Package com.mapr.synth.samplers

Examples of com.mapr.synth.samplers.SchemaSampler

200
201
202
203
204
205
206
207
208
209
210
211
212
213
        for (int i = 0; i < 20; i++) {
            out.write(i + "\t" + (i * i) + "\n");
        }
        out.close();

        SchemaSampler s = new SchemaSampler(Resources.asCharSource(Resources.getResource("schema008.json"), Charsets.UTF_8).read());

        for (int k = 0; k < 1000; k++) {
            JsonNode r = s.sample();
            assertEquals(6, r.get("x").get("x").asInt() + r.get("x").get("y").asInt());
            int i = r.get("y").get("a").asInt();
            assertEquals(i * i, r.get("y").get("b").asInt());
        }
    }
View Full Code Here
212
213
214
215
216
217
218
219
220
221
222
223
224
225
        }
    }

    @Test
    public void testJoin() throws IOException {
        SchemaSampler s = new SchemaSampler(Resources.asCharSource(Resources.getResource("schema009.json"), Charsets.UTF_8).read());

        for (int k = 0; k < 10; k++) {
            JsonNode r = s.sample();
            assertEquals("3,6,8", r.get("x").asText());
            assertTrue(r.get("y").asInt() >= 1 && r.get("y").asInt() < 5);
            assertTrue(r.get("z").asText().matches("(xyz(,xyz)*)?"));
        }
    }
View Full Code Here
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
        }
    }

    @Test
    public void testEvents() throws IOException, ParseException {
        SchemaSampler s = new SchemaSampler(Resources.asCharSource(Resources.getResource("schema012.json"), Charsets.UTF_8).read());
        long t = System.currentTimeMillis();

        SimpleDateFormat df0 = new SimpleDateFormat("yyyy-MM-dd");
        SimpleDateFormat df1 = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
        SimpleDateFormat df2 = new SimpleDateFormat("MM/dd/yyyy HH:mm:ss");
        JsonNode old = s.sample();

        long old1 = df0.parse(old.get("foo1").asText()).getTime();
        assertTrue(Math.abs(old1 - t) < TimeUnit.MILLISECONDS.convert(1, TimeUnit.DAYS));

        long old2 = df1.parse(old.get("foo2").asText()).getTime();
        assertEquals((double) old2, df1.parse("2014-01-01 00:00:00").getTime(), 10.0);

        long old3 = df2.parse(old.get("foo3").asText()).getTime();
        assertEquals(old3, df1.parse("2014-02-01 00:00:00").getTime(), 10);

        double sum1 = 0;
        double sum2 = 0;
        double sum3 = 0;

        final int N = 10000;

        for (int k = 0; k < N; k++) {
            JsonNode r = s.sample();

            long t1 = df0.parse(r.get("foo1").asText()).getTime();
            sum1 += t1 - old1;
            old1 = t1;
View Full Code Here
270
271
272
273
274
275
276
277
278
279
280
281
282
283
        assertEquals(2000, sum3 / N, 2000 * 0.03);
    }

    @Test
    public void testFlatten() throws IOException {
        SchemaSampler s = new SchemaSampler(Resources.asCharSource(Resources.getResource("schema010.json"), Charsets.UTF_8).read());

        for (int k = 0; k < 10; k++) {
            JsonNode r = s.sample();
            assertEquals(k, r.get("id").asInt());
            assertTrue(r.get("stuff").isArray());
            assertEquals(1, r.get("stuff").get(0).asInt());
            assertEquals(2, r.get("stuff").get(1).asInt());
            assertEquals(3, r.get("stuff").get(2).asInt());
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
        }

        Joiner withTab = Joiner.on("\t");

        // first generate lots of user definitions
        SchemaSampler users = new SchemaSampler(Resources.asCharSource(Resources.getResource("user-schema.txt"), Charsets.UTF_8).read());
        File userFile = File.createTempFile("user", "tsv");
        BufferedWriter out = Files.newBufferedWriter(userFile.toPath(), Charsets.UTF_8);
        for (int i = 0; i < opts.users; i++) {
            out.write(withTab.join(users.sample()));
            out.newLine();
        }
        out.close();

        // now generate a session for each user
        Splitter onTabs = Splitter.on("\t");
        Splitter onComma = Splitter.on(",");

        Random gen = new Random();
        SchemaSampler intermediate = new SchemaSampler(Resources.asCharSource(Resources.getResource("hit_step.txt"), Charsets.UTF_8).read());

        final int COUNTRY = users.getFieldNames().indexOf("country");
        final int CAMPAIGN = intermediate.getFieldNames().indexOf("campaign_list");
        final int SEARCH_TERMS = intermediate.getFieldNames().indexOf("search_keywords");
        Preconditions.checkState(COUNTRY >= 0, "Need country field in user schema");
        Preconditions.checkState(CAMPAIGN >= 0, "Need campaign_list field in step schema");
        Preconditions.checkState(SEARCH_TERMS >= 0, "Need search_keywords field in step schema");

        out = Files.newBufferedWriter(new File(opts.out).toPath(), Charsets.UTF_8);

        for (String line : Files.readAllLines(userFile.toPath(), Charsets.UTF_8)) {
            long t = (long) (TimeUnit.MILLISECONDS.convert(30, TimeUnit.DAYS) * gen.nextDouble());
            List<String> user = Lists.newArrayList(onTabs.split(line));

            // pick session length
            int n = (int) Math.floor(-30 * Math.log(gen.nextDouble()));

            for (int i = 0; i < n; i++) {
                // time on page
                int dt = (int) Math.floor(-20000 * Math.log(gen.nextDouble()));
                t += dt;

                // hit specific values
                JsonNode step = intermediate.sample();

                // check for purchase
                double p = 0.01;
                List<String> campaigns = Lists.newArrayList(onComma.split(step.get("campaign_list").asText()));
                List<String> keywords = Lists.newArrayList(onComma.split(step.get("search_keywords").asText()));
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
            }
            Preconditions.checkArgument(outputDir.exists() && outputDir.isDirectory(),
                    String.format("Couldn't create directory %s", opts.output));
        }

        final SchemaSampler sampler = new SchemaSampler(opts.schema);
        final AtomicLong rowCount = new AtomicLong();

        final List<ReportingWorker> tasks = Lists.newArrayList();
        int limit = (opts.count + opts.threads - 1) / opts.threads;
        int remaining = opts.count;
View Full Code Here
TOP

Related Classes of com.mapr.synth.samplers.SchemaSampler

  • com.fasterxml.jackson.databind.node.ObjectNode
  • com.fasterxml.jackson.databind.ObjectMapper
  • com.mapr.PurchaseLog
  • com.mapr.synth.CommonPointOfCompromiseTest
  • com.mapr.synth.SchemaSamplerTest
  • com.mapr.synth.Synth
    • 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.