fhueske commented on a change in pull request #6823: [FLINK-10134] UTF-16
support for TextInputFormat bug refixed
URL: https://github.com/apache/flink/pull/6823#discussion_r225543780
##########
File path:
flink-core/src/main/java/org/apache/flink/api/common/io/DelimitedInputFormat.java
##########
@@ -504,16 +533,119 @@ private void initBuffers() {
this.end = false;
}
+ /**
+ * Set file bom encoding.
+ *
+ * @param split
+ */
+ private void setBomFileCharset(FileInputSplit split) {
+ try {
+ String filePath = split.getPath().toString();
+ if (this.fileBomCharsetMap.containsKey(filePath)) {
+ this.bomCharset =
this.fileBomCharsetMap.get(filePath);
+ } else {
+ byte[] checkBytes = new byte[]{(byte) 0x00,
(byte) 0xFE, (byte) 0xFF, (byte) 0xEF, (byte) 0xBB, (byte) 0xBF};
+ byte[] bomBuffer = new byte[4];
+
+ if (this.splitStart != 0) {
+ this.stream.seek(0);
+ this.stream.read(bomBuffer, 0,
bomBuffer.length);
+ this.stream.seek(split.getStart());
+ } else {
+ System.arraycopy(this.readBuffer, 0,
bomBuffer, 0, 3);
+ }
+
+ if ((bomBuffer[0] == checkBytes[0]) &&
(bomBuffer[1] == checkBytes[0]) && (bomBuffer[2] == checkBytes[1])
+ && (bomBuffer[3] == checkBytes[2])) {
+ this.bomCharset =
Charset.forName("UTF-32BE");
+ } else if ((bomBuffer[0] == checkBytes[2]) &&
(bomBuffer[1] == checkBytes[1]) && (bomBuffer[2] == checkBytes[0])
+ && (bomBuffer[3] == checkBytes[0])) {
+ this.bomCharset =
Charset.forName("UTF-32LE");
+ } else if ((bomBuffer[0] == checkBytes[3]) &&
(bomBuffer[1] == checkBytes[4]) && (bomBuffer[2] == checkBytes[5])) {
+ this.bomCharset =
Charset.forName("UTF-8");
+ } else if ((bomBuffer[0] == checkBytes[1]) &&
(bomBuffer[1] == checkBytes[2])) {
+ this.bomCharset =
Charset.forName("UTF-16BE");
+ } else if ((bomBuffer[0] == checkBytes[2]) &&
(bomBuffer[1] == checkBytes[1])) {
+ this.bomCharset =
Charset.forName("UTF-16LE");
+ } else {
+ this.bomCharset =
Charset.forName(charsetName);
+ }
+
+ this.fileBomCharsetMap.put(filePath,
this.bomCharset);
+ }
+ } catch (Exception e) {
+ LOG.warn("Failed to get file bom encoding.");
+ this.bomCharset = Charset.forName(charsetName);
+ }
+
+ setParasByCharset(this.bomCharset);
+ }
+
+ /**
+ * Set stepSize, delimiterNewLinePos, delimiterCarrageReturnPos by
charset.
+ *
+ * @param charset
+ */
+ private void setParasByCharset(Charset charset) {
+ int stepSize;
+ byte[] delimiterBytes = new byte[]{'\n'};
+ switch (charset.toString()) {
+ case "UTF-8":
+ stepSize = 1;
+ break;
+ case "UTF-16":
+ stepSize = 2;
+ this.delimiterNewLinePos = 1;
+ delimiterBytes = new byte[]{(byte) 0x00, '\n'};
+ break;
+ case "UTF-16LE":
+ stepSize = 2;
+ delimiterBytes = new byte[]{'\n', (byte) 0x00};
+ this.delimiterNewLinePos = 0;
+ break;
+ case "UTF-16BE":
+ stepSize = 2;
+ this.delimiterNewLinePos = 1;
+ delimiterBytes = new byte[]{(byte) 0x00, '\n'};
+ break;
+ case "UTF-32":
Review comment:
We should not have "UTF-32" but only "UTF-32BE"
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
