From 742a05261905975052558aec4794e9a533643c50 Mon Sep 17 00:00:00 2001
From: David Roberts <dave.roberts@elastic.co>
Date: Fri, 10 Nov 2017 15:14:28 +0000
Subject: [PATCH] [ML] Account for the possibility of C++ log messages being
 UTF-16 (elastic/x-pack-elasticsearch#2952)

On Windows, log4cxx always writes to stderr in UTF-16, and we get the
logs from C++ to Java by redirecting stderr to our named pipe.  Hence
the log handler in Java needs to tolerate the log stream it's reading
being either UTF-16 (for Windows) or UTF-8 (for other platforms).

Fixes elastic/machine-learning-cpp#385

Original commit: elastic/x-pack-elasticsearch@89237d71255558c7b4452d6c7b8ae24bc5d1aa32
---
 .../xpack/ml/job/process/logging/CppLogMessageHandler.java  | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/plugin/src/main/java/org/elasticsearch/xpack/ml/job/process/logging/CppLogMessageHandler.java b/plugin/src/main/java/org/elasticsearch/xpack/ml/job/process/logging/CppLogMessageHandler.java
index d0948dee88f..c57096e0e60 100644
--- a/plugin/src/main/java/org/elasticsearch/xpack/ml/job/process/logging/CppLogMessageHandler.java
+++ b/plugin/src/main/java/org/elasticsearch/xpack/ml/job/process/logging/CppLogMessageHandler.java
@@ -205,6 +205,12 @@ public class CppLogMessageHandler implements Closeable {
                 parseMessage(xContent, bytesRef.slice(from, nextMarker - from));
             }
             from = nextMarker + 1;
+            if (from < bytesRef.length() && bytesRef.get(from) == (byte) 0) {
+                // This is to work around the problem of log4cxx on Windows
+                // outputting UTF-16 instead of UTF-8.  For full details see
+                // https://github.com/elastic/machine-learning-cpp/issues/385
+                ++from;
+            }
         }
         if (from >= bytesRef.length()) {
             return null;