TIKA-2807 -- extract sdt content from within textbox in docx

CHANGES.txt

@@ -7,6 +7,7 @@ Release 2.0.0 - ???
 
 
 Release 1.21 - ????
+   * Extract text from SDT element within textboxes in .docx files (TIKA-2807).
 
    * Try to handle truncated OOXML files more robustly (TIKA-2765).
 

tika-parsers/src/main/java/org/apache/tika/parser/microsoft/ooxml/XWPFWordExtractorDecorator.java

@@ -328,9 +328,11 @@ private void extractParagraph(XWPFParagraph paragraph, XWPFListManager listManag
             xhtml.characters(footnameText + "\n");
         }
 
-        // Also extract any paragraphs embedded in text boxes:
+        // Also extract any paragraphs embedded in text boxes
+        //Note "w:txbxContent//"...must look for all descendant paragraphs
+        //not just the immediate children of txbxContent -- TIKA-2807
         if (config.getIncludeShapeBasedContent()) {
-            for (XmlObject embeddedParagraph : paragraph.getCTP().selectPath("declare namespace w='http://schemas.openxmlformats.org/wordprocessingml/2006/main' declare namespace wps='http://schemas.microsoft.com/office/word/2010/wordprocessingShape' .//*/wps:txbx/w:txbxContent/w:p")) {
+            for (XmlObject embeddedParagraph : paragraph.getCTP().selectPath("declare namespace w='http://schemas.openxmlformats.org/wordprocessingml/2006/main' declare namespace wps='http://schemas.microsoft.com/office/word/2010/wordprocessingShape' .//*/wps:txbx/w:txbxContent//w:p")) {
                 extractParagraph(new XWPFParagraph(CTP.Factory.parse(embeddedParagraph.xmlText()), paragraph.getBody()), listManager, xhtml);
             }
         }

tika-parsers/src/test/java/org/apache/tika/parser/microsoft/ooxml/OOXMLParserTest.java

@@ -1058,6 +1058,15 @@ public void testTextInsideTextBox() throws Exception {
         assertContains("This text is inside of a text box in the footer of the document.", xml);
     }
 
+    //TIKA-2807
+    @Test
+    public void testSDTInTextBox() throws Exception {
+        String xml = getXML("testWORD_sdtInTextBox.docx").xml;
+        System.out.println(xml);
+        assertContains("rich-text-content-control_inside-text-box", xml);
+        assertContainsCount("inside-text", xml, 1);
+    }
+
     //TIKA-2346
     @Test
     public void testTurningOffTextBoxExtraction() throws Exception {

tika-parsers/src/test/java/org/apache/tika/parser/microsoft/ooxml/SXWPFExtractorTest.java

@@ -860,5 +860,12 @@ public void testTextDecorationNested() throws Exception {
         assertNotContained("unde ", txt);
     }
 
+    //TIKA-2807
+    @Test
+    public void testSDTInTextBox() throws Exception {
+        String xml = getXML("testWORD_sdtInTextBox.docx", parseContext).xml;
+        assertContains("rich-text-content-control_inside-text-box", xml);
+        assertContainsCount("inside-text", xml, 1);
+    }
 
 }