Catch spurious available() throws

jhy · jhy · commit 49e58c202883 · 2026-03-06T19:03:57.000+11:00
Fixes #2474 Was causing CI flakes on the JDK 8 runner; quite possibly in other environments too.
diff --git a/CHANGES.md b/CHANGES.md
@@ -5,6 +5,7 @@
 ### Bug Fixes
 * Android (R8/ProGuard): added a rule to ignore the optional `re2j` dependency when not present. [#2459](https://github.com/jhy/jsoup/issues/2459)
 * In `NodeTraversor`, removing or replacing the current node during `head()` no longer re-visits the replacement node, preventing loops. Traversal now continues correctly from nodes that occupy the original position after mutation, and will not advance past the original root subtree. Also, clarified in the documentation which inserted nodes are visited during the current traversal. [#2472](https://github.com/jhy/jsoup/issues/2472)
+* Parsing during charset sniffing no longer fails if an advisory `available()` call throws `IOException`, as seen on JDK 8 `HttpURLConnection`. [#2474](https://github.com/jhy/jsoup/issues/2474)
 
 ## 1.22.1 (2026-Jan-01)
 
diff --git a/src/main/java/org/jsoup/internal/SimpleBufferedInput.java b/src/main/java/org/jsoup/internal/SimpleBufferedInput.java
@@ -80,7 +80,11 @@ private void fill() throws IOException {
             bufLength = read + bufPos;
             capRemaining -= read;
             while (byteBuf.length - bufLength > 0 && capRemaining > 0) { // read in more if we have space, without blocking
-                if (in.available() < 1) break;
+                try {
+                    if (in.available() < 1) break;
+                } catch (IOException e) {
+                    break; // available() is advisory; keep the bytes we've already buffered
+                }
                 toRead = Math.min(byteBuf.length - bufLength, capRemaining);
                 if (toRead <= 0) break;
                 read = in.read(byteBuf, bufLength, toRead);
@@ -116,8 +120,7 @@ public int available() throws IOException {
         if (buffered > 0) {
             return buffered; // doesn't include those in.available(), but mostly used as a block test
         }
-        int avail = inReadFully ? 0 : in.available();
-        return avail;
+        return inReadFully ? 0 : in.available();
     }
 
     void capRemaining(int newRemaining) {
diff --git a/src/main/java/org/jsoup/internal/SimpleStreamReader.java b/src/main/java/org/jsoup/internal/SimpleStreamReader.java
@@ -43,7 +43,7 @@ public int read(char[] charArray, int off, int len) throws IOException {
         while (true) {
             CoderResult result = decoder.decode(byteBuf, charBuf, readFully);
             if (result.isUnderflow()) {
-                if (readFully || !charBuf.hasRemaining() || (charBuf.position() > 0) && !(in.available() > 0))
+                if (readFully || !charBuf.hasRemaining() || (charBuf.position() > 0) && !hasAvailableBytes())
                     break;
                 int read = bufferUp();
                 if (read < 0) {
@@ -64,6 +64,14 @@ public int read(char[] charArray, int off, int len) throws IOException {
         return charBuf.position();
     }
 
+    private boolean hasAvailableBytes() {
+        try {
+            return in.available() > 0;
+        } catch (IOException e) {
+            return false; // available() is advisory; a real read can still consume buffered bytes or reach EOF
+        }
+    }
+
     private int bufferUp() throws IOException {
         assert byteBuf != null; // already validated ^
         byteBuf.compact();
diff --git a/src/test/java/org/jsoup/helper/DataUtilTest.java b/src/test/java/org/jsoup/helper/DataUtilTest.java
@@ -418,4 +418,25 @@ public int available() {
         }
     }
 
+    @Test
+    void charsetSniffingIgnoresAdvisoryAvailableIOException() throws IOException {
+        // https://github.com/jhy/jsoup/issues/2474
+        // JDK 8's HttpURLConnection stream may throw from available() once the peer has closed the socket;
+        // that advisory failure does not mean we can't still consume bytes already buffered or read to clean EOF.
+        String html = "<!doctype html><html><head><title>One</title></head><body>Two</body></html>";
+        byte[] bytes = html.getBytes(StandardCharsets.UTF_8);
+        InputStream stream = new FilterInputStream(new ByteArrayInputStream(bytes)) {
+            @Override
+            public int available() throws IOException {
+                throw new IOException("Stream closed.");
+            }
+        };
+        ControllableInputStream in = ControllableInputStream.wrap(stream, 0);
+
+        Document doc = DataUtil.parseInputStream(in, null, "http://example.com/", Parser.htmlParser());
+
+        assertEquals("One", doc.title());
+        assertEquals("Two", doc.body().text());
+    }
+
 }
