[email protected]">Andre Chalom</a>,  
+   <a href="mailto:[email protected]">Alexandre Adalardo</a>, <a href="mailto:[email protected]">Ayana Martins</a>
+   e <a href="mailto:[email protected]">Paulo Prado</a>.</p>
+<p>Visite nossa <a href="https://github.com/andrechalom/Rsampling-shiny">página de projeto</a>
+   e a página da <a href="https://github.com/lageIBUSP/Rsampling">biblioteca Rsampling </a> no
+	 <a href="https://github.com">github</a>,
+	 para aprender mais sobre o projeto e enviar sugestões ou reportar bugs.</p>
+<p>Rsampling e Rsampling-shiny têm os poderes da Galinha Esférica.</p>
+</div>
+<div style="clear: both;"></div>