Github user ueshin commented on a diff in the pull request: https://github.com/apache/spark/pull/19769#discussion_r151597335 --- Diff: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedParquetRecordReader.java --- @@ -105,13 +108,23 @@ */ private static final MemoryMode DEFAULT_MEMORY_MODE = MemoryMode.ON_HEAP; + private TimeZone convertTz = null; + /** * Implementation of RecordReader API. */ @Override public void initialize(InputSplit inputSplit, TaskAttemptContext taskAttemptContext) throws IOException, InterruptedException, UnsupportedOperationException { super.initialize(inputSplit, taskAttemptContext); + Configuration hadoopConf = taskAttemptContext.getConfiguration(); + // this is *not* the raw value in the sql conf -- we've already looked at the createdBy field + // for this particular file and adjusted that value accordingly by this point. + boolean doTsConversion = + Boolean.valueOf(hadoopConf.get(SQLConf.PARQUET_INT96_TIMESTAMP_CONVERSION().key())); + if (doTsConversion) { + convertTz = TimeZone.getTimeZone(hadoopConf.get(SQLConf.SESSION_LOCAL_TIMEZONE().key())); --- End diff -- We can use `DateTimeUtils.getTimeZone()` which caches timezone object instead of `TimeZone.getTimeZone()` which calculates timezone object each time.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org