zeroshade commented on code in PR #38367:
URL: https://github.com/apache/arrow/pull/38367#discussion_r1376389878
##########
go/parquet/file/file_reader_test.go:
##########
@@ -385,3 +387,56 @@ func TestDeltaLengthByteArrayPackingWithNulls(t
*testing.T) {
assert.NotNil(t, readData[0])
}
}
+
+func TestRleBooleanEncodingFileRead(t *testing.T) {
+ dir := os.Getenv("PARQUET_TEST_DATA")
+ if dir == "" {
+ t.Skip("no path supplied with PARQUET_TEST_DATA")
+ }
+ assert.DirExists(t, dir)
+
+ props := parquet.NewReaderProperties(memory.DefaultAllocator)
+ fileReader, err := file.OpenParquetFile(path.Join(dir,
"rle_boolean_encoding.parquet"),
+ false, file.WithReadProps(props))
+ require.NoError(t, err)
+ defer fileReader.Close()
+
+ assert.Equal(t, 1, fileReader.NumRowGroups())
+ rgr := fileReader.RowGroup(0)
+ assert.EqualValues(t, 68, rgr.NumRows())
+
+ rdr, err := rgr.Column(0)
+ require.NoError(t, err)
+ brdr := rdr.(*file.BooleanColumnChunkReader)
+
+ values := make([]bool, 68)
+ defLvls, repLvls := make([]int16, 68), make([]int16, 68)
+ total, read, err := brdr.ReadBatch(68, values, defLvls, repLvls)
+ require.NoError(t, err)
+
+ assert.EqualValues(t, 68, total)
+ md, err := rgr.MetaData().ColumnChunk(0)
+ require.NoError(t, err)
+ stats, err := md.Statistics()
+ require.NoError(t, err)
+ assert.EqualValues(t, total-stats.NullCount(), read)
+
+ expected := []bool{
+ true, false, true, true, false, false,
+ true, true, true, false, false, true, true,
+ false, true, true, false, false, true, true,
+ false, true, true, false, false, true, true,
+ true, false, false, false, false, true, true,
+ false, true, true, false, false, true, true,
+ true, false, false, true, true, false, false,
+ true, true, true, false, true, true, false,
+ true, true, false, false, true, true, true,
+ }
+ expectedNulls := []int{2, 15, 23, 38, 48, 60}
+
+ for _, v := range expectedNulls {
+ assert.Zero(t, defLvls[v])
+ }
Review Comment:
updated test to confirm other values are 1
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]