public SeparatedValuesTable(string fileName, string separator, bool hasHeader, int skipLines)
{
var file = new FileInfo(fileName);
using (var stream = new StreamReader(file.OpenRead()))
{
var line = string.Empty;
var currentLine = 0;
while (!stream.EndOfStream && ((line = stream.ReadLine()) == string.Empty || currentLine < skipLines))
{
currentLine += 1;
}
var columns = line.Split(new[] { separator }, StringSplitOptions.None);
if (hasHeader)
{
Columns = columns
.Select((header, i) => (ISchemaColumn) new SchemaColumn(SeparatedValuesHelper.MakeHeaderNameValidColumnName(header), i, typeof(string)))
.ToArray();
}
else
{
Columns = columns
.Select((f, i) => (ISchemaColumn) new SchemaColumn(string.Format(SeparatedValuesHelper.AutoColumnName, i + 1), i, typeof(string)))
.ToArray();
}
}
}
private void ProcessFile(SeparatedValueFile csvFile, BlockingCollection <IReadOnlyList <DataSources.IObjectResolver> > chunkedSource)
{
var file = new FileInfo(csvFile.FilePath);
if (!file.Exists)
{
chunkedSource.Add(new List <EntityResolver <object[]> >());
return;
}
var nameToIndexMap = new Dictionary <string, int>();
var indexToMethodAccess = new Dictionary <int, Func <object[], object> >();
var indexToNameMap = new Dictionary <int, string>();
var endWorkToken = _context.EndWorkToken;
using (var stream = CreateStreamFromFile(file))
{
using (var reader = new StreamReader(stream, Encoding.UTF8))
{
SkipLines(reader, csvFile);
using (var csvReader = new CsvReader(reader))
{
csvReader.Configuration.Delimiter = csvFile.Separator;
csvReader.Read();
var header = csvReader.Context.Record;
for (var i = 0; i < header.Length; ++i)
{
var headerName = csvFile.HasHeader ? SeparatedValuesHelper.MakeHeaderNameValidColumnName(header[i]) : string.Format(SeparatedValuesHelper.AutoColumnName, i + 1);
nameToIndexMap.Add(headerName, i);
indexToNameMap.Add(i, headerName);
var i1 = i;
indexToMethodAccess.Add(i, row => row[i1]);
}
}
}
}
using (var stream = CreateStreamFromFile(file))
{
using (var reader = new StreamReader(stream))
{
SkipLines(reader, csvFile);
using (var csvReader = new CsvReader(reader))
{
csvReader.Configuration.BadDataFound = context => { };
csvReader.Configuration.Delimiter = csvFile.Separator;
int i = 1, j = 11;
var list = new List <EntityResolver <object[]> >(100);
var rowsToRead = 1000;
const int rowsToReadBase = 100;
if (csvFile.HasHeader)
{
csvReader.Read(); //skip header.
}
while (csvReader.Read())
{
var rawRow = csvReader.Context.Record;
list.Add(new EntityResolver <object[]>(ParseRecords(rawRow, indexToNameMap), nameToIndexMap, indexToMethodAccess));
if (i++ < rowsToRead)
{
continue;
}
i = 1;
if (j > 1)
{
j -= 1;
}
rowsToRead = rowsToReadBase * j;
chunkedSource.Add(list, endWorkToken);
list = new List <EntityResolver <object[]> >(rowsToRead);
}
chunkedSource.Add(list, endWorkToken);
}
}
}
}
