CSV-196-master: More changes (#16)

marklogic · Dec 27, 2024 · bdd152f · bdd152f
1 parent 8387f79
commit bdd152f
Show file tree

Hide file tree

Showing 6 changed files with 21 additions and 30 deletions.
diff --git a/src/main/java/org/apache/commons/csv/CSVParser.java b/src/main/java/org/apache/commons/csv/CSVParser.java
@@ -539,13 +539,14 @@ public CSVParser(final Reader reader, final CSVFormat format, final long charact
      * @param recordNumber
      *            The next record number to assign.
      * @param charset
-     *            The character encoding to be used for the reader.
+     *            The character encoding to be used for the reader when enableByteTracking is true.
+     * @param enableByteTracking
+     *           {@code true} to enable byte tracking for the parser; {@code false} to disable it.
      * @throws IllegalArgumentException
      *             If the parameters of the format are inconsistent or if either the reader or format is null.
      * @throws IOException
      *             If there is a problem reading the header or skipping the first record.
      * @throws CSVException Thrown on invalid input.
-     * @since 1.13.0.
      */
     private CSVParser(final Reader reader, final CSVFormat format, final long characterOffset, final long recordNumber,
         final Charset charset, final boolean enableByteTracking)

diff --git a/src/main/java/org/apache/commons/csv/CSVRecord.java b/src/main/java/org/apache/commons/csv/CSVRecord.java
@@ -51,7 +51,7 @@ public final class CSVRecord implements Serializable, Iterable<String> {
     /**
      * The start byte of this record as a character byte in the source stream.
      */
-    private final long characterByte;
+    private final long bytePosition;
 
     /** The accumulated comments (if any) */
     private final String comment;
@@ -65,24 +65,14 @@ public final class CSVRecord implements Serializable, Iterable<String> {
     /** The parser that originates this record. This is not serialized. */
     private final transient CSVParser parser;
 
-    CSVRecord(final CSVParser parser, final String[] values, final String comment, final long recordNumber,
-            final long characterPosition) {
-        this.recordNumber = recordNumber;
-        this.values = values != null ? values : Constants.EMPTY_STRING_ARRAY;
-        this.parser = parser;
-        this.comment = comment;
-        this.characterPosition = characterPosition;
-        this.characterByte = 0L;
-    }
-
     CSVRecord(final CSVParser parser, final String[] values,  final String comment, final long recordNumber,
-            final long characterPosition, final long characterByte) {
+            final long characterPosition, final long bytePosition) {
         this.recordNumber = recordNumber;
         this.values = values != null ? values : Constants.EMPTY_STRING_ARRAY;
         this.parser = parser;
         this.comment = comment;
         this.characterPosition = characterPosition;
-        this.characterByte = characterByte;
+        this.bytePosition = bytePosition;
     }
     /**
      * Returns a value by {@link Enum}.
@@ -164,8 +154,8 @@ public long getCharacterPosition() {
      *
      * @return the start byte of this record as a character byte in the source stream.
      */
-    public long getCharacterByte() {
-        return characterByte;
+    public long getBytePosition() {
+        return bytePosition;
     }
 
     /**

diff --git a/src/main/java/org/apache/commons/csv/ExtendedBufferedReader.java b/src/main/java/org/apache/commons/csv/ExtendedBufferedReader.java
@@ -147,7 +147,7 @@ public int read() throws IOException {
             lineNumber++;
         }
         if (encoder != null) {
-            this.bytesRead += getCharBytes(current);
+            this.bytesRead += getEncodedCharLength(current);
         }
         lastChar = current;
         position++;
@@ -180,7 +180,7 @@ public int read() throws IOException {
      * @return the byte length of the character.
      * @throws CharacterCodingException if the character cannot be encoded.
      */
-    private long getCharBytes(int current) throws CharacterCodingException {
+    private int getEncodedCharLength(int current) throws CharacterCodingException {
         final char cChar = (char) current;
         final char lChar = (char) lastChar;
         if (!Character.isSurrogate(cChar)) {

diff --git a/src/test/java/org/apache/commons/csv/CSVParserTest.java b/src/test/java/org/apache/commons/csv/CSVParserTest.java
@@ -718,22 +718,22 @@ public void testGetRecordThreeBytesRead() throws Exception {
             assertNotNull(record = parser.nextRecord());
             assertEquals(1, record.getRecordNumber());
             assertEquals(code.indexOf('i'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), record.getCharacterPosition());
+            assertEquals(record.getBytePosition(), record.getCharacterPosition());
 
             assertNotNull(record = parser.nextRecord());
             assertEquals(2, record.getRecordNumber());
             assertEquals(code.indexOf('1'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), record.getCharacterPosition());
+            assertEquals(record.getBytePosition(), record.getCharacterPosition());
 
             assertNotNull(record = parser.nextRecord());
             assertEquals(3, record.getRecordNumber());
             assertEquals(code.indexOf('2'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), 95);
+            assertEquals(record.getBytePosition(), 95);
 
             assertNotNull(record = parser.nextRecord());
             assertEquals(4, record.getRecordNumber());
             assertEquals(code.indexOf('3'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), 154);
+            assertEquals(record.getBytePosition(), 154);
         };
 
     }
@@ -755,20 +755,20 @@ public void testGetRecordFourBytesRead() throws Exception {
             assertNotNull(record = parser.nextRecord());
             assertEquals(1, record.getRecordNumber());
             assertEquals(code.indexOf('i'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), record.getCharacterPosition());
+            assertEquals(record.getBytePosition(), record.getCharacterPosition());
 
             assertNotNull(record = parser.nextRecord());
             assertEquals(2, record.getRecordNumber());
             assertEquals(code.indexOf('1'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), record.getCharacterPosition());
+            assertEquals(record.getBytePosition(), record.getCharacterPosition());
             assertNotNull(record = parser.nextRecord());
             assertEquals(3, record.getRecordNumber());
             assertEquals(code.indexOf('2'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), 26);
+            assertEquals(record.getBytePosition(), 26);
             assertNotNull(record = parser.nextRecord());
             assertEquals(4, record.getRecordNumber());
             assertEquals(code.indexOf('3'), record.getCharacterPosition());
-            assertEquals(record.getCharacterByte(), 43);
+            assertEquals(record.getBytePosition(), 43);
         }
     }
 

diff --git a/src/test/java/org/apache/commons/csv/CSVRecordTest.java b/src/test/java/org/apache/commons/csv/CSVRecordTest.java
@@ -85,7 +85,7 @@ record = parser.iterator().next();
     @Test
     public void testCSVRecordNULLValues() throws IOException {
         try (CSVParser parser = CSVParser.parse("A,B\r\nONE,TWO", CSVFormat.DEFAULT.withHeader())) {
-            final CSVRecord csvRecord = new CSVRecord(parser, null, null, 0L, 0L);
+            final CSVRecord csvRecord = new CSVRecord(parser, null, null, 0L, 0L, 0L);
             assertEquals(0, csvRecord.size());
             assertThrows(IllegalArgumentException.class, () -> csvRecord.get("B"));
         }

diff --git a/src/test/java/org/apache/commons/csv/JiraCsv196Test.java b/src/test/java/org/apache/commons/csv/JiraCsv196Test.java
@@ -42,7 +42,7 @@ public void parseThreeBytes() throws IOException {
         long[] charByteKey = {0, 89, 242, 395};
         int idx = 0;
         for (CSVRecord record : parser) {
-            assertEquals(charByteKey[idx++], record.getCharacterByte());
+            assertEquals(charByteKey[idx++], record.getBytePosition());
         }
         parser.close();
     }
@@ -63,7 +63,7 @@ public void parseFourBytes() throws IOException {
         long[] charByteKey = {0, 84, 701, 1318, 1935};
         int idx = 0;
         for (CSVRecord record : parser) {
-            assertEquals(charByteKey[idx++], record.getCharacterByte());
+            assertEquals(charByteKey[idx++], record.getBytePosition());
         }
         parser.close();
     }