net.optionfactory.csv.parser.DefaultCsvParser Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of cassava Show documentation
a csv parser
The newest version!
package net.optionfactory.csv.parser;

import java.io.Reader;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.concurrent.ConcurrentHashMap;
import net.optionfactory.csv.parser.CsvParserSpi.Record;
import net.optionfactory.csv.parser.javacc.ParseException;

public class DefaultCsvParser implements CsvParser {

    @Override
    public CsvParsingResult parse(Separator format, Trimming trimming, RecordsTrimming recordsTrimming, String sourceName, Reader reader) {
        final CsvParserSpi parser = format.parser(reader);
        final CsvParsingResult result = CsvParsingResult.empty();
        try {
            final Record record = parser.record(trimming);
            if (record == null) {
                result.problems.add(CsvParsingProblems.noSchema(sourceName));
                return result;
            }
            result.schema = record.values;
            final CsvParsingResult dataResult = parse(format, trimming, recordsTrimming, sourceName, parser, result.schema);
            result.problems = dataResult.problems;
            result.records = dataResult.records;
            return result;
        } catch (ParseException ex) {
            final Source s = Source.of(ex.currentToken.beginLine, ex.currentToken.beginColumn, ex.currentToken.endLine, ex.currentToken.endColumn);
            final List> expectedOneOf = new ArrayList<>();
            for (int[] expectedTokenSequence : ex.expectedTokenSequences) {
                final List expected = new ArrayList<>();
                for (int e : expectedTokenSequence) {
                    expected.add(format.tokenImage(e));
                }
                expectedOneOf.add(expected);
            }
            result.problems.add(CsvParsingProblems.unparseableSchema(sourceName, s, ex.currentToken.image, expectedOneOf));
            return result;
        }
    }

    @Override
    public CsvParsingResult parse(Separator f, Trimming trimming, RecordsTrimming recordsTrimming, String sourceName, Reader r, List externalSchema) {
        return parse(f, trimming, recordsTrimming, sourceName, f.parser(r), externalSchema);
    }

    private static CsvParsingResult parse(Separator format, Trimming trimming, RecordsTrimming recordsTrimming, String sourceName, CsvParserSpi parser, List externalSchema) {
        final CsvParsingResult result = CsvParsingResult.withSchema(externalSchema);
        for (;;) {
            try {
                final Record recordOrNull = parser.record(trimming);
                if (recordOrNull == null) {
                    while (RecordsTrimming.EmptyLinesAtEnd == recordsTrimming && !result.records.isEmpty() && result.records.get(result.records.size() - 1).values().stream().allMatch(f -> f.isEmpty())) {
                        result.records.remove(result.records.size() - 1);
                    }
                    return result;
                }
                if (recordOrNull.values.size() != externalSchema.size()) {
                    result.problems.add(CsvParsingProblems.schemaViolation(sourceName, recordOrNull.source, recordOrNull.values.toString(), externalSchema));
                    continue;
                }
                final Map values = new ConcurrentHashMap<>();
                for (int i = 0; i != recordOrNull.values.size(); i++) {
                    final String fieldName = externalSchema.get(i);
                    final String fieldValue = recordOrNull.values.get(i);
                    values.put(fieldName, fieldValue);
                }
                result.records.add(values);
            } catch (ParseException ex) {
                final Source s = Source.of(ex.currentToken.beginLine, ex.currentToken.beginColumn, ex.currentToken.endLine, ex.currentToken.endColumn);
                final List> expectedOneOf = new ArrayList<>();
                for (int[] expectedTokenSequence : ex.expectedTokenSequences) {
                    final List expected = new ArrayList<>();
                    for (int e : expectedTokenSequence) {
                        expected.add(format.tokenImage(e));
                    }
                    expectedOneOf.add(expected);
                }
                result.problems.add(CsvParsingProblems.unparseableRecord(sourceName, s, ex.currentToken.image, expectedOneOf));
                try {
                    parser.skipToNextRecord();
                } catch (ParseException neverHappens) {
                }
            }
        }
    }
}