Merge pull request #89 from ITRS-Group/master

Changed logic in Parser _readHeaderLine

Author: cowtowncoder

src/main/java/com/fasterxml/jackson/dataformat/csv/CsvParser.java

@@ -41,7 +41,7 @@ public enum Feature // implements FormatFeature // for 2.7
          * Default value is false, as per <a href="http://tools.ietf.org/html/rfc4180">RFC-4180</a>.
          */
         TRIM_SPACES(false),
-        
+
         /**
          * Feature that determines how stream of records (usually CSV lines, but sometimes
          * multiple lines when linefeeds are included in quoted values) is exposed:
@@ -802,27 +802,36 @@ protected JsonToken _handleArrayValue() throws IOException
     /**
      * Method called to process the expected header line
      */
-    protected void _readHeaderLine() throws IOException
-    {
-        /* Two separate cases:
-         * 
-         * (a) We already have a Schema with columns; if so, header will be skipped
-         * (b) Otherwise, need to find column definitions; empty one is not acceptable
+    protected void _readHeaderLine() throws IOException {
+        /*
+            When the header line is present and the settings ask for it
+            to be processed, two different options are possible:
+
+            a) The schema has been populated.  In this case, build a new
+               schema where the order matches the *actual* order in which
+               the given CSV file offers its columns, iif _schema.reordersColumns()
+               is set to true; there cases the consumer of the csv file
+               knows about the columns but not necessarily the order in
+               which they are defined.
+
+            b) The schema has not been populated.  In this case, build a
+               default schema based on the columns found in the header.
          */
 
-        if (_schema.size() > 0) { // case (a); skip all/any
-            while (_reader.nextString() != null) { }
+        if (_schema.size() > 0 && !_schema.reordersColumns()) {
+            //noinspection StatementWithEmptyBody
+            while (_reader.nextString() != null) { /* does nothing */ }
             return;
         }
-        // case (b); read all
+
+        // either the schema is empty or reorder columns flag is set
         String name;
-        // base setting on existing schema, but drop columns
         CsvSchema.Builder builder = _schema.rebuild().clearColumns();
-        
+
         while ((name = _reader.nextString()) != null) {
             // one more thing: always trim names, regardless of config settings
             name = name.trim();
-            
+
             // See if "old" schema defined type; if so, use that type...
             CsvSchema.Column prev = _schema.column(name);
             if (prev != null) {
@@ -831,6 +840,7 @@ protected void _readHeaderLine() throws IOException
                 builder.addColumn(name);
             }
         }
+
         // Ok: did we get any  columns?
         CsvSchema newSchema = builder.build();
         int size = newSchema.size();

src/main/java/com/fasterxml/jackson/dataformat/csv/CsvSchema.java

@@ -81,6 +81,7 @@ public class CsvSchema
     protected final static int ENCODING_FEATURE_USE_HEADER = 0x0001;
     protected final static int ENCODING_FEATURE_SKIP_FIRST_DATA_ROW = 0x0002;
     protected final static int ENCODING_FEATURE_ALLOW_COMMENTS = 0x0004;
+    protected final static int ENCODING_FEATURE_REORDER_COLUMNS = 0x0008;
 
     protected final static int DEFAULT_ENCODING_FEATURES = 0;
 
@@ -454,6 +455,19 @@ public Builder setUseHeader(boolean b) {
             return this;
         }
 
+        /**
+         * Use in combination with setUseHeader.  When use header flag is
+         * is set, this setting will reorder the columns defined in this
+         * schema to match the order set by the header.
+         *
+         * @param b         Enable / Disable this setting
+         * @return          This Builder instance
+         */
+        public Builder setReorderColumns(boolean b) {
+            _feature(ENCODING_FEATURE_REORDER_COLUMNS, b);
+            return this;
+        }
+
         /**
          * Method for specifying whether Schema should indicate that
          * the first line that is not a header (if header handling enabled)
@@ -774,6 +788,18 @@ public CsvSchema withUseHeader(boolean state) {
         return _withFeature(ENCODING_FEATURE_USE_HEADER, state);
     }
 
+    /**
+     * Returns a clone of this instance by changing or setting the
+     * column reordering flag
+     *
+     * @param state     New value for setting
+     * @return          A copy of itself, ensuring the setting for
+     *                  the column reordering feature.
+     */
+    public CsvSchema withColumnReordering(boolean state) {
+        return _withFeature(ENCODING_FEATURE_REORDER_COLUMNS, state);
+    }
+
     /**
      * Helper method for constructing and returning schema instance that
      * is similar to this one, except that it will be using header line.
@@ -968,6 +994,7 @@ public String getSchemaType() {
      */
 
     public boolean usesHeader() { return (_features & ENCODING_FEATURE_USE_HEADER) != 0; }
+    public boolean reordersColumns() { return (_features & ENCODING_FEATURE_REORDER_COLUMNS) != 0; }
     public boolean skipsFirstDataRow() { return (_features & ENCODING_FEATURE_SKIP_FIRST_DATA_ROW) != 0; }
     public boolean allowsComments() { return (_features & ENCODING_FEATURE_ALLOW_COMMENTS) != 0; }