/// <summary>
///
/// </summary>
/// <param name="fileAndContent"></param>
/// <param name="maxClusterSize"></param>
/// <returns></returns>
public static TextFileClassifier GetInstance(
IDictionary <string, string> fileAndContent,
int maxClusterSize = -1)
{
Queue <Task <Point> > tasks = new Queue <Task <Point> >();
foreach (KeyValuePair <string, string> kvp in fileAndContent)
{
tasks.Enqueue(
new Task <Point>(
() =>
{
return(new Text(kvp.Value, kvp.Key));
}
)
);
}
QueueBasedTaskRunner <Point> runner = new QueueBasedTaskRunner <Point>(tasks);
runner.RunParallel();
TextFileClassifier TheInstance = new TextFileClassifier();
TheInstance.ClusterIdentifier = new KMinSpanningTree(runner.GetResult().ToArray(), maxClusterSize);
return(TheInstance);
}
/// <summary>
/// Get the instance prividing the directory containing the text files you want to classfiy
/// </summary>
/// <param name="directory">
/// A string that points to the directory of targeted files.
/// </param>
/// <param name="maxClusterSize">
/// The maximum size you of the clusters you want to set.
/// </param>
/// <returns>
/// An instacne of TextFileClassifier.
/// </returns>
public static TextFileClassifier GetInstance(string directory, int maxClusterSize = -1)
{
DirectoryInfo dinfo = new DirectoryInfo(directory);
if (!dinfo.Exists)
{
throw new ArgumentException("Directory doesn't exist.");
}
IDictionary <string, string> FilesAndContent = Basic.GetContentForAllFiles(
directory,
SettingsManager.FileSearchRecursive,
"txt");
TextFileClassifier TheInstance = GetInstance(FilesAndContent, maxClusterSize);
TheInstance.Directory = directory;
return(TheInstance);
}