Fix reading html from InputStream

jtablesaw · Mar 24, 2019 · d410a90 · d410a90
1 parent 5a156c2
commit d410a90
Showing 1 changed file with 11 additions and 3 deletions.
diff --git a/html/src/main/java/tech/tablesaw/io/html/HtmlReader.java b/html/src/main/java/tech/tablesaw/io/html/HtmlReader.java
@@ -1,11 +1,12 @@
 package tech.tablesaw.io.html;
 
 import java.io.IOException;
-import java.io.Reader;
+import java.io.InputStream;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.stream.Stream;
 
+import org.jsoup.Jsoup;
 import org.jsoup.nodes.Document;
 import org.jsoup.nodes.Element;
 import org.jsoup.parser.Parser;
@@ -33,8 +34,15 @@ public static void register(ReaderRegistry registry) {
 
     @Override
     public Table read(HtmlReadOptions options) throws IOException {
-        Reader reader = options.source().createReader(null);
-        Document doc = Parser.htmlParser().parseInput(reader, "");
+        Document doc;
+        InputStream inputStream = options.source().inputStream();
+        if (inputStream != null) {
+            // Reader must support mark, so can't use InputStreamReader
+            // Parse the InputStream directly
+            doc = Jsoup.parse(inputStream, null, "");
+        } else {
+            doc = Parser.htmlParser().parseInput(options.source().createReader(null), "");
+        }
         Elements tables = doc.select("table");
         if (tables.size() != 1) {
             throw new IllegalStateException(