DbfDataReader

You can use CdxFile to open .cdx files and search for specific keys. The search results provide RecordIndex values (one-based) which can be converted to zero-based indices for use with DbfTable.Seek or DbfTable.Read.

using DbfDataReader.Cdx;

var dbfPath = "path/file.dbf";
var cdxPath = "path/file.cdx";

using (var dbfTable = new DbfTable(dbfPath))
using (var cdxFile = new CdxFile(cdxPath, dbfTable.CurrentEncoding))
{
    var tagNames = cdxFile.TagNames;             // the named indexes ("tags") in the file

    var index = cdxFile.GetIndex("CONTACT_ID");  // one tag; index.KeyExpression describes the key
    var dbfRecord = new DbfRecord(dbfTable);

    foreach (var entry in index.Search("C0000000042"))
    {
        dbfTable.Seek(entry.RecordIndex);
        dbfTable.Read(dbfRecord);
        // dbfRecord now holds the matching row
    }
}

Install the DbfDataReader NuGet package using the dotnet CLI. The library targets net10.0 and netstandard2.1.

dotnet add package DbfDataReader

To benchmark the performance of automatic index usage (comparing index seeks against full scans for various query types like character seeks, equality seeks, and range scans), run the following command. This uses a generated 50,000-row table with a compound index on ID and CODE.

dotnet run -c Release --project test/DbfDataReader.Benchmarks -- --filter "*IndexQueryBenchmarks*"

To compare DbfDataReader performance against other libraries (like Sylvan, NDbf, or DbfData) for reading every field of every record, use the following command.

Note: You must set the version via the DbfDataReaderVersion environment variable. Using the -p:DbfDataReaderVersion MSBuild property will not work because it does not reach the BenchmarkDotNet generated child build. You can verify the version being measured by checking the // DbfDataReader under benchmark: line in the benchmark log.

DbfDataReaderVersion=1.1.0 dotnet run -c Release --project test/DbfDataReader.Benchmarks -- --filter "*DbfDataReaderBenchmarks*" --job short

You can treat a folder of DBF files as a database by using DbfDbConnection. This allows you to execute SQL commands against the files within a specified directory. The connection string must include the Folder option.

To use it, set the ConnectionString with the Folder path, call .Open(), and then create a command to execute SQL text.

var dbConnection = new DbfDbConnection(string.Empty, string.Empty);
dbConnection.ConnectionString = $"Folder=./test/fixtures;SkipDeletedRecords=false";
dbConnection.Open();

var dbCommand = dbConnection.CreateCommand();
dbCommand.CommandText = "select * from dbase_03.dbf;";

var reader = await dbCommand.ExecuteReaderAsync();
while (await reader.ReadAsync())
{
    var valueCol1 = reader.GetString(0);
    var valueCol11 = reader.GetDecimal(10);
}

To execute a query that returns a single value (like a count or a specific column value), cast the created command to DbfDbCommand and use ExecuteScalar().

var command = (DbfDbCommand)dbConnection.CreateCommand();
command.CommandText = "select Point_ID from dbase_03.dbf where Point_ID = @id";
command.Parameters.AddWithValue("@id", "A1");
var value = command.ExecuteScalar();

Since DbfDataReader implements DbDataReader, you can pass it directly to SqlBulkCopy.WriteToServer to perform high-performance imports into MS SQL Server.

var options = new DbfDataReaderOptions
{
    SkipDeletedRecords = true
};

var dbfPath = "path/file.dbf";
using (var dbfDataReader = new DbfDataReader(dbfPath, options))
{
    using (var bulkCopy = new SqlBulkCopy(connection))
    {
        bulkCopy.DestinationTableName = "DestinationTableName";

        try
        {
            bulkCopy.WriteToServer(dbfDataReader);
        }
        catch (Exception ex)
        {
            Console.WriteLine($"Error importing: dbf file: '{dbfPath}', exception: {ex.Message}");
        }
    }
}

To iterate through rows manually, use DbfTable.Read(DbfRecord). You can check DbfRecord.IsDeleted to decide whether to skip records marked as deleted.

var skipDeleted = true;

var dbfPath = "path/file.dbf";
using (var dbfTable = new DbfTable(dbfPath, Encoding.UTF8))
{
    var dbfRecord = new DbfRecord(dbfTable);

    while (dbfTable.Read(dbfRecord))
    {
        if (skipDeleted && dbfRecord.IsDeleted)
        {
            continue;
        }

        foreach (var dbfValue in dbfRecord.Values)
        {
            var stringValue = dbfValue.ToString();
            var obj = dbfValue.GetValue();
        }
    }
}

The DbfDataReader class implements DbDataReader, allowing you to use standard ADO.NET methods like GetString, GetDecimal, GetDateTime, and GetInt32. You can configure it using DbfDataReaderOptions (e.g., SkipDeletedRecords).

var options = new DbfDataReaderOptions
{
    SkipDeletedRecords = true
};

var dbfPath = "path/file.dbf";
using (var dbfDataReader = new DbfDataReader(dbfPath, options))
{
    while (dbfDataReader.Read())
    {
        var valueCol1 = dbfDataReader.GetString(0);
        var valueCol2 = dbfDataReader.GetDecimal(1);
        var valueCol3 = dbfDataReader.GetDateTime(2);
        var valueCol4 = dbfDataReader.GetInt32(3);
    }
}

Use DbfTable to open a .dbf file and inspect its metadata, such as the version description, record count, and column definitions (name, type, length, and decimal count).

var dbfPath = "path/file.dbf";
using (var dbfTable = new DbfTable(dbfPath, Encoding.UTF8))
{
    var header = dbfTable.Header;

    var versionDescription = header.VersionDescription;
    var hasMemo = dbfTable.Memo != null;
    var recordCount = header.RecordCount;

    foreach (var dbfColumn in dbfTable.Columns)
    {
        var name = dbfColumn.ColumnName;
        var columnType = dbfColumn.ColumnType;
        var length = dbfColumn.Length;
        var decimalCount = dbfColumn.DecimalCount;
    }
}