From 51652bb0c99e7b603340cbc883da4fa049cc5d57 Mon Sep 17 00:00:00 2001
From: boB Rudis <bob@rudis.net>
Date: Sat, 10 Sep 2016 15:06:28 -0400
Subject: [PATCH] pre-CRAN flight check

---
 R/tidy.r         | 11 +++++++----
 README.Rmd       |  2 +-
 README.md        |  4 ++--
 man/tidy_html.Rd |  5 +++--
 4 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/R/tidy.r b/R/tidy.r
index 6944e4b..5d8fe7a 100644
--- a/R/tidy.r
+++ b/R/tidy.r
@@ -69,10 +69,11 @@
 #' res <- GET("http://rud.is/test/untidy.html")
 #'
 #' # look at the original, un-tidy source
-#' cat(content(res, as="text"))
+#' cat(content(res, as="text", encoding="UTF-8"))
 #'
 #' # see the tidied version
-#' cat(tidy_html(content(res, as="text"), list(TidyDocType="html5", TidyWrapLen=200)))
+#' cat(tidy_html(content(res, as="text", encoding="UTF-8"),
+#'               list(TidyDocType="html5", TidyWrapLen=200)))
 #'
 #' # but, you could also just do:
 #' cat(tidy_html(url("http://rud.is/test/untidy.html")))
@@ -128,8 +129,10 @@ tidy_html.HTMLInternalDocument <- function(content, options=list(TidyXhtmlOut=TR
 #' @rdname tidy_html
 tidy_html.connection <- function(content, options=list(TidyXhtmlOut=TRUE)) {
 
-  content <- paste0(readLines(content), collapse="")
+  html <- paste0(readLines(content, warn=FALSE), collapse="")
+  close(content)
+
   .Call('htmltidy_tidy_html_int', PACKAGE='htmltidy',
-        source=content, options=options)
+        source=html, options=options)
 
 }
diff --git a/README.Rmd b/README.Rmd
index d78c011..5cec1e2 100644
--- a/README.Rmd
+++ b/README.Rmd
@@ -1,7 +1,7 @@
 ---
 output: rmarkdown::github_document
 ---
-[![Travis-CI Build Status](https://travis-ci.org/hrbrmstr/htmltidy.svg?branch=master)](https://travis-ci.org/hrbrmstr/htmltidy) [![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/htmltidy)](https://cran.r-project.org/package=htmltidy)
+[![Travis-CI Build Status](https://travis-ci.org/hrbrmstr/htmltidy.svg?branch=master)](https://travis-ci.org/hrbrmstr/htmltidy) <!-- [![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/htmltidy)](https://cran.r-project.org/package=htmltidy) -->
 
 <!-- README.md is generated from README.Rmd. Please edit that file -->
 
diff --git a/README.md b/README.md
index baa3610..c5d5d21 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,5 @@
 
-[![Travis-CI Build Status](https://travis-ci.org/hrbrmstr/htmltidy.svg?branch=master)](https://travis-ci.org/hrbrmstr/htmltidy) [![CRAN\_Status\_Badge](http://www.r-pkg.org/badges/version/htmltidy)](https://cran.r-project.org/package=htmltidy)
+[![Travis-CI Build Status](https://travis-ci.org/hrbrmstr/htmltidy.svg?branch=master)](https://travis-ci.org/hrbrmstr/htmltidy) <!-- [![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/htmltidy)](https://cran.r-project.org/package=htmltidy) -->
 
 <!-- README.md is generated from README.Rmd. Please edit that file -->
 `htmltidy` — Clean up gnarly HTML/XHTML
@@ -197,7 +197,7 @@ sum(map_int(book, nchar))
 ## [1] 207501
 system.time(tidy_book <- tidy_html(book))
 ##    user  system elapsed 
-##   0.022   0.001   0.022
+##   0.021   0.001   0.023
 ```
 
 (It's usually between 20 & 25 milliseconds to process those 202 kilobytes of HTML.) Not too shabby.
diff --git a/man/tidy_html.Rd b/man/tidy_html.Rd
index dca275e..13783c2 100644
--- a/man/tidy_html.Rd
+++ b/man/tidy_html.Rd
@@ -96,10 +96,11 @@ library(httr)
 res <- GET("http://rud.is/test/untidy.html")
 
 # look at the original, un-tidy source
-cat(content(res, as="text"))
+cat(content(res, as="text", encoding="UTF-8"))
 
 # see the tidied version
-cat(tidy_html(content(res, as="text"), list(TidyDocType="html5", TidyWrapLen=200)))
+cat(tidy_html(content(res, as="text", encoding="UTF-8"),
+              list(TidyDocType="html5", TidyWrapLen=200)))
 
 # but, you could also just do:
 cat(tidy_html(url("http://rud.is/test/untidy.html")))