TIKA-2931 (apache#281)

* need to lowercase the output file to match the format passed to tesseract cmd line * need to lowercase the output file to match the format passed to tesseract cmd line * pull original code from apache/tika * Send informational output to System.err, not System.out so commandline json response is clean
conradbm · Sep 4, 2019 · 056304f · 056304f
1 parent e939504
commit 056304f
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 3 deletions.
diff --git a/tika-app/src/main/java/org/apache/tika/cli/TikaCLI.java b/tika-app/src/main/java/org/apache/tika/cli/TikaCLI.java
@@ -1051,7 +1051,7 @@ public void parseEmbedded(InputStream inputStream, ContentHandler contentHandler
                     throw new IOException("unable to create directory \"" + parent + "\"");
                 }
             }
-            System.out.println("Extracting '"+name+"' ("+contentType+") to " + outputFile);
+            System.err.println("Extracting '"+name+"' ("+contentType+") to " + outputFile);
 
             try (FileOutputStream os = new FileOutputStream(outputFile)) {
                 if (inputStream instanceof TikaInputStream) {

diff --git a/tika-app/src/test/java/org/apache/tika/cli/TikaCLITest.java b/tika-app/src/test/java/org/apache/tika/cli/TikaCLITest.java
@@ -49,17 +49,22 @@ public class TikaCLITest {
 
     /* Test members */
     private ByteArrayOutputStream outContent = null;
+    private ByteArrayOutputStream errContent = null;
     private PrintStream stdout = null;
+    private PrintStream stderr = null;
     private File testDataFile = new File("src/test/resources/test-data");
     private URI testDataURI = testDataFile.toURI();
     private String resourcePrefix;
 
     @Before
     public void setUp() throws Exception {
         outContent = new ByteArrayOutputStream();
+        errContent = new ByteArrayOutputStream();
         resourcePrefix = testDataURI.toString();
         stdout = System.out;
+        stderr = System.err;
         System.setOut(new PrintStream(outContent, true, UTF_8.name()));
+        System.setErr(new PrintStream(errContent, true, UTF_8.name()));
     }
 
     /**
@@ -258,11 +263,12 @@ public void testListSupportedTypes() throws Exception{
     }
 
     /**
-     * Tears down the test. Returns the System.out
+     * Tears down the test. Returns the System.out and System.err
      */
     @After
     public void tearDown() throws Exception {
         System.setOut(stdout);
+        System.setErr(stderr);
     }
 
     @Test
@@ -402,7 +408,7 @@ public void testZipWithSubdirs() throws Exception {
         new File("subdir/foo.txt").delete();
         new File("subdir").delete();
         TikaCLI.main(params);
-        String content = outContent.toString(UTF_8.name());
+        String content = errContent.toString(UTF_8.name());
         assertTrue(content.contains("Extracting 'subdir/foo.txt'"));
         // clean up. TODO: These should be in target.
         new File("target/subdir/foo.txt").delete();