public override BaseColumn Add(BaseColumn column, bool inPlace = false)
{
if (Length != column.Length)
{
throw new ArgumentException(Strings.MismatchedColumnLengths, nameof(column));
}
StringColumn ret = inPlace ? this : Clone();
for (long i = 0; i < Length; i++)
{
ret[i] += column[i].ToString();
}
return(ret);
}
public override BaseColumn Add(BaseColumn column)
{
// TODO: Using indexing is VERY inefficient here. Each indexer call will find the "right" buffer and then return the value
if (Length != column.Length)
{
throw new ArgumentException(Strings.MismatchedColumnLengths, nameof(column));
}
StringColumn ret = Clone();
for (long i = 0; i < Length; i++)
{
ret[i] += column[i].ToString();
}
return(ret);
}
public override BaseColumn Add <T>(T value, bool inPlace = false)
{
StringColumn ret = inPlace ? this : Clone();
string valString = value.ToString();
for (int i = 0; i < ret._stringBuffers.Count; i++)
{
IList <string> buffer = ret._stringBuffers[i];
int bufferLen = buffer.Count;
for (int j = 0; j < bufferLen; j++)
{
buffer[j] += valString;
}
}
return(ret);
}
/// <summary>
/// Reads a text file as a DataFrame.
/// Follows pandas API.
/// </summary>
/// <param name="createStream">function which creates a stream</param>
/// <param name="separator">column separator</param>
/// <param name="header">has a header or not</param>
/// <param name="columnNames">column names (can be empty)</param>
/// <param name="dataTypes">column types (can be empty)</param>
/// <param name="numberOfRowsToRead">number of rows to read not including the header(if present)</param>
/// <param name="guessRows">number of rows used to guess types</param>
/// <param name="addIndexColumn">add one column with the row index</param>
/// <returns>DataFrame</returns>
public static DataFrame ReadStream(Func <StreamReader> createStream,
char separator = ',', bool header = true,
string[] columnNames = null, Type[] dataTypes = null,
long numberOfRowsToRead = -1, int guessRows = 10, bool addIndexColumn = false)
{
var linesForGuessType = new List <string[]>();
long rowline = 0;
int numberOfColumns = 0;
if (header == true && numberOfRowsToRead != -1)
{
numberOfRowsToRead++;
}
// First pass: schema and number of rows.
using (var st = createStream())
{
string line = st.ReadLine();
while (line != null)
{
if ((numberOfRowsToRead == -1) || rowline < numberOfRowsToRead)
{
if (linesForGuessType.Count < guessRows)
{
var spl = line.Split(separator);
if (header && rowline == 0)
{
if (columnNames == null)
{
columnNames = spl;
}
}
else
{
linesForGuessType.Add(spl);
numberOfColumns = Math.Max(numberOfColumns, spl.Length);
}
}
}
++rowline;
if (rowline == numberOfRowsToRead)
{
break;
}
line = st.ReadLine();
}
}
if (linesForGuessType.Count == 0)
{
throw new FormatException(Strings.EmptyFile);
}
List <BaseColumn> columns = new List <BaseColumn>(numberOfColumns);
// Guesses types and adds columns.
for (int i = 0; i < numberOfColumns; ++i)
{
Type kind = GuessKind(i, linesForGuessType);
if (kind == typeof(bool))
{
BaseColumn boolColumn = new PrimitiveColumn <bool>(columnNames == null ? "Column" + i.ToString() : columnNames[i], header == true ? rowline - 1 : rowline);
columns.Add(boolColumn);
}
else if (kind == typeof(float))
{
BaseColumn floatColumn = new PrimitiveColumn <float>(columnNames == null ? "Column" + i.ToString() : columnNames[i], header == true ? rowline - 1 : rowline);
columns.Add(floatColumn);
}
else if (kind == typeof(string))
{
BaseColumn stringColumn = new StringColumn(columnNames == null ? "Column" + i.ToString() : columnNames[i], header == true ? rowline - 1 : rowline);
columns.Add(stringColumn);
}
else
{
throw new NotSupportedException(nameof(kind));
}
}
// Fills values.
using (StreamReader st = createStream())
{
string line = st.ReadLine();
rowline = 0;
while (line != null && (numberOfRowsToRead == -1 || rowline < numberOfRowsToRead))
{
var spl = line.Split(separator);
if (header && rowline == 0)
{
// Skips.
}
else
{
AppendRow(columns, header == true ? rowline - 1 : rowline, spl);
}
++rowline;
line = st.ReadLine();
}
}
if (addIndexColumn)
{
PrimitiveColumn <int> indexColumn = new PrimitiveColumn <int>("IndexColumn", columns[0].Length);
for (int i = 0; i < columns[0].Length; i++)
{
indexColumn[i] = i;
}
columns.Insert(0, indexColumn);
}
return(new DataFrame(columns));
}
