From: Maxim Valyanskiy Date: Wed, 4 Aug 2010 12:43:58 +0000 (+0000) Subject: hwpf: append any character data before paragraphs to first paragraph X-Git-Tag: REL_3_7_BETA2~7 X-Git-Url: https://source.dussan.org/?a=commitdiff_plain;h=79c4de256f76c7b372333552975c8701c7f6a798;p=poi.git hwpf: append any character data before paragraphs to first paragraph (workaround for bug#48075) git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@982238 13f79535-47bb-0310-9956-ffa450edef68 --- diff --git a/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Paragraph.java b/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Paragraph.java index a1462ffe0a..087c6c5c16 100644 --- a/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Paragraph.java +++ b/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Paragraph.java @@ -101,6 +101,14 @@ public class Paragraph extends Range implements Cloneable { _istd = papx.getIstd(); } + protected Paragraph(PAPX papx, Range parent, int start) + { + super(Math.max(parent._start, start), Math.min(parent._end, papx.getEnd()), parent); + _props = papx.getParagraphProperties(_doc.getStyleSheet()); + _papx = papx.getSprmBuf(); + _istd = papx.getIstd(); + } + public short getStyleIndex() { return _istd; diff --git a/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Range.java b/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Range.java index 7c9b541d7c..df9cb0c46d 100644 --- a/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Range.java +++ b/src/scratchpad/src/org/apache/poi/hwpf/usermodel/Range.java @@ -830,7 +830,11 @@ public class Range { // TODO -instantiable superclass if (props.getIlfo() > 0) { pap = new ListEntry(papx, this, _doc.getListTables()); } else { - pap = new Paragraph(papx, this); + if (((index + _parStart)==0) && papx.getStart()>0) { + pap = new Paragraph(papx, this, 0); + } else { + pap = new Paragraph(papx, this); + } } return pap; diff --git a/src/scratchpad/testcases/org/apache/poi/hwpf/extractor/TestWordExtractor.java b/src/scratchpad/testcases/org/apache/poi/hwpf/extractor/TestWordExtractor.java index 22eaf0ade4..ea69824daa 100644 --- a/src/scratchpad/testcases/org/apache/poi/hwpf/extractor/TestWordExtractor.java +++ b/src/scratchpad/testcases/org/apache/poi/hwpf/extractor/TestWordExtractor.java @@ -298,4 +298,14 @@ public final class TestWordExtractor extends TestCase { assertTrue(text.contains("\u0425\u0425\u0425\u0425\u0425")); assertTrue(text.contains("\u0423\u0423\u0423\u0423\u0423")); } + + public void testFirstParagraphFix() throws Exception { + extractor = new WordExtractor( + POIDataSamples.getDocumentInstance().openResourceAsStream("MBD001D0B89.doc") + ); + + String text = extractor.getText(); + + assertTrue(text.startsWith("\u041f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0435")); + } } diff --git a/test-data/document/MBD001D0B89.doc b/test-data/document/MBD001D0B89.doc new file mode 100644 index 0000000000..386b5305be Binary files /dev/null and b/test-data/document/MBD001D0B89.doc differ