Examples with Page org.embulk.spi.Page used on opensource projects

Example 1 with Page

use of org.embulk.spi.Page in project embulk by embulk.

the class Pages method toObjects.

// TODO use streaming and return Iterable
public static List<Object[]> toObjects(Schema schema, Iterable<Page> pages) {
    ImmutableList.Builder<Object[]> builder = ImmutableList.builder();
    Iterator<Page> ite = pages.iterator();
    try (PageReader reader = new PageReader(schema)) {
        while (ite.hasNext()) {
            reader.setPage(ite.next());
            while (reader.nextRecord()) {
                builder.add(toObjects(reader));
            }
        }
    }
    return builder.build();
}

Example 2 with Page

use of org.embulk.spi.Page in project embulk by embulk.

the class GuessExecutor method guessParserConfig.

private ConfigDiff guessParserConfig(Buffer sample, ConfigSource config, List<PluginType> guessPlugins, final int guessParserSampleBufferBytes) {
    // repeat guessing upto 10 times
    ConfigDiff lastGuessed = Exec.newConfigDiff();
    for (int i = 0; i < 10; i++) {
        // include last-guessed config to run guess input
        ConfigSource originalConfig = config.deepCopy().merge(lastGuessed);
        ConfigSource guessInputConfig = originalConfig.deepCopy();
        guessInputConfig.getNestedOrSetEmpty("parser").set("type", // override in.parser.type so that FileInputRunner.run uses GuessParserPlugin
        "system_guess").set("guess_plugins", guessPlugins).set("orig_config", originalConfig).set("guess_parser_sample_buffer_bytes", guessParserSampleBufferBytes);
        // run FileInputPlugin
        final FileInputRunner input = new FileInputRunner(new BufferFileInputPlugin(sample));
        ConfigDiff guessed;
        try {
            input.transaction(guessInputConfig, new InputPlugin.Control() {

                public List<TaskReport> run(TaskSource inputTaskSource, Schema schema, int taskCount) {
                    if (taskCount == 0) {
                        throw new NoSampleException("No input files to guess");
                    }
                    input.run(inputTaskSource, null, 0, new PageOutput() {

                        @Override
                        public void add(Page page) {
                            // TODO exception class
                            throw new RuntimeException("Input plugin must be a FileInputPlugin to guess parser configuration");
                        }

                        @Override
                        public void finish() {
                        }

                        @Override
                        public void close() {
                        }
                    });
                    throw new AssertionError("Guess executor must throw GuessedNoticeError");
                }
            });
            throw new AssertionError("Guess executor must throw GuessedNoticeError");
        } catch (GuessedNoticeError error) {
            guessed = lastGuessed.deepCopy().merge(error.getGuessedConfig());
        }
        // merge to the last-guessed config
        if (lastGuessed.equals(guessed)) {
            // not changed
            return lastGuessed;
        }
        lastGuessed = guessed;
    }
    return lastGuessed;
}

Example 3 with Page

use of org.embulk.spi.Page in project embulk by embulk.

the class CsvFormatterPlugin method open.

@Override
public PageOutput open(TaskSource taskSource, final Schema schema, FileOutput output) {
    final PluginTask task = taskSource.loadTask(PluginTask.class);
    final LineEncoder encoder = new LineEncoder(output, task);
    final TimestampFormatter[] timestampFormatters = Timestamps.newTimestampColumnFormatters(task, schema, task.getColumnOptions());
    final char delimiter = task.getDelimiterChar();
    final QuotePolicy quotePolicy = task.getQuotePolicy();
    final char quote = task.getQuoteChar() != '\0' ? task.getQuoteChar() : '"';
    final char escape = task.getEscapeChar().or(quotePolicy == QuotePolicy.NONE ? '\\' : quote);
    final String newlineInField = task.getNewlineInField().getString();
    final String nullString = task.getNullString();
    // create a file
    encoder.nextFile();
    // write header
    if (task.getHeaderLine()) {
        writeHeader(schema, encoder, delimiter, quotePolicy, quote, escape, newlineInField, nullString);
    }
    return new PageOutput() {

        private final PageReader pageReader = new PageReader(schema);

        private final String delimiterString = String.valueOf(delimiter);

        public void add(Page page) {
            pageReader.setPage(page);
            while (pageReader.nextRecord()) {
                schema.visitColumns(new ColumnVisitor() {

                    public void booleanColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            addValue(Boolean.toString(pageReader.getBoolean(column)));
                        } else {
                            addNullString();
                        }
                    }

                    public void longColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            addValue(Long.toString(pageReader.getLong(column)));
                        } else {
                            addNullString();
                        }
                    }

                    public void doubleColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            addValue(Double.toString(pageReader.getDouble(column)));
                        } else {
                            addNullString();
                        }
                    }

                    public void stringColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            addValue(pageReader.getString(column));
                        } else {
                            addNullString();
                        }
                    }

                    public void timestampColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            Timestamp value = pageReader.getTimestamp(column);
                            addValue(timestampFormatters[column.getIndex()].format(value));
                        } else {
                            addNullString();
                        }
                    }

                    public void jsonColumn(Column column) {
                        addDelimiter(column);
                        if (!pageReader.isNull(column)) {
                            Value value = pageReader.getJson(column);
                            addValue(value.toJson());
                        } else {
                            addNullString();
                        }
                    }

                    private void addDelimiter(Column column) {
                        if (column.getIndex() != 0) {
                            encoder.addText(delimiterString);
                        }
                    }

                    private void addValue(String v) {
                        encoder.addText(setEscapeAndQuoteValue(v, delimiter, quotePolicy, quote, escape, newlineInField, nullString));
                    }

                    private void addNullString() {
                        encoder.addText(nullString);
                    }
                });
                encoder.addNewLine();
            }
        }

        public void finish() {
            encoder.finish();
        }

        public void close() {
            encoder.close();
        }
    };
}

Example 4 with Page

use of org.embulk.spi.Page in project embulk by embulk.

the class SamplingParserPlugin method runFileInputSampling.

public static Buffer runFileInputSampling(final FileInputRunner runner, ConfigSource inputConfig, ConfigSource sampleBufferConfig) {
    final SampleBufferTask sampleBufferTask = sampleBufferConfig.loadConfig(SampleBufferTask.class);
    // override in.parser.type so that FileInputRunner creates SamplingParserPlugin
    ConfigSource samplingInputConfig = inputConfig.deepCopy();
    samplingInputConfig.getNestedOrSetEmpty("parser").set("type", "system_sampling").set("sample_buffer_bytes", sampleBufferTask.getSampleBufferBytes());
    samplingInputConfig.set("decoders", null);
    try {
        runner.transaction(samplingInputConfig, new InputPlugin.Control() {

            public List<TaskReport> run(TaskSource taskSource, Schema schema, int taskCount) {
                if (taskCount == 0) {
                    throw new NoSampleException("No input files to read sample data");
                }
                int maxSize = -1;
                int maxSizeTaskIndex = -1;
                for (int taskIndex = 0; taskIndex < taskCount; taskIndex++) {
                    try {
                        runner.run(taskSource, schema, taskIndex, new PageOutput() {

                            @Override
                            public void add(Page page) {
                                // TODO exception class
                                throw new RuntimeException("Input plugin must be a FileInputPlugin to guess parser configuration");
                            }

                            public void finish() {
                            }

                            public void close() {
                            }
                        });
                    } catch (NotEnoughSampleError ex) {
                        if (maxSize < ex.getSize()) {
                            maxSize = ex.getSize();
                            maxSizeTaskIndex = taskIndex;
                        }
                        continue;
                    }
                }
                if (maxSize <= 0) {
                    throw new NoSampleException("All input files are empty");
                }
                taskSource.getNested("ParserTaskSource").set("force", true);
                try {
                    runner.run(taskSource, schema, maxSizeTaskIndex, new PageOutput() {

                        @Override
                        public void add(Page page) {
                            // TODO exception class
                            throw new RuntimeException("Input plugin must be a FileInputPlugin to guess parser configuration");
                        }

                        public void finish() {
                        }

                        public void close() {
                        }
                    });
                } catch (NotEnoughSampleError ex) {
                    throw new NoSampleException("All input files are smaller than minimum sampling size");
                }
                throw new NoSampleException("All input files are smaller than minimum sampling size");
            }
        });
        throw new AssertionError("SamplingParserPlugin must throw SampledNoticeError");
    } catch (SampledNoticeError error) {
        return error.getSample();
    }
}