Adding a boolean to drive byte tracking opt-in behavior (#14)

Adding a boolean to drive byte tracking opt-in behavior
apache · Dec 3, 2024 · 27511be · 27511be
1 parent 344f282
commit 27511be
Show file tree

Hide file tree

Showing 4 changed files with 17 additions and 8 deletions.
diff --git a/src/main/java/org/apache/commons/csv/CSVParser.java b/src/main/java/org/apache/commons/csv/CSVParser.java
@@ -153,6 +153,7 @@ public static class Builder extends AbstractStreamBuilder<CSVParser, Builder> {
         private CSVFormat format;
         private long characterOffset;
         private long recordNumber = 1;
+        private boolean enableByteTracking = false;
 
         /**
          * Constructs a new instance.
@@ -164,7 +165,7 @@ protected Builder() {
         @SuppressWarnings("resource")
         @Override
         public CSVParser get() throws IOException {
-            return new CSVParser(getReader(), format != null ? format : CSVFormat.DEFAULT, characterOffset, recordNumber, getCharset());
+            return new CSVParser(getReader(), format != null ? format : CSVFormat.DEFAULT, characterOffset, recordNumber, getCharset(), enableByteTracking);
         }
 
         /**
@@ -200,6 +201,11 @@ public Builder setRecordNumber(final long recordNumber) {
             return asThis();
         }
 
+        public Builder setEnableByteTracking(final boolean enableByteTracking) {
+            this.enableByteTracking = enableByteTracking;
+            return asThis();
+        }
+
     }
 
     final class CSVRecordIterator implements Iterator<CSVRecord> {
@@ -507,7 +513,7 @@ public CSVParser(final Reader reader, final CSVFormat format) throws IOException
     @SuppressWarnings("resource")
     public CSVParser(final Reader reader, final CSVFormat format, final long characterOffset, final long recordNumber)
         throws IOException {
-            this(reader, format, characterOffset, recordNumber, null);
+            this(reader, format, characterOffset, recordNumber, null, false);
         }
 
     /**
@@ -535,12 +541,13 @@ public CSVParser(final Reader reader, final CSVFormat format, final long charact
      * @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)
+    private CSVParser(final Reader reader, final CSVFormat format, final long characterOffset, final long recordNumber,
+        final Charset charset, final boolean enableByteTracking)
         throws IOException {
         Objects.requireNonNull(reader, "reader");
         Objects.requireNonNull(format, "format");
         this.format = format.copy();
-        this.lexer = new Lexer(format, new ExtendedBufferedReader(reader, charset));
+        this.lexer = new Lexer(format, new ExtendedBufferedReader(reader, charset, enableByteTracking));
         this.csvRecordIterator = new CSVRecordIterator();
         this.headers = createHeaders();
         this.characterOffset = characterOffset;

diff --git a/src/main/java/org/apache/commons/csv/ExtendedBufferedReader.java b/src/main/java/org/apache/commons/csv/ExtendedBufferedReader.java
@@ -67,9 +67,9 @@ final class ExtendedBufferedReader extends UnsynchronizedBufferedReader {
         super(reader);
     }
 
-    ExtendedBufferedReader(final Reader reader, Charset charset) {
+    ExtendedBufferedReader(final Reader reader, Charset charset, boolean enableByteTracking) {
         super(reader);
-        if (charset != null) {
+        if (charset != null && enableByteTracking) {
             encoder = charset.newEncoder();
         }
     }

diff --git a/src/test/java/org/apache/commons/csv/CSVParserTest.java b/src/test/java/org/apache/commons/csv/CSVParserTest.java
@@ -711,7 +711,7 @@ public void testGetRecordThreeBytesRead() throws Exception {
             .setDelimiter(',')
             .setQuote('\'')
             .get();
-        try (CSVParser parser = CSVParser.builder().setReader(new StringReader(code)).setFormat(format).setCharset(UTF_8).get() ) {
+        try (CSVParser parser = CSVParser.builder().setReader(new StringReader(code)).setFormat(format).setCharset(UTF_8).setEnableByteTracking(true).get() ) {
             CSVRecord record = new CSVRecord(parser, null, null, 1L, 0L, 0L);
 
             assertEquals(0, parser.getRecordNumber());
@@ -748,7 +748,7 @@ public void testGetRecordFourBytesRead() throws Exception {
             .setDelimiter(',')
             .setQuote('\'')
             .get();
-        try (CSVParser parser = CSVParser.builder().setReader(new StringReader(code)).setFormat(format).setCharset(UTF_8).get()) {
+        try (CSVParser parser = CSVParser.builder().setReader(new StringReader(code)).setFormat(format).setCharset(UTF_8).setEnableByteTracking(true).get()) {
             CSVRecord record = new CSVRecord(parser, null, null, 1L, 0L, 0L);
 
             assertEquals(0, parser.getRecordNumber());

diff --git a/src/test/java/org/apache/commons/csv/JiraCsv196Test.java b/src/test/java/org/apache/commons/csv/JiraCsv196Test.java
@@ -37,6 +37,7 @@ public void parseThreeBytes() throws IOException {
             .setFormat(format)
             .setReader(getTestInput("org/apache/commons/csv/CSV-196/japanese.csv"))
             .setCharset(StandardCharsets.UTF_8)
+            .setEnableByteTracking(true)
             .get();
         long[] charByteKey = {0, 89, 242, 395};
         int idx = 0;
@@ -57,6 +58,7 @@ public void parseFourBytes() throws IOException {
             .setFormat(format)
             .setReader(getTestInput("org/apache/commons/csv/CSV-196/emoji.csv"))
             .setCharset(StandardCharsets.UTF_8)
+            .setEnableByteTracking(true)
             .get();
         long[] charByteKey = {0, 84, 701, 1318, 1935};
         int idx = 0;