Excel Reading (XLSX) using SAX Default Handler (#10877)

- Implement Excel reading as a SAXMLParser.

(cherry picked from commit 63ed629)

distribution/lib/Standard/Table/0.0.0-dev/src/Excel/Excel_Workbook.enso

@@ -49,7 +49,7 @@ type Excel_Workbook
        - file: The file to load.
        - xls_format: Whether to use the old XLS format (default is XLSX).
     new : File | Temporary_File -> Boolean  -> Excel_Workbook
-    new file:(File | Temporary_File) xls_format=False =
+    new file:(File | Temporary_File) xls_format:Boolean=False =
         file_for_errors = if file.is_a Temporary_File then Nothing else file
 
         continuation raw_file =
@@ -73,7 +73,7 @@ type Excel_Workbook
        - xls_format: Whether to use the old XLS format (default is XLSX).
        - file: Optional file reference.
     from_stream : Input_Stream -> Boolean -> File | Nothing -> Excel_Workbook
-    from_stream stream xls_format=False file=Nothing = Excel_Reader.handle_bad_format file <|
+    from_stream stream xls_format:Boolean=False file=Nothing = Excel_Reader.handle_bad_format file <|
         temp_file = Temporary_File.from_stream_light stream
         Excel_Workbook.new temp_file xls_format
 
@@ -89,8 +89,8 @@ type Excel_Workbook
     ## PRIVATE
        ICON metadata
        Returns the list of databases (or catalogs) for the connection.
-    databases : Nothing
-    databases self = Nothing
+    databases : Vector (Text | Nothing)
+    databases self = [Nothing]
 
     ## PRIVATE
        ICON metadata
@@ -109,7 +109,7 @@ type Excel_Workbook
        Arguments:
         - database: The target file to open as an Excel_Workbook.
     set_database : Text | File -> Excel_Workbook ! Illegal_Argument
-    set_database self database =
+    set_database self database:(Text | File) =
         if database == self.database then self else
             file = File.new database
             if file.exists && file.is_directory.not then Excel_Workbook.new file self.xls_format else
@@ -163,7 +163,7 @@ type Excel_Workbook
        Gets the names of all the named ranges.
     named_ranges : Vector Text
     named_ranges self = self.with_java_workbook java_workbook->
-        Vector.from_polyglot_array (ExcelReader.readRangeNames java_workbook)
+        Vector.from_polyglot_array java_workbook.getRangeNames
 
     ## PRIVATE
        ICON metadata

std-bits/table/src/main/java/org/enso/table/excel/ExcelConnectionPool.java

@@ -20,6 +20,7 @@
 import org.apache.poi.poifs.filesystem.POIFSFileSystem;
 import org.apache.poi.ss.usermodel.Workbook;
 import org.apache.poi.xssf.usermodel.XSSFWorkbook;
+import org.enso.table.excel.xssfreader.XSSFReaderWorkbook;
 
 public class ExcelConnectionPool {
   public static final ExcelConnectionPool INSTANCE = new ExcelConnectionPool();
@@ -64,7 +65,7 @@ public ReadOnlyExcelConnection openReadOnlyConnection(File file, ExcelFileFormat
         record.refCount = 1;
         record.file = file;
         record.format = format;
-        record.workbook = openWorkbook(file, format, false);
+        record.reopen(true);
         records.put(key, record);
         return new ReadOnlyExcelConnection(this, key, record);
       }
@@ -212,10 +213,10 @@ static class ConnectionRecord {
     private int refCount;
     private File file;
     private ExcelFileFormat format;
-    private Workbook workbook;
+    private ExcelWorkbook workbook;
     private IOException initializationException = null;
 
-    <T> T withWorkbook(Function<Workbook, T> action) throws IOException {
+    <T> T withWorkbook(Function<ExcelWorkbook, T> action) throws IOException {
       synchronized (this) {
         return action.apply(accessCurrentWorkbook());
       }
@@ -238,7 +239,10 @@ void reopen(boolean throwOnFailure) throws IOException {
         }
 
         try {
-          workbook = openWorkbook(file, format, false);
+          workbook =
+              format == ExcelFileFormat.XLSX
+                  ? new XSSFReaderWorkbook(file.getAbsolutePath())
+                  : ExcelWorkbook.forPOIUserModel(openWorkbook(file, format, false));
         } catch (IOException e) {
           initializationException = e;
           if (throwOnFailure) {
@@ -248,7 +252,7 @@ void reopen(boolean throwOnFailure) throws IOException {
       }
     }
 
-    private Workbook accessCurrentWorkbook() throws IOException {
+    private ExcelWorkbook accessCurrentWorkbook() throws IOException {
       synchronized (this) {
         if (workbook == null) {
           if (initializationException != null) {
@@ -278,7 +282,7 @@ private static Workbook openWorkbook(File file, ExcelFileFormat format, boolean
           throw e;
         }
       }
-      case XLSX -> {
+      case XLSX, XLSX_FALLBACK -> {
         try {
           PackageAccess access = writeAccess ? PackageAccess.READ_WRITE : PackageAccess.READ;
           OPCPackage pkg = OPCPackage.open(file, access);
@@ -300,7 +304,7 @@ private static Workbook openWorkbook(File file, ExcelFileFormat format, boolean
   private static Workbook createEmptyWorkbook(ExcelFileFormat format) {
     return switch (format) {
       case XLS -> new HSSFWorkbook();
-      case XLSX -> new XSSFWorkbook();
+      case XLSX, XLSX_FALLBACK -> new XSSFWorkbook();
     };
   }
 

std-bits/table/src/main/java/org/enso/table/excel/ExcelFileFormat.java

@@ -2,5 +2,6 @@
 
 public enum ExcelFileFormat {
   XLS,
-  XLSX
+  XLSX,
+  XLSX_FALLBACK
 }

std-bits/table/src/main/java/org/enso/table/excel/ExcelHeaders.java

@@ -57,7 +57,7 @@ private static String[] readRowAsHeaders(
 
     String[] output = new String[currentEndCol - startCol + 1];
     for (int col = startCol; col <= currentEndCol; col++) {
-      String cellText = row.getFormattedCell(col);
+      String cellText = row.getCellText(col);
       String name = cellText.isEmpty() ? "" : deduplicator.makeUnique(cellText);
 
       output[col - startCol] = name;

std-bits/table/src/main/java/org/enso/table/excel/ExcelRange.java

@@ -197,7 +197,7 @@ public static ExcelRange expandSingleCell(ExcelRange excelRange, ExcelSheet shee
 
     Context context = Context.getCurrent();
     while (currentRow != null && !currentRow.isEmpty(excelRange.getLeftColumn(), rightColumn)) {
-      rightColumn = currentRow.findEndRight(rightColumn);
+      rightColumn = findEndRight(currentRow, rightColumn);
       bottomRow++;
       currentRow = sheet.get(bottomRow);
 
@@ -212,6 +212,16 @@ public static ExcelRange expandSingleCell(ExcelRange excelRange, ExcelSheet shee
         bottomRow - 1);
   }
 
+  private static int findEndRight(ExcelRow row, int start) {
+    Context context = Context.getCurrent();
+    int column = start;
+    while (!row.isEmpty(column + 1)) {
+      column++;
+      context.safepoint();
+    }
+    return column;
+  }
+
   /**
    * @param index The index to the next character after the parsed value
    * @param value Parsed integer value or 0 if not valid