From f3c25f02d587fd8585a3f0d468195829a13592b6 Mon Sep 17 00:00:00 2001
From: Robert Muir <rmuir@apache.org>
Date: Sat, 8 May 2010 00:42:04 +0000
Subject: [PATCH] SOLR-1865: ignore BOMs in SolrResourceLoader.getLines

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@942288 13f79535-47bb-0310-9956-ffa450edef68
---
 solr/CHANGES.txt                              |  4 +++
 .../apache/solr/core/SolrResourceLoader.java  |  3 ++
 .../apache/solr/core/ResourceLoaderTest.java  | 31 ++++++++++++++++++-
 .../test/test-files/solr/conf/stopwithbom.txt |  1 +
 4 files changed, 38 insertions(+), 1 deletion(-)
 create mode 100644 solr/src/test/test-files/solr/conf/stopwithbom.txt

diff --git a/solr/CHANGES.txt b/solr/CHANGES.txt
index 85b95218168..f93f162e148 100644
--- a/solr/CHANGES.txt
+++ b/solr/CHANGES.txt
@@ -347,6 +347,10 @@ Other Changes
   (Chris Male via rmuir)
 
 * SOLR-1851: luceneAutoCommit no longer has any effect - it has been remove (Mark Miller)
+
+* SOLR-1865: SolrResourceLoader.getLines ignores Byte Order Markers (BOMs) at the
+  beginning of input files, these are often created by editors such as Windows
+  Notepad.  (rmuir, hossman)
   
   
 Build
diff --git a/solr/src/java/org/apache/solr/core/SolrResourceLoader.java b/solr/src/java/org/apache/solr/core/SolrResourceLoader.java
index 114def4977e..40b2b557728 100644
--- a/solr/src/java/org/apache/solr/core/SolrResourceLoader.java
+++ b/solr/src/java/org/apache/solr/core/SolrResourceLoader.java
@@ -309,6 +309,9 @@ public class SolrResourceLoader implements ResourceLoader
 
       lines = new ArrayList<String>();
       for (String word=null; (word=input.readLine())!=null;) {
+        // skip initial bom marker
+        if (lines.isEmpty() && word.length() > 0 && word.charAt(0) == '\uFEFF')
+          word = word.substring(1);
         // skip comments
         if (word.startsWith("#")) continue;
         word=word.trim();
diff --git a/solr/src/test/org/apache/solr/core/ResourceLoaderTest.java b/solr/src/test/org/apache/solr/core/ResourceLoaderTest.java
index 5b386f0a585..8c77ca2143d 100644
--- a/solr/src/test/org/apache/solr/core/ResourceLoaderTest.java
+++ b/solr/src/test/org/apache/solr/core/ResourceLoaderTest.java
@@ -30,6 +30,9 @@ import org.apache.solr.util.plugin.ResourceLoaderAware;
 import org.apache.solr.util.plugin.SolrCoreAware;
 
 import java.io.File;
+import java.io.InputStream;
+import java.util.Arrays;
+import java.util.List;
 
 public class ResourceLoaderTest extends TestCase 
 {
@@ -88,4 +91,30 @@ public class ResourceLoaderTest extends TestCase
       catch( SolrException ex ) { } // OK
     }
   }
-}
\ No newline at end of file
+  
+  public void testBOMMarkers() throws Exception {
+    final String fileWithBom = "stopwithbom.txt";
+    SolrResourceLoader loader = new SolrResourceLoader(null);
+
+    // preliminary sanity check
+    InputStream bomStream = loader.openResource(fileWithBom);
+    try {
+      final byte[] bomExpected = new byte[] { -17, -69, -65 };
+      final byte[] firstBytes = new byte[3];
+      
+      assertEquals("Should have been able to read 3 bytes from bomStream",
+                   3, bomStream.read(firstBytes));
+
+      assertTrue("This test only works if " + fileWithBom + 
+                 " contains a BOM -- it appears someone removed it.", 
+                 Arrays.equals(bomExpected, firstBytes));
+    } finally {
+      try { bomStream.close(); } catch (Exception e) { /* IGNORE */ }
+    }
+
+    // now make sure getLines skips the BOM...
+    List<String> lines = loader.getLines(fileWithBom);
+    assertEquals(1, lines.size());
+    assertEquals("BOMsAreEvil", lines.get(0));
+  }
+}
diff --git a/solr/src/test/test-files/solr/conf/stopwithbom.txt b/solr/src/test/test-files/solr/conf/stopwithbom.txt
new file mode 100644
index 00000000000..eb5f6e1c0f8
--- /dev/null
+++ b/solr/src/test/test-files/solr/conf/stopwithbom.txt
@@ -0,0 +1 @@
+BOMsAreEvil