Class CSVLoader

java.lang.Object
com.sparsity.sparksee.io.CSVLoader
public class CSVLoader extends Object

  • Method Details

    • loadEdges

      public static void loadEdges(Graph graph, String fileName, String edgeType, String tailNodeType, String headNodeType, int tail, int head, String separator, boolean directed, boolean header, MissingEndpoint onMissingTail, MissingEndpoint onMissingHead, int[] columns, String[] attrNames, DataType[] dataTypes, AttributeKind[] attrKinds)
      Loads edges from a CSV file.

      nodeType[in] The type of the edge to load. If it does not exist, it creates it

      Parameters:
      graph - [in] The graph to load the edges into
      fileName - [in] The name of the file
      edgeType - null
      tailNodeType - [in] The type of the tail nodes. If it does not exist and onMissingTail is set to "Create", it creates it. Otherwise, an exception is thrown
      headNodeType - [in] The type of the head nodes. It it does not exist and onMissingHead is set to "Create", it creates it. Otherwise, an exception is thrown
      tail - [in] The tail column index. Default: 0
      head - [in] The head column index. Default: 1
      separator - [in] The column separator. Default: ","
      directed - [in] True if this edge is directed or not. False otherwise. Default: True
      header - [in] True if the CSV contains a header. False otherwise. Default: True
      onMissingTail - [in] The policy to follow when a tail is missing. Default: "IsError"
      onMissingHead - [in] The policy to follow when a head is missing. Default: "IsError"
      columns - [in] The list of columns to load. tail and head columns must be in this list if this list is specified. If the list is empty, all columns are loaded. Default: empty
      attrNames - [in] The attribute names. If this list is empty and hasHeader is set to true, the header values are used as attribute names. Default: empty
      dataTypes - [in] The dataTypes of the attributes if this do not already exist. If this list is empty, the method tries to infer the data type from the first non-header row. Default: empty
      attrKinds - [in] The attributeKinds of the attributes if these do not already exist. If this list is empty, the attributeKind of the created attributes are set to Basic. Default: empty

    • loadNodes

      public static void loadNodes(Graph graph, String fileName, String nodeType, String separator, boolean header, int[] columns, String[] attrNames, DataType[] dataTypes, AttributeKind[] attrKinds)
      Loads nodes from a CSV file.

      Parameters:
      graph - [in] The graph to load the nodes into
      fileName - [in] The name of the file
      nodeType - [in] The type of the node to load. If it does not exist, it creates it
      separator - [in] The separator of the CSV file. Default: ","
      header - [in] True if the CSV contains a header. False otherwise. Default: True
      columns - [in] The list of columns to load. tail and head columns mus be in this list if this list is specified. Default: all columns
      attrNames - [in] The attribute names. If this list is empty and hasHeader is set to true, the header values are used as attribute names. Default: empty
      dataTypes - [in] The dataTypes of the attributes if this do not already exist. Default: empty If this list is empty, the method tries to infer the data type from the first non-header row
      attrKinds - [in] The attributeKinds of the attributes if these do not already exist. Default: empty If this list is empty, the attributeKind of the created attributes are set to Basic. Default: empty