Package com.asakusafw.runtime.directio.hadoop

Examples of com.asakusafw.runtime.directio.hadoop.BlockMap

140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
                        context.getDataType().getSimpleName(),
                        status.getPath(),
                        orc.getNumberOfRows(),
                        orc.getRawDataSize()));
            }
            BlockMap blockMap = BlockMap.create(
                    status.getPath().toString(),
                    status.getLen(),
                    BlockMap.computeBlocks(context.getFileSystem(), status),
                    false);
            // TODO configurable split
            for (StripeInformation stripe : orc.getStripes()) {
                long begin = stripe.getOffset();
                long end = begin + stripe.getLength();
                DirectInputFragment fragment = blockMap.get(begin, end);
                if (LOG.isDebugEnabled()) {
                    LOG.debug(MessageFormat.format(
                            "Detect ORCFile stripe: path={0}, rows={1}, range={2}+{3}, allocation={4}",
                            fragment.getPath(),
                            stripe.getNumberOfRows(),
View Full Code Here
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
                LOG.info(MessageFormat.format(
                        "Analyzing Parquet file metadata ({0}): {1}",
                        context.getDataType().getSimpleName(),
                        status.getPath()));
            }
            BlockMap blockMap = BlockMap.create(
                    status.getPath().toString(),
                    status.getLen(),
                    BlockMap.computeBlocks(context.getFileSystem(), status),
                    false);
            for (BlockMetaData block : footer.getParquetMetadata().getBlocks()) {
                if (block.getColumns().isEmpty()) {
                    continue;
                }
                long begin = Long.MAX_VALUE;
                long end = -1L;
                for (ColumnChunkMetaData column : block.getColumns()) {
                    long offset = column.getFirstDataPageOffset();
                    long size = column.getTotalSize();
                    begin = Math.min(begin, offset);
                    end = Math.max(end, offset + size);
                }
                assert begin >= 0;
                assert end >= 0;
                DirectInputFragment fragment = blockMap.get(begin, end);
                if (LOG.isDebugEnabled()) {
                    LOG.debug(MessageFormat.format(
                            "Detect Parquet file block: path={0}, rows={1}, range={2}+{3}, allocation={4}",
                            status.getPath(),
                            block.getRowCount(),
View Full Code Here
99
100
101
102
103
104
105
106
107
108
109
            FileStatus[] statuses = fs.globStatus(path);
            if (statuses == null) {
                continue;
            }
            for (FileStatus status : statuses) {
                BlockMap blockMap = BlockMap.create(
                        status.getPath().toString(),
                        status.getLen(),
                        BlockMap.computeBlocks(fs, status),
                        false);
                results.addAll(computeSplits(status.getPath(), blockMap, splitSize));
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
    /**
     * simple case for computing splits.
     */
    @Test
    public void splits_simple() {
        BlockMap blocks = blocks("testing", m(10));
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(new Path("testing"), blocks, m(64));

        assertThat(splits, hasSize(1));
        FileSplit s0 = find(splits, 0);
        assertThat(s0.getLength(), is(m(10)));
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
    /**
     * computing splits with already aligned blocks.
     */
    @Test
    public void splits_aligned() {
        BlockMap blocks = blocks("testing", TemporaryFile.BLOCK_SIZE, TemporaryFile.BLOCK_SIZE);
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(new Path("testing"), blocks, m(64));

        assertThat(splits, hasSize(2));

        FileSplit s0 = find(splits, 0);
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
    /**
     * computing splits without unaligned blocks.
     */
    @Test
    public void splits_unaligned() {
        BlockMap blocks = blocks("testing", TemporaryFile.BLOCK_SIZE - 10, TemporaryFile.BLOCK_SIZE);
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(new Path("testing"), blocks, m(128));

        assertThat(splits, hasSize(2));

        FileSplit s0 = find(splits, 0);
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
    /**
     * computing splits with aligned blocks plus.
     */
    @Test
    public void splits_aligned_rest() {
        BlockMap blocks = blocks("testing", TemporaryFile.BLOCK_SIZE, TemporaryFile.BLOCK_SIZE + 10);
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(new Path("testing"), blocks, m(64));

        assertThat(splits, hasSize(2));

        FileSplit s0 = find(splits, 0);
View Full Code Here
120
121
122
123
124
125
126
127
128
129
130
    /**
     * computing splits with forcibly splitting.
     */
    @Test
    public void splits_force() {
        BlockMap blocks = blocks("testing", TemporaryFile.BLOCK_SIZE * 10);
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(
                new Path("testing"), blocks, TemporaryFile.BLOCK_SIZE + 1);

        assertThat(splits, hasSize(5));

View Full Code Here
143
144
145
146
147
148
149
150
151
152
153
    /**
     * computing splits w/ suppress.
     */
    @Test
    public void splits_suppress() {
        BlockMap blocks = blocks("testing", TemporaryFile.BLOCK_SIZE * 10);
        List<FileSplit> splits = TemporaryInputFormat.computeSplits(new Path("testing"), blocks, 0);

        assertThat(splits, hasSize(1));
        FileSplit s0 = find(splits, 0);
        assertThat(s0.getLength(), is((long) TemporaryFile.BLOCK_SIZE * 10));
 
View Full Code Here
TOP

Related Classes of com.asakusafw.runtime.directio.hadoop.BlockMap

  • com.asakusafw.directio.hive.orc.AbstractOrcFileFormat
  • com.asakusafw.directio.hive.parquet.AbstractParquetFileFormat
  • com.asakusafw.runtime.directio.DirectInputFragment
  • com.asakusafw.runtime.stage.input.TemporaryInputFormat
  • com.asakusafw.runtime.stage.input.TemporaryInputFormatTest
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.