REPORT-50649 导出-生僻字导出问题 Excel导出

fine-poi/src/main/java/com/fr/third/v2/org/apache/poi/xssf/streaming/SheetDataWriter.java

@@ -377,14 +377,22 @@ public class SheetDataWriter {
                 default:
                     // YK: XmlBeans silently replaces all ISO control characters ( < 32) with question marks.
                     // the same rule applies to unicode surrogates and "not a character" symbols.
-                    if( c < ' ' || Character.isLowSurrogate(c) || Character.isHighSurrogate(c) ||
-                            ('\uFFFE' <= c && c <= '\uFFFF')) {
+                    // Excel SXSSF deals with special chars.
+                    // https://svn.apache.org/viewvc/poi/trunk/src/ooxml/java/org/apache/poi/xssf/streaming/SheetDataWriter.java?r1=1791720&r2=1800705&pathrev=1800705&diff_format=f
+                    if (replaceWithQuestionMark(c)) {
                         if (counter > last) {
                             sb.append(chars, last, counter - last);
                         }
                         sb.append('?');
                         last = counter + 1;
                     }
+                    else if (Character.isHighSurrogate(c) || Character.isLowSurrogate(c)) {
+                        if (counter > last) {
+                            sb.append(chars, last, counter - last);
+                        }
+                        sb.append(c);
+                        last = counter + 1;
+                    }
                     else if (c > 127) {
                         if (counter > last) {
                             sb.append(chars, last, counter - last);
@@ -412,4 +420,8 @@ public class SheetDataWriter {
         _out.close();
         return _fd.delete();
     }
+
+    private boolean replaceWithQuestionMark(char c) {
+        return c < ' ' || '\uFFFE' <= c;
+    }
 }

fine-poi/src/main/java/com/fr/third/v2/org/apache/xmlbeans/impl/store/Saver.java

@@ -273,6 +273,23 @@ abstract class Saver
         end.toEnd();
     }
 
+    /**
+     * Test if a character is valid in xml character content. See
+     * http://www.w3.org/TR/REC-xml#NT-Char
+     * Excel XSSF deals with special chars.
+     * https://svn.apache.org/viewvc/xmlbeans/trunk/src/store/org/apache/xmlbeans/impl/store/Saver.java?r1=1025773&r2=1832298&pathrev=1832298&diff_format=h
+     */
+    static boolean isBadChar ( char ch )
+    {
+        return ! (
+                Character.isHighSurrogate(ch) ||
+                        Character.isLowSurrogate(ch) ||
+                        (ch >= 0x20 && ch <= 0xD7FF) ||
+                        (ch >= 0xE000 && ch <= 0xFFFD) ||
+                        (ch == 0x9) || (ch == 0xA) || (ch == 0xD)
+        );
+    }
+
     protected boolean saveNamespacesFirst ( )
     {
         return _saveNamespacesFirst;
@@ -1548,21 +1565,6 @@ abstract class Saver
             }
         }
 
-        /**
-         * Test if a character is valid in xml character content. See
-         * http://www.w3.org/TR/REC-xml#NT-Char
-         */
-
-        private boolean isBadChar ( char ch )
-        {
-            return ! (
-                (ch >= 0x20 && ch <= 0xD7FF ) ||
-                (ch >= 0xE000 && ch <= 0xFFFD) ||
-                (ch >= 0x10000 && ch <= 0x10FFFF) ||
-                (ch == 0x9) || (ch == 0xA) || (ch == 0xD)
-                );
-        }
-
         /**
          * Test if a character is to be replaced with an escaped value
          */
@@ -2187,20 +2189,6 @@ abstract class Saver
             }
         }
 
-        /**
-         * Test if a character is valid in xml character content. See
-         * http://www.w3.org/TR/REC-xml#NT-Char
-         */
-        private boolean isBadChar ( char ch )
-        {
-            return ! (
-                (ch >= 0x20 && ch <= 0xD7FF ) ||
-                (ch >= 0xE000 && ch <= 0xFFFD) ||
-                (ch >= 0x10000 && ch <= 0x10FFFF) ||
-                (ch == 0x9) || (ch == 0xA) || (ch == 0xD)
-                );
-        }
-
         private void emitLiteral ( String literal )
         {
             // TODO: systemId production http://www.w3.org/TR/REC-xml/#NT-SystemLiteral