Implemented wrapping content in <html> structure so set encoding, see…

… Issue 1 and 2 (#1).
dankito · Aug 12, 2018 · 8d4b042 · 8d4b042
1 parent 9d5a8c3
commit 8d4b042
Showing 1 changed file with 41 additions and 1 deletion.
diff --git a/src/main/kotlin/net/dankito/readability4j/Article.kt b/src/main/kotlin/net/dankito/readability4j/Article.kt
@@ -20,11 +20,29 @@ open class Article(
         var articleContent: Element? = null
 
         /**
-         * HTML string of processed article content
+         * HTML string of processed article content in a &lt;div> element.
+         *
+         * Therefore no encoding is applied, see [contentWithUtf8Encoding] or issue
+         * [https://github.com/dankito/Readability4J/issues/1].
          */
         val content: String?
                 get() = articleContent?.html() // TODO: but this removes paging information (pages in top node <div id="readability-content">)
 
+        /**
+         * [content] returns a &lt;div> element.
+         *
+         * As the only way in HTML to set an encoding is via &lt;head>&lt;meta charset=""> tag, therefore no explicit
+         * encoding is applied to it.
+         * As a result non-ASCII characters may get displayed incorrectly.
+         *
+         * So this method wraps [content] in &lt;html>&lt;head>&lt;meta charset="utf-8"/>&lt;/head>&lt;body>&lt;!--
+         * content-->&lt;/body>&lt;/html> so that UTF-8 encoding gets applied.
+         *
+         * See [https://github.com/dankito/Readability4J/issues/1] for more info.
+         */
+        val contentWithUtf8Encoding: String?
+                get() = getContentWithEncoding("utf-8")
+
         val textContent: String?
                 get() = articleContent?.text()
 
@@ -49,4 +67,26 @@ open class Article(
          */
         var dir: String? = null
 
+
+        /**
+         * [content] returns a &lt;div> element.
+         *
+         * As the only way in HTML to set an encoding is via &lt;head>&lt;meta charset=""> tag, therefore no explicit
+         * encoding is applied to it.
+         * As a result non-ASCII characters may get displayed incorrectly.
+         *
+         * So this method wraps [content] in &lt;html>&lt;head>&lt;meta charset="[encoding]"/>&lt;/head>&lt;body>&lt;!--
+         * content-->&lt;/body>&lt;/html> so that encoding gets applied.
+         *
+         * See [https://github.com/dankito/Readability4J/issues/1] for more info.
+         */
+        fun getContentWithEncoding(encoding: String): String? {
+                content?.let { content ->
+                        return "<html>\n  <head>\n    <meta charset=\"$encoding\"/>\n  </head>\n  <body>\n    " +
+                                "$content\n  </body>\n<html>"
+                }
+
+                return null
+        }
+
 }