Package com.splout.db.hadoop

Examples of com.splout.db.hadoop.TableBuilder

36
37
38
39
40
41
42
43
44
45
46
47
  private static Schema SCHEMA_2 = new Schema("schema2", Fields.parse("id2:string, value2:string"));
  private static Schema SCHEMA_3 = new Schema("schema3", Fields.parse("id3:int, value3:string"));

  @Test
  public void testCorrectTablespace() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1").build();
    Table table2 = new TableBuilder(SCHEMA_2).addCSVTextFile("foo2.txt").partitionBy("id2").build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
    builder.build();
  }

  @Test
  public void testCorrectTablespaceMultiplePartitionBy() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1", "value1").build();
    Table table2 = new TableBuilder(SCHEMA_2).addCSVTextFile("foo2.txt").partitionBy("id2", "value2").build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
75
76
    builder.build();
  }
 
  @Test
  public void testCorrectTablespaceWithReplicated() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1").build();
    Table table2 = new TableBuilder(SCHEMA_2).addCSVTextFile("foo2.txt").partitionBy("id2").build();
    Table table3 = new TableBuilder(SCHEMA_3).addCSVTextFile("foo3.txt").replicateToAll().build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
    builder.add(table3);
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
93
  // ---- Number of partitions must be specified for Tablespace ---- //
 
  @Test(expected=TablespaceBuilderException.class)
  public void testMissingNPartitions() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1").build();
    Table table2 = new TableBuilder(SCHEMA_2).addCSVTextFile("foo2.txt").partitionBy("id2").build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
View Full Code Here
96
97
98
99
100
101
102
103
104
105
106
107
 
  // ---- At least one table must be partitioned ---- //
 
  @Test(expected=TablespaceBuilderException.class)
  public void testNoPartitionTable() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").replicateToAll().build();
    Table table2 = new TableBuilder(SCHEMA_2).addCSVTextFile("foo2.txt").replicateToAll().build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
   
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
 
  // ---- There must not be collision between table names (schema names) ---- //
 
  @Test(expected=TablespaceBuilderException.class)
  public void testTableNameCollision() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1").build();
    Table table2 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo2.txt").partitionBy("id1").build();
   
    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
   
View Full Code Here
128
129
130
131
132
133
134
135
136
137
138
139
   
  // ---- Partition field must be of same kind across tables ---- //
 
  @Test(expected=TablespaceBuilderException.class)
  public void testInvalidCoPartition() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("id1").build();
    Table table2 = new TableBuilder(SCHEMA_3).addCSVTextFile("foo2.txt").partitionBy("id3").build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
View Full Code Here
142
143
144
145
146
147
148
149
150
151
152
153
    builder.build();
  }
 
  @Test(expected=TablespaceBuilderException.class)
  public void testInvalidCoPartitionMultipleFields() throws TableBuilderException, TablespaceBuilderException {
    Table table1 = new TableBuilder(SCHEMA_1).addCSVTextFile("foo1.txt").partitionBy("value1", "id1").build();
    Table table2 = new TableBuilder(SCHEMA_3).addCSVTextFile("foo2.txt").partitionBy("value3", "id3").build();

    TablespaceBuilder builder = new TablespaceBuilder();
    builder.add(table1);
    builder.add(table2);
View Full Code Here
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
      // define the schema of the input files: projectcode, pagename, pageviews, bytes
      Schema fileSchema = new Schema("pagecountsfile",
          Fields.parse("projectcode:string, pagename:string, pageviews:int, bytes:long"));

      // instantiate a TableBuilder
      TableBuilder tableBuilder = new TableBuilder(tableSchema);

      // for every input file...
      for(FileStatus fileStatus : fileStatuses) {
        String fileName = fileStatus.getPath().getName().toString();
        // strip the date and the hour from the file name
        String fileDate = fileName.split("-")[1];
        String fileHour = fileName.split("-")[2].substring(0, 2);
        // instantiate a custom RecordProcessor to process the records of this file
        PageCountsRecordProcessor recordProcessor = new PageCountsRecordProcessor(tableSchema, fileDate,
            fileHour);
        // use the tableBuilder method for adding each of the files to the mix
        tableBuilder.addCSVTextFile(fileStatus.getPath(), ' ', TupleTextInputFormat.NO_QUOTE_CHARACTER,
            TupleTextInputFormat.NO_ESCAPE_CHARACTER, false, false, TupleTextInputFormat.NO_NULL_STRING,
            fileSchema, recordProcessor);
      }

      // partition the dataset by pagename - which should give a fair even distribution.
      tableBuilder.partitionBy("pagename");
      // create a compound index on pagename, date so that typical queries for the dataset will be fast
      tableBuilder.createIndex("pagename", "date");

      long nonExactPageSize = memoryForIndexing / 32000; // number of pages
      int pageSize = (int) Math.pow(2, (int) Math.round(Math.log(nonExactPageSize) / Math.log(2)));
      Log.info("Pagesize = " + pageSize + " as memory for indexing was [" + memoryForIndexing
          + "] and there are 32000 pages.");

      tableBuilder.initialSQL("pragma page_size=" + pageSize);
      // insertion order is very important for optimizing query speed because it makes data be co-located in disk
      tableBuilder.insertionSortOrder(OrderBy.parse("pagename:asc, date:asc"));

      // instantiate a TablespaceBuilder
      TablespaceBuilder tablespaceBuilder = new TablespaceBuilder();

      // we will partition this dataset in as many partitions as:
      tablespaceBuilder.setNPartitions(nPartitions);
      tablespaceBuilder.add(tableBuilder.build());
      // we turn a specific SQLite pragma on for making autocomplete queries fast
      tablespaceBuilder.initStatements("pragma case_sensitive_like=true;");

      HadoopUtils.deleteIfExists(outFs, outPath);
View Full Code Here
TOP

Related Classes of com.splout.db.hadoop.TableBuilder

  • com.datasalt.pangool.io.Schema
  • com.datasalt.pangool.io.Schema.Field
  • com.datasalt.pangool.tuplemr.mapred.lib.input.CascadingTupleInputFormat
  • com.datasalt.pangool.tuplemr.mapred.lib.input.HCatTupleInputFormat
  • com.datasalt.pangool.tuplemr.mapred.lib.input.TupleInputFormat
  • com.datasalt.pangool.tuplemr.mapred.lib.input.TupleTextInputFormat
  • com.splout.db.examples.PageCountsExample
  • com.splout.db.hadoop.TableSpec.FieldIndex
  • com.splout.db.hadoop.TestTablespaceBuilder
  • org.apache.hadoop.fs.Path
    • 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.