public void Split(ESplitType splitType, int splitSize,string fileSuffixXpath)
{
_data.Read();
//Capture all of the items before the first element
while (_data.NodeType != XmlNodeType.Element)
{
_output.WriteNode(_data, true);
}
//Prepare the document header and footer sections for use later...
_output.WriteStartElement(_data.Name);
_output.WriteAttributes(_data, true);
var header = _outputBuilder.ToString() + ">";
var footer = "</" + _data.Name + ">";
//skip past the first node
_data.Read();
int elementCounter = 0;
int splitCount = 0;
bool emptyDocument = false;
FileInfo dataFI = new FileInfo(_dataPath);
string outputFileName = dataFI.Name.Substring(0,dataFI.Name.LastIndexOf("."));
StreamWriter writer = null;
while (!_data.EOF)
{
//Only count the nodes that interest us...
if (!IgnorableNodeType(_data.NodeType))
elementCounter++;
else
emptyDocument = false;
//copy everything from the reader
_output.WriteNode(_data, true);
if ((splitType == ESplitType.ElementCount && elementCounter >= splitSize) || (splitType == ESplitType.Filesize && (_outputBuilder.Length - header.Length) >= splitSize*1000))
{
////construct the final XML string
ExportXmlFile(footer, writer, outputFileName, splitCount);
//reset counters
splitCount++;
elementCounter = 0;
//reset string builder
_outputBuilder.Length = 0;
_outputBuilder.Append(header);
emptyDocument = true;
}
}
//If there is anything left, export it - no footer required!
if(_outputBuilder.Length > 0)
ExportXmlFile("", writer, outputFileName, splitCount);
SplitTotal = splitCount;
}
private static void GetSplitConfig()
{
var config = new System.Xml.XmlDocument();
config.Load(_configPath);
var split = config.SelectSingleNode("/config/split");
if (split == null)
{
throw new Exception("No <split> element defined in config.");
}
if (split.Attributes["type"] == null || split.Attributes["size"] == null)
{
throw new Exception("type, size, directoryname attributes were not defined for <split> element in config");
}
//get config
switch (split.Attributes["type"].Value.ToLower())
{
case "elementcount":
_splitType = ESplitType.ElementCount;
break;
case "filesize":
_splitType = ESplitType.Filesize;
break;
default:
throw new Exception("Unrecognized split type --> must be elementcount or filesize");
}
if (!int.TryParse(split.Attributes["size"].Value, out _splitSize))
{
throw new Exception("size attribute not integer value");
}
//now all config parameters obtained need to:
//1) create a folder to place all of the chunks of origial data
//2) find the folder for the output and name appropriately for each one
if (split.Attributes["inputdirectoryname"] != null)
{
_inputDirectoryName = split.Attributes["inputdirectoryname"].Value;
}
if (split.Attributes["outputdirectoryname"] != null)
{
_outputDirectoryName = split.Attributes["outputdirectoryname"].Value;
}
if (split.Attributes["filesuffix"] != null)
{
_fileSuffixXpath = split.Attributes["filesuffix"].Value;
}
//input chunks folder - get the folder of the _dataPath
FileInfo _dataFileInfo = new FileInfo(_dataPath);
//_inputChunkingFolder = _dataFileInfo.Directory.FullName + "\\" + _inputDirectoryName;
_inputChunkingFolder = Path.Combine(_dataFileInfo.Directory.FullName, _inputDirectoryName);
Directory.CreateDirectory(_inputChunkingFolder);
//rdf output chunks folder - get the folder of the _outputPath
_outputChunkingFolder = Path.Combine(_outputPath, _outputDirectoryName);
Directory.CreateDirectory(_outputChunkingFolder);
}
private static void GetSplitConfig()
{
var config = new System.Xml.XmlDocument();
config.Load(_configPath);
var split = config.SelectSingleNode("/config/split");
if (split == null)
throw new Exception("No <split> element defined in config.");
if (split.Attributes["type"] == null || split.Attributes["size"] == null)
throw new Exception("type, size, directoryname attributes were not defined for <split> element in config");
//get config
switch (split.Attributes["type"].Value.ToLower())
{
case "elementcount":
_splitType = ESplitType.ElementCount;
break;
case "filesize":
_splitType = ESplitType.Filesize;
break;
default:
throw new Exception("Unrecognized split type --> must be elementcount or filesize");
}
if (!int.TryParse(split.Attributes["size"].Value, out _splitSize))
throw new Exception("size attribute not integer value");
//now all config parameters obtained need to:
//1) create a folder to place all of the chunks of origial data
//2) find the folder for the output and name appropriately for each one
if (split.Attributes["inputdirectoryname"] != null)
_inputDirectoryName = split.Attributes["inputdirectoryname"].Value;
if (split.Attributes["outputdirectoryname"] != null)
_outputDirectoryName = split.Attributes["outputdirectoryname"].Value;
if (split.Attributes["filesuffix"] != null)
_fileSuffixXpath = split.Attributes["filesuffix"].Value;
//input chunks folder - get the folder of the _dataPath
FileInfo _dataFileInfo = new FileInfo(_dataPath);
//_inputChunkingFolder = _dataFileInfo.Directory.FullName + "\\" + _inputDirectoryName;
_inputChunkingFolder = Path.Combine(_dataFileInfo.Directory.FullName, _inputDirectoryName);
Directory.CreateDirectory(_inputChunkingFolder);
//rdf output chunks folder - get the folder of the _outputPath
_outputChunkingFolder = Path.Combine(_outputPath, _outputDirectoryName);
Directory.CreateDirectory(_outputChunkingFolder);
}
public void Split(ESplitType splitType, int splitSize, string fileSuffixXpath)
{
_data.Read();
//Capture all of the items before the first element
while (_data.NodeType != XmlNodeType.Element)
{
_output.WriteNode(_data, true);
}
//Prepare the document header and footer sections for use later...
_output.WriteStartElement(_data.Name);
_output.WriteAttributes(_data, true);
var header = _outputBuilder.ToString() + ">";
var footer = "</" + _data.Name + ">";
//skip past the first node
_data.Read();
int elementCounter = 0;
int splitCount = 0;
bool emptyDocument = false;
FileInfo dataFI = new FileInfo(_dataPath);
string outputFileName = dataFI.Name.Substring(0, dataFI.Name.LastIndexOf("."));
StreamWriter writer = null;
while (!_data.EOF)
{
//Only count the nodes that interest us...
if (!IgnorableNodeType(_data.NodeType))
{
elementCounter++;
}
else
{
emptyDocument = false;
}
//copy everything from the reader
_output.WriteNode(_data, true);
if ((splitType == ESplitType.ElementCount && elementCounter >= splitSize) || (splitType == ESplitType.Filesize && (_outputBuilder.Length - header.Length) >= splitSize * 1000))
{
////construct the final XML string
ExportXmlFile(footer, writer, outputFileName, splitCount);
//reset counters
splitCount++;
elementCounter = 0;
//reset string builder
_outputBuilder.Length = 0;
_outputBuilder.Append(header);
emptyDocument = true;
}
}
//If there is anything left, export it - no footer required!
if (_outputBuilder.Length > 0)
{
ExportXmlFile("", writer, outputFileName, splitCount);
}
SplitTotal = splitCount;
}
