Example to go with refactor for https://bz.apache.org/bugzilla/show_bug.cgi?id=58219

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1694536 13f79535-47bb-0310-9956-ffa450edef68

2 files changed, 62 insertions, 1 deletions

diff --git a/src/examples/src/org/apache/poi/xssf/streaming/examples/HybridStreaming.java b/src/examples/src/org/apache/poi/xssf/streaming/examples/HybridStreaming.java
new file mode 100644
index 0000000000..99af0356a1
--- /dev/null
+++ b/src/examples/src/org/apache/poi/xssf/streaming/examples/HybridStreaming.java
@@ -0,0 +1,57 @@
+package org.apache.poi.xssf.streaming.examples;
+
+import java.io.FileInputStream;
+import java.io.IOException;
+import java.io.InputStream;
+
+import org.apache.poi.xssf.eventusermodel.ReadOnlySharedStringsTable;
+import org.apache.poi.xssf.eventusermodel.XSSFSheetXMLHandler;
+import org.apache.poi.xssf.eventusermodel.XSSFSheetXMLHandler.SheetContentsHandler;
+import org.apache.poi.xssf.usermodel.XSSFComment;
+import org.apache.poi.xssf.usermodel.XSSFSheet;
+import org.apache.poi.xssf.usermodel.XSSFWorkbook;
+import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTSheet;
+import org.xml.sax.SAXException;
+
+/**
+ * This demonstrates how a hybrid approach to workbook read can be taken, using 
+ */
+public class HybridStreaming {
+    
+    private static final String SHEET_TO_STREAM = "large sheet";
+
+    public static void main(String[] args) throws IOException, SAXException {
+        InputStream sourceBytes = new FileInputStream("/path/too/workbook.xlsx");
+        XSSFWorkbook workbook = new XSSFWorkbook(sourceBytes) {
+            /** Avoid DOM parse of large sheet */
+            public void parseSheet(java.util.Map<String,XSSFSheet> shIdMap, CTSheet ctSheet) {
+                if (SHEET_TO_STREAM.equals(ctSheet.getName())) {
+                    return;
+                }
+            };
+        };
+        
+        // Having avoided a DOM-based parse of the sheet, we can stream it instead.
+        ReadOnlySharedStringsTable strings = new ReadOnlySharedStringsTable(workbook.getPackage());
+        new XSSFSheetXMLHandler(workbook.getStylesSource(), strings, createSheetContentsHandler(), false);
+        workbook.close();
+    }
+
+    private static SheetContentsHandler createSheetContentsHandler() {
+        return new SheetContentsHandler() {
+            
+            public void startRow(int rowNum) {
+            }
+            
+            public void headerFooter(String text, boolean isHeader, String tagName) {
+            }
+            
+            public void endRow(int rowNum) {
+            }
+            
+            public void cell(String cellReference, String formattedValue, XSSFComment comment) {
+            }
+        };
+    }
+
+}
diff --git a/src/ooxml/java/org/apache/poi/xssf/usermodel/XSSFWorkbook.java b/src/ooxml/java/org/apache/poi/xssf/usermodel/XSSFWorkbook.java
index cb9ba9a5fe..4b4a30e0b6 100644
--- a/src/ooxml/java/org/apache/poi/xssf/usermodel/XSSFWorkbook.java
+++ b/src/ooxml/java/org/apache/poi/xssf/usermodel/XSSFWorkbook.java
@@ -401,7 +401,11 @@ public class XSSFWorkbook extends POIXMLDocument implements Workbook, Iterable<X
         }
     }
 
-    protected void parseSheet(Map<String, XSSFSheet> shIdMap, CTSheet ctSheet) {
+    /**
+     * Not normally to be called externally, but possibly to be overridden to avoid
+     * the DOM based parse of large sheets (see examples).
+     */
+    public void parseSheet(Map<String, XSSFSheet> shIdMap, CTSheet ctSheet) {
         XSSFSheet sh = shIdMap.get(ctSheet.getId());
         if(sh == null) {
             logger.log(POILogger.WARN, "Sheet with name " + ctSheet.getName() + " and r:id " + ctSheet.getId()+ " was defined, but didn't exist in package, skipping");