Fix to work if ImageMagick isn't present. Fix forbidden APIs.
Project: http://git-wip-us.apache.org/repos/asf/tika/repo Commit: http://git-wip-us.apache.org/repos/asf/tika/commit/6f16480f Tree: http://git-wip-us.apache.org/repos/asf/tika/tree/6f16480f Diff: http://git-wip-us.apache.org/repos/asf/tika/diff/6f16480f Branch: refs/heads/master Commit: 6f16480f7b1c6b9677626f215f52505d4b61a31a Parents: 1c6cff8 Author: Chris Mattmann <[email protected]> Authored: Wed Jul 6 23:39:05 2016 -0700 Committer: Chris Mattmann <[email protected]> Committed: Wed Jul 6 23:39:05 2016 -0700 ---------------------------------------------------------------------- .../main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java | 4 ++-- .../java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/tika/blob/6f16480f/tika-parsers/src/main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java ---------------------------------------------------------------------- diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java b/tika-parsers/src/main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java index f28016a..7d842c4 100644 --- a/tika-parsers/src/main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java +++ b/tika-parsers/src/main/java/org/apache/tika/parser/ocr/TesseractOCRParser.java @@ -161,7 +161,7 @@ public class TesseractOCRParser extends AbstractParser { try { Process proc = Runtime.getRuntime().exec("python -h"); - BufferedReader stdInput = new BufferedReader(new InputStreamReader(proc.getInputStream())); + BufferedReader stdInput = new BufferedReader(new InputStreamReader(proc.getInputStream(), "UTF-8")); if(stdInput.read() != -1) { hasPython = true; } @@ -284,7 +284,7 @@ public class TesseractOCRParser extends AbstractParser { if(hasPython()) { try { executor.execute(cmdLine); - angle = outputStream.toString().trim(); + angle = outputStream.toString("UTF-8").trim(); } catch(Exception e) { e.printStackTrace(); } http://git-wip-us.apache.org/repos/asf/tika/blob/6f16480f/tika-parsers/src/test/java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java ---------------------------------------------------------------------- diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java index 5c511a0..cc0288f 100644 --- a/tika-parsers/src/test/java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java +++ b/tika-parsers/src/test/java/org/apache/tika/parser/ocr/TesseractOCRParserTest.java @@ -180,7 +180,7 @@ public class TesseractOCRParserTest extends TikaTest { "/test-properties/TesseractOCR.properties"); TesseractOCRConfig config = new TesseractOCRConfig(stream); String[] CheckCmd = {config.getImageMagickPath() + TesseractOCRParser.getImageMagickProg()}; - assertEquals(true,ExternalParser.check(CheckCmd)); + assumeTrue(ExternalParser.check(CheckCmd)); } @Test
