Java TypeInfoUtils.getTypeInfosFromTypeString方法代码示例

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
private StructTypeInfo getAllReadTypeInfo( final String columnNameProperty , final String columnTypeProperty ){
  ArrayList<TypeInfo> fieldTypes = TypeInfoUtils.getTypeInfosFromTypeString( columnTypeProperty );
  ArrayList<String> columnNames = new ArrayList<String>();
  if ( columnNameProperty != null && 0 < columnNameProperty.length() ) {
    String[] columnNameArray = columnNameProperty.split(",");
    for( int i = 0 ; i < columnNameArray.length ; i++ ){
      columnNames.add( columnNameArray[i] );
      filedIndexMap.put( columnNameArray[i] , i );
    }
  }
  StructTypeInfo rootType = new StructTypeInfo();

  rootType.setAllStructFieldNames( columnNames );
  rootType.setAllStructFieldTypeInfos( fieldTypes );

  return rootType;
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
private StructTypeInfo getColumnProjectionTypeInfo( final String columnNameProperty , final String columnTypeProperty , final String projectionColumnNames ){
  Set<String> columnNameSet = new HashSet<String>();
  for( String columnName : projectionColumnNames.split(",") ){
    columnNameSet.add( columnName );
  }

  ArrayList<TypeInfo> fieldTypes = TypeInfoUtils.getTypeInfosFromTypeString( columnTypeProperty );
  String[] splitNames = columnNameProperty.split(",");

  ArrayList<String> projectionColumnNameList = new ArrayList<String>();
  ArrayList<TypeInfo> projectionFieldTypeList = new ArrayList<TypeInfo>();
  for( int i = 0 ; i < fieldTypes.size() ; i++ ){
    if( columnNameSet.contains( splitNames[i] ) ){
      projectionColumnNameList.add( splitNames[i] );
      projectionFieldTypeList.add( fieldTypes.get(i) );
    }
    filedIndexMap.put( splitNames[i] , i );
  }
  StructTypeInfo rootType = new StructTypeInfo();

  rootType.setAllStructFieldNames( projectionColumnNameList );
  rootType.setAllStructFieldTypeInfos( projectionFieldTypeList );

  return rootType;
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
static StandardStructObjectInspector structObjectInspector(Properties tableProperties) {
    // extract column info - don't use Hive constants as they were renamed in 0.9 breaking compatibility
    // the column names are saved as the given inspector to #serialize doesn't preserves them (maybe because it's an external table)
    // use the class since StructType requires it ...
    List<String> columnNames = StringUtils.tokenize(tableProperties.getProperty(HiveConstants.COLUMNS), ",");
    List<TypeInfo> colTypes = TypeInfoUtils.getTypeInfosFromTypeString(tableProperties.getProperty(HiveConstants.COLUMNS_TYPES));

    // create a standard writable Object Inspector - used later on by serialization/deserialization
    List<ObjectInspector> inspectors = new ArrayList<ObjectInspector>();

    for (TypeInfo typeInfo : colTypes) {
        inspectors.add(TypeInfoUtils.getStandardWritableObjectInspectorFromTypeInfo(typeInfo));
    }

    return ObjectInspectorFactory.getStandardStructObjectInspector(columnNames, inspectors);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
/**
 * An initialization function used to gather information about the table.
 * Typically, a SerDe implementation will be interested in fgthe list of
 * column names and their types. That information will be used to help perform
 * actual serialization and deserialization of data.
 */
@Override
public void initialize(Configuration conf, Properties tbl)
        throws SerDeException {
    // Get a list of the table's column names.
    String colNamesStr = tbl.getProperty(serdeConstants.LIST_COLUMNS);
    colNames = Arrays.asList(colNamesStr.split(","));

    // Get a list of TypeInfos for the columns. This list lines up with
    // the list of column names.
    String colTypesStr = tbl.getProperty(serdeConstants.LIST_COLUMN_TYPES);
    List<TypeInfo> colTypes =
            TypeInfoUtils.getTypeInfosFromTypeString(colTypesStr);

    rowTypeInfo =
            (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(colNames, colTypes);
    rowOI =
            TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(rowTypeInfo);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
/**
 * An initialization function used to gather information about the table.
 * Typically, a SerDe implementation will be interested in the list of
 * column names and their types. That information will be used to help perform
 * actual serialization and deserialization of data.
 */
@Override
public void initialize(Configuration conf, Properties tbl)
		throws SerDeException {
	// Get a list of the table's column names.
	String colNamesStr = tbl.getProperty(serdeConstants.LIST_COLUMNS);
	colNames = Arrays.asList(colNamesStr.split(","));

	// Get a list of TypeInfos for the columns. This list lines up with
	// the list of column names.
	String colTypesStr = tbl.getProperty(serdeConstants.LIST_COLUMN_TYPES);
	List<TypeInfo> colTypes =
			TypeInfoUtils.getTypeInfosFromTypeString(colTypesStr);

	rowTypeInfo =
			(StructTypeInfo) TypeInfoFactory.getStructTypeInfo(colNames, colTypes);
	rowOI =
			TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(rowTypeInfo);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
@Override
public void initialize(final Configuration conf, final Properties tbl)
		throws SerDeException {
	log.debug("conf="+conf);
	log.debug("tblProperties="+tbl);
	final String facetType = tbl.getProperty(ConfigurationUtil.SOLR_FACET_MAPPING);
	final String columnString = tbl.getProperty(ConfigurationUtil.SOLR_COLUMN_MAPPING);
	if (StringUtils.isBlank(facetType)) {
		if (StringUtils.isBlank(columnString)) {
			throw new SerDeException("No facet mapping found, using "+ ConfigurationUtil.SOLR_COLUMN_MAPPING);
		}
		final String[] columnNamesArray = ConfigurationUtil.getAllColumns(columnString);
		colNames = Arrays.asList(columnNamesArray);
		log.debug(ConfigurationUtil.SOLR_COLUMN_MAPPING+" = " + colNames);
		row = new ArrayList<Object>(columnNamesArray.length);
	} else {
		row = new ArrayList<Object>(2);
		colNames = Arrays.asList(StringUtils.split(tbl.getProperty(Constants.LIST_COLUMNS),","));
	}
	
	colTypes = TypeInfoUtils.getTypeInfosFromTypeString(tbl.getProperty(Constants.LIST_COLUMN_TYPES));
	rowTypeInfo = (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(colNames, colTypes);
	rowOI = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(rowTypeInfo);
	log.debug("colNames="+colNames+" rowIO="+rowOI);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
@Override
public void initialize(Configuration conf, Properties tbl) throws SerDeException {
  String columnNameProperty = tbl.getProperty(LIST_COLUMNS);
  String columnTypeProperty = tbl.getProperty(LIST_COLUMN_TYPES);
  List<String> columnNames = Arrays.asList(columnNameProperty.split(","));
  List<TypeInfo> columnTypes = TypeInfoUtils.getTypeInfosFromTypeString(columnTypeProperty);

  List<ObjectInspector> columnObjectInspectors = new ArrayList<ObjectInspector>(columnNames.size());
  ObjectInspector colObjectInspector;
  for (int col = 0; col < columnNames.size(); col++) {
    colObjectInspector = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(columnTypes.get(col));
    columnObjectInspectors.add(colObjectInspector);
  }

  cachedObjectInspector = ObjectInspectorFactory
    .getColumnarStructObjectInspector(columnNames, columnObjectInspectors);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
/**
 * An initialization function used to gather information about the table.
 * Typically, a SerDe implementation will be interested in the list of
 * column names and their types. That information will be used to help
 * perform actual serialization and deserialization of data.
 */
@Override
public void initialize(final Configuration conf, final Properties tbl)
		throws SerDeException {
	// Get a list of the table's column names.
	final String colNamesStr = tbl.getProperty(serdeConstants.LIST_COLUMNS);
	// Jai...change column names to lower case.
	colNames = Arrays.asList(colNamesStr.toLowerCase().split(","));
	// Get a list of TypeInfos for the columns. This list lines up with
	// the list of column names.
	final String colTypesStr = tbl
			.getProperty(serdeConstants.LIST_COLUMN_TYPES);
	final List<TypeInfo> colTypes = TypeInfoUtils
			.getTypeInfosFromTypeString(colTypesStr);
	rowTypeInfo = (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(
			colNames, colTypes);
	rowOI = TypeInfoUtils
			.getStandardJavaObjectInspectorFromTypeInfo(rowTypeInfo);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
public void initialize(String columnNameProperty, String columnTypeProperty) {
  List<String> columnNames = Arrays.asList(columnNameProperty.split(","));
  List<TypeInfo> columnTypes = TypeInfoUtils.getTypeInfosFromTypeString(columnTypeProperty);
  StructTypeInfo rowTypeInfo =
      (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(columnNames, columnTypes);
  rowOI = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(rowTypeInfo);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
private ObjectInspector createLazyPhoenixInspector(Configuration conf, Properties tbl) throws SerDeException {
	List<String> columnNameList = Arrays.asList(tbl.getProperty(serdeConstants.LIST_COLUMNS).split(PhoenixStorageHandlerConstants.COMMA));
	List<TypeInfo> columnTypeList = TypeInfoUtils.getTypeInfosFromTypeString(tbl.getProperty(serdeConstants.LIST_COLUMN_TYPES));
	
	List<ObjectInspector> columnObjectInspectors = Lists.newArrayListWithExpectedSize(columnTypeList.size());
	
	for (TypeInfo typeInfo : columnTypeList) {
		columnObjectInspectors.add(PhoenixObjectInspectorFactory.createObjectInspector(typeInfo, serdeParams));
       }

	return LazyObjectInspectorFactory.getLazySimpleStructObjectInspector(columnNameList,
			columnObjectInspectors, null, serdeParams.getSeparators()[0], serdeParams, ObjectInspectorOptions.JAVA);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
@Override
public void initialize(Configuration conf, Properties tbl) throws SerDeException {
    String columnNameProperty = tbl.getProperty(IOConstants.COLUMNS);
    String columnTypeProperty = tbl.getProperty(IOConstants.COLUMNS_TYPES);

    if (Strings.isEmpty(columnNameProperty)) {
        columnNames = new ArrayList<String>();
    } else {
        columnNames = Arrays.asList(columnNameProperty.split(","));
    }
    if (Strings.isEmpty(columnTypeProperty)) {
        columnTypes = TypeInfoUtils.getTypeInfosFromTypeString(StringUtils.repeat("string", ":", columnNames.size()));
    } else {
        columnTypes = TypeInfoUtils.getTypeInfosFromTypeString(columnTypeProperty);
    }
    if (columnNames.size() != columnTypes.size()) {
        throw new IllegalArgumentException("IndexRHiveSerde initialization failed. Number of column " +
                "name and column type differs. columnNames = " + columnNames + ", columnTypes = " +
                columnTypes);
    }

    TypeInfo rowTypeInfo = TypeInfoFactory.getStructTypeInfo(columnNames, columnTypes);
    this.objInspector = new ArrayWritableObjectInspector((StructTypeInfo) rowTypeInfo);

    stats = new SerDeStats();
    serdeSize = 0;
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
@Override
public void initialize(Configuration conf, Properties tblProperties) throws SerDeException {
  colNames = Arrays.asList(tblProperties.getProperty(Constants.LIST_COLUMNS).split(","));
  colTypes = TypeInfoUtils.getTypeInfosFromTypeString(tblProperties.getProperty(Constants.LIST_COLUMN_TYPES));
  typeInfo = (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(colNames, colTypes);
  inspector = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(typeInfo);
  row = new ArrayList<>();
  enableFieldMapping = Boolean.valueOf(tblProperties.getProperty(ENABLE_FIELD_MAPPING, "false"));
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
@Override
public void initialize(Configuration conf, Properties tblProperties) throws SerDeException {

  colNames = Arrays.asList(tblProperties.getProperty(Constants.LIST_COLUMNS).split(","));
  colTypes = TypeInfoUtils
      .getTypeInfosFromTypeString(tblProperties.getProperty(Constants.LIST_COLUMN_TYPES));
  typeInfo = (StructTypeInfo) TypeInfoFactory.getStructTypeInfo(colNames, colTypes);
  inspector = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(typeInfo);
  row = new ArrayList<>();
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
/**
 * Initializes the RecordServiceSerde based on the table schema.
 */
@Override
public final void initialize(final Configuration conf, final Properties tbl)
    throws SerDeException {
  final List<TypeInfo> columnTypes;
  final String columnNameProperty = tbl.getProperty(IOConstants.COLUMNS);
  final String columnTypeProperty = tbl.getProperty(IOConstants.COLUMNS_TYPES);

  if (columnNameProperty.length() == 0) {
    columnNames_ = new ArrayList<String>();
  } else {
    columnNames_ = Arrays.asList(columnNameProperty.split(","));
  }

  if (columnTypeProperty.length() == 0) {
    columnTypes = new ArrayList<TypeInfo>();
  } else {
    columnTypes = TypeInfoUtils.getTypeInfosFromTypeString(columnTypeProperty);
  }

  if (columnNames_.size() != columnTypes.size()) {
    throw new IllegalArgumentException("Initialization failed. Number of column " +
      "names and column types differs. columnNames = " + columnNames_ +
      ", columnTypes = " + columnTypes);
  }

  // Initialize the ObjectInspector based on the column type information in the table.
  final TypeInfo rowTypeInfo =
      TypeInfoFactory.getStructTypeInfo(columnNames_, columnTypes);
  objInspector_ = new RecordServiceObjectInspector((StructTypeInfo) rowTypeInfo);
}
 

import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; //导入方法依赖的package包/类
public SolrReader(JobConf conf, SolrSplit split) throws IOException {
	log.debug("jobConf=" + conf);

	List<Integer> readColIDs = getReadColumnIDs(conf);
	facetMapping = conf.get(ConfigurationUtil.SOLR_FACET_MAPPING);
	if (StringUtils.isBlank(facetMapping)) {
		String columnString = conf.get(ConfigurationUtil.SOLR_COLUMN_MAPPING);
		if (StringUtils.isBlank(columnString)) {
			throw new IOException("no column mapping found!");
		}
		solrColumns = ConfigurationUtil.getAllColumns(columnString);
		if (readColIDs.size() > solrColumns.length) {
			throw new IOException("read column count larger than that in column mapping string!");
		}
	} else {
		if (readColIDs.size() != 2) {
			throw new IOException("read column should be 2 with facet mapping");
		}
		solrColumns = conf.get(Constants.LIST_COLUMNS).split(",");
	}

	if (conf.get(Constants.LIST_COLUMNS) != null) {
		hiveColNames = Arrays.asList(StringUtils.split(conf.get(Constants.LIST_COLUMNS), ","));
	}

	if (conf.get(Constants.LIST_COLUMN_TYPES) != null) {
		hiveColTypes = TypeInfoUtils.getTypeInfosFromTypeString(conf.get(Constants.LIST_COLUMN_TYPES));
		for (TypeInfo ti : hiveColTypes) {
			rowOI.add(TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(ti));
		}
	}

	this.split = split;
	SolrTable table = new SolrTable(conf);
	cursor = table.getCursor((int) split.getStart(), (int) split.getLength());
}