use of org.apache.hadoop.hive.ql.metadata.CheckConstraint in project hive by apache.
the class DDLTask method describeTable.
/**
* Write the description of a table to a file.
*
* @param db
* The database in question.
* @param descTbl
* This is the table we're interested in.
* @return Returns 0 when execution succeeds and above 0 if it fails.
* @throws HiveException
* Throws this exception if an unexpected error occurs.
* @throws MetaException
*/
private int describeTable(Hive db, DescTableDesc descTbl) throws HiveException, MetaException {
String colPath = descTbl.getColumnPath();
String tableName = descTbl.getTableName();
// describe the table - populate the output stream
Table tbl = db.getTable(tableName, false);
if (tbl == null) {
throw new HiveException(ErrorMsg.INVALID_TABLE, tableName);
}
Partition part = null;
if (descTbl.getPartSpec() != null) {
part = db.getPartition(tbl, descTbl.getPartSpec(), false);
if (part == null) {
throw new HiveException(ErrorMsg.INVALID_PARTITION, StringUtils.join(descTbl.getPartSpec().keySet(), ','), tableName);
}
tbl = part.getTable();
}
DataOutputStream outStream = getOutputStream(descTbl.getResFile());
try {
LOG.debug("DDLTask: got data for {}", tableName);
List<FieldSchema> cols = null;
List<ColumnStatisticsObj> colStats = null;
Deserializer deserializer = tbl.getDeserializer(true);
if (deserializer instanceof AbstractSerDe) {
String errorMsgs = ((AbstractSerDe) deserializer).getConfigurationErrors();
if (errorMsgs != null && !errorMsgs.isEmpty()) {
throw new SQLException(errorMsgs);
}
}
if (colPath.equals(tableName)) {
cols = (part == null || tbl.getTableType() == TableType.VIRTUAL_VIEW) ? tbl.getCols() : part.getCols();
if (!descTbl.isFormatted()) {
cols.addAll(tbl.getPartCols());
}
if (tbl.isPartitioned() && part == null) {
// No partitioned specified for partitioned table, lets fetch all.
Map<String, String> tblProps = tbl.getParameters() == null ? new HashMap<String, String>() : tbl.getParameters();
Map<String, Long> valueMap = new HashMap<>();
Map<String, Boolean> stateMap = new HashMap<>();
for (String stat : StatsSetupConst.supportedStats) {
valueMap.put(stat, 0L);
stateMap.put(stat, true);
}
PartitionIterable parts = new PartitionIterable(db, tbl, null, conf.getIntVar(HiveConf.ConfVars.METASTORE_BATCH_RETRIEVE_MAX));
int numParts = 0;
for (Partition partition : parts) {
Map<String, String> props = partition.getParameters();
Boolean state = StatsSetupConst.areBasicStatsUptoDate(props);
for (String stat : StatsSetupConst.supportedStats) {
stateMap.put(stat, stateMap.get(stat) && state);
if (props != null && props.get(stat) != null) {
valueMap.put(stat, valueMap.get(stat) + Long.parseLong(props.get(stat)));
}
}
numParts++;
}
for (String stat : StatsSetupConst.supportedStats) {
StatsSetupConst.setBasicStatsState(tblProps, Boolean.toString(stateMap.get(stat)));
tblProps.put(stat, valueMap.get(stat).toString());
}
tblProps.put(StatsSetupConst.NUM_PARTITIONS, Integer.toString(numParts));
tbl.setParameters(tblProps);
}
} else {
if (descTbl.isFormatted()) {
// when column name is specified in describe table DDL, colPath will
// will be table_name.column_name
String colName = colPath.split("\\.")[1];
String[] dbTab = Utilities.getDbTableName(tableName);
List<String> colNames = new ArrayList<String>();
colNames.add(colName.toLowerCase());
if (null == part) {
if (tbl.isPartitioned()) {
Map<String, String> tblProps = tbl.getParameters() == null ? new HashMap<String, String>() : tbl.getParameters();
if (tbl.isPartitionKey(colNames.get(0))) {
FieldSchema partCol = tbl.getPartColByName(colNames.get(0));
cols = Collections.singletonList(partCol);
PartitionIterable parts = new PartitionIterable(db, tbl, null, conf.getIntVar(HiveConf.ConfVars.METASTORE_BATCH_RETRIEVE_MAX));
ColumnInfo ci = new ColumnInfo(partCol.getName(), TypeInfoUtils.getTypeInfoFromTypeString(partCol.getType()), null, false);
ColStatistics cs = StatsUtils.getColStatsForPartCol(ci, parts, conf);
ColumnStatisticsData data = new ColumnStatisticsData();
ColStatistics.Range r = cs.getRange();
StatObjectConverter.fillColumnStatisticsData(partCol.getType(), data, r == null ? null : r.minValue, r == null ? null : r.maxValue, r == null ? null : r.minValue, r == null ? null : r.maxValue, r == null ? null : r.minValue.toString(), r == null ? null : r.maxValue.toString(), cs.getNumNulls(), cs.getCountDistint(), null, cs.getAvgColLen(), cs.getAvgColLen(), cs.getNumTrues(), cs.getNumFalses());
ColumnStatisticsObj cso = new ColumnStatisticsObj(partCol.getName(), partCol.getType(), data);
colStats = Collections.singletonList(cso);
StatsSetupConst.setColumnStatsState(tblProps, colNames);
} else {
cols = Hive.getFieldsFromDeserializer(colPath, deserializer);
List<String> parts = db.getPartitionNames(dbTab[0].toLowerCase(), dbTab[1].toLowerCase(), (short) -1);
AggrStats aggrStats = db.getAggrColStatsFor(dbTab[0].toLowerCase(), dbTab[1].toLowerCase(), colNames, parts);
colStats = aggrStats.getColStats();
if (parts.size() == aggrStats.getPartsFound()) {
StatsSetupConst.setColumnStatsState(tblProps, colNames);
} else {
StatsSetupConst.removeColumnStatsState(tblProps, colNames);
}
}
tbl.setParameters(tblProps);
} else {
cols = Hive.getFieldsFromDeserializer(colPath, deserializer);
colStats = db.getTableColumnStatistics(dbTab[0].toLowerCase(), dbTab[1].toLowerCase(), colNames);
}
} else {
List<String> partitions = new ArrayList<String>();
partitions.add(part.getName());
cols = Hive.getFieldsFromDeserializer(colPath, deserializer);
colStats = db.getPartitionColumnStatistics(dbTab[0].toLowerCase(), dbTab[1].toLowerCase(), partitions, colNames).get(part.getName());
}
} else {
cols = Hive.getFieldsFromDeserializer(colPath, deserializer);
}
}
PrimaryKeyInfo pkInfo = null;
ForeignKeyInfo fkInfo = null;
UniqueConstraint ukInfo = null;
NotNullConstraint nnInfo = null;
DefaultConstraint dInfo = null;
CheckConstraint cInfo = null;
if (descTbl.isExt() || descTbl.isFormatted()) {
pkInfo = db.getPrimaryKeys(tbl.getDbName(), tbl.getTableName());
fkInfo = db.getForeignKeys(tbl.getDbName(), tbl.getTableName());
ukInfo = db.getUniqueConstraints(tbl.getDbName(), tbl.getTableName());
nnInfo = db.getNotNullConstraints(tbl.getDbName(), tbl.getTableName());
dInfo = db.getDefaultConstraints(tbl.getDbName(), tbl.getTableName());
cInfo = db.getCheckConstraints(tbl.getDbName(), tbl.getTableName());
}
fixDecimalColumnTypeName(cols);
// In case the query is served by HiveServer2, don't pad it with spaces,
// as HiveServer2 output is consumed by JDBC/ODBC clients.
boolean isOutputPadded = !SessionState.get().isHiveServerQuery();
formatter.describeTable(outStream, colPath, tableName, tbl, part, cols, descTbl.isFormatted(), descTbl.isExt(), isOutputPadded, colStats, pkInfo, fkInfo, ukInfo, nnInfo, dInfo, cInfo);
LOG.debug("DDLTask: written data for {}", tableName);
} catch (SQLException e) {
throw new HiveException(e, ErrorMsg.GENERIC_ERROR, tableName);
} finally {
IOUtils.closeStream(outStream);
}
return 0;
}
use of org.apache.hadoop.hive.ql.metadata.CheckConstraint in project hive by apache.
the class SemanticAnalyzer method getCheckConstraintExpr.
private ExprNodeDesc getCheckConstraintExpr(Table tbl, Operator input, RowResolver inputRR, String dest) throws SemanticException {
CheckConstraint cc = null;
try {
cc = Hive.get().getEnabledCheckConstraints(tbl.getDbName(), tbl.getTableName());
} catch (HiveException e) {
throw new SemanticException(e);
}
if (cc == null || cc.getCheckConstraints().isEmpty()) {
return null;
}
// build a map which tracks the name of column in input's signature to corresponding table column name
// this will be used to replace column references in CHECK expression AST with corresponding column name
// in input
Map<String, String> col2Cols = new HashMap<>();
List<ColumnInfo> colInfos = input.getSchema().getSignature();
int colIdx = 0;
if (updating(dest)) {
// if this is an update we need to skip the first col since it is row id
colIdx = 1;
}
for (FieldSchema fs : tbl.getCols()) {
// since SQL is case insenstive just to make sure that the comparison b/w column names
// and check expression's column reference work convert the key to lower case
col2Cols.put(fs.getName().toLowerCase(), colInfos.get(colIdx).getInternalName());
colIdx++;
}
List<String> checkExprStrs = cc.getCheckExpressionList();
TypeCheckCtx typeCheckCtx = new TypeCheckCtx(inputRR);
ExprNodeDesc checkAndExprs = null;
for (String checkExprStr : checkExprStrs) {
try {
ParseDriver parseDriver = new ParseDriver();
ASTNode checkExprAST = parseDriver.parseExpression(checkExprStr);
// replace column references in checkExprAST with corresponding columns in input
replaceColumnReference(checkExprAST, col2Cols, inputRR);
Map<ASTNode, ExprNodeDesc> genExprs = TypeCheckProcFactory.genExprNode(checkExprAST, typeCheckCtx);
ExprNodeDesc checkExpr = genExprs.get(checkExprAST);
// Check constraint fails only if it evaluates to false, NULL/UNKNOWN should evaluate to TRUE
ExprNodeDesc notFalseCheckExpr = TypeCheckProcFactory.DefaultExprProcessor.getFuncExprNodeDesc("isnotfalse", checkExpr);
if (checkAndExprs == null) {
checkAndExprs = notFalseCheckExpr;
} else {
checkAndExprs = TypeCheckProcFactory.DefaultExprProcessor.getFuncExprNodeDesc("and", checkAndExprs, notFalseCheckExpr);
}
} catch (Exception e) {
throw new SemanticException(e);
}
}
return checkAndExprs;
}
use of org.apache.hadoop.hive.ql.metadata.CheckConstraint in project hive by apache.
the class ConstraintExprGenerator method getCheckConstraintExpr.
private T getCheckConstraintExpr(Table tbl, List<ColumnInfo> inputColInfos, RowResolver inputRR, boolean isUpdateStatement) throws SemanticException {
CheckConstraint cc = null;
try {
cc = Hive.get().getEnabledCheckConstraints(tbl.getDbName(), tbl.getTableName());
} catch (HiveException e) {
throw new SemanticException(e);
}
if (cc == null || cc.getCheckConstraints().isEmpty()) {
return null;
}
// build a map which tracks the name of column in input's signature to corresponding table column name
// this will be used to replace column references in CHECK expression AST with corresponding column name
// in input
Map<String, String> col2Cols = new HashMap<>();
int colIdx = 0;
if (isUpdateStatement) {
// if this is an update we need to skip the first col since it is row id
colIdx = 1;
}
for (FieldSchema fs : tbl.getCols()) {
// since SQL is case insenstive just to make sure that the comparison b/w column names
// and check expression's column reference work convert the key to lower case
col2Cols.put(fs.getName().toLowerCase(), inputColInfos.get(colIdx).getInternalName());
colIdx++;
}
List<String> checkExprStrs = cc.getCheckExpressionList();
TypeCheckCtx typeCheckCtx = new TypeCheckCtx(inputRR);
T checkAndExprs = null;
for (String checkExprStr : checkExprStrs) {
try {
ParseDriver parseDriver = new ParseDriver();
ASTNode checkExprAST = parseDriver.parseExpression(checkExprStr);
// replace column references in checkExprAST with corresponding columns in input
replaceColumnReference(checkExprAST, col2Cols, inputRR);
Map<ASTNode, T> genExprs = typeCheckProcFactory.genExprNode(checkExprAST, typeCheckCtx);
T checkExpr = genExprs.get(checkExprAST);
// Check constraint fails only if it evaluates to false, NULL/UNKNOWN should evaluate to TRUE
T notFalseCheckExpr = exprProcessor.getFuncExprNodeDesc("isnotfalse", checkExpr);
if (checkAndExprs == null) {
checkAndExprs = notFalseCheckExpr;
} else {
checkAndExprs = exprProcessor.getFuncExprNodeDesc("and", checkAndExprs, notFalseCheckExpr);
}
} catch (Exception e) {
throw new SemanticException(e);
}
}
return checkAndExprs;
}
use of org.apache.hadoop.hive.ql.metadata.CheckConstraint in project hive by apache.
the class DDLPlanUtils method getAlterTableStmtCheckConstraint.
public void getAlterTableStmtCheckConstraint(CheckConstraint ck, List<String> constraints) {
if (!CheckConstraint.isNotEmpty(ck)) {
return;
}
Map<String, List<CheckConstraint.CheckConstraintCol>> checkConstraints = ck.getCheckConstraints();
for (String constraintName : checkConstraints.keySet()) {
List<CheckConstraintCol> checkConstraintCols = checkConstraints.get(constraintName);
if (checkConstraintCols != null && checkConstraintCols.size() > 0) {
for (CheckConstraintCol col : checkConstraintCols) {
ST command = new ST(ALTER_TABLE_ADD_CHECK_CONSTRAINT);
command.add(DATABASE_NAME, ck.getDatabaseName());
command.add(TABLE_NAME, ck.getTableName());
command.add(CONSTRAINT_NAME, constraintName);
command.add(CHECK_EXPRESSION, col.getCheckExpression());
command.add(ENABLE, col.getEnable());
command.add(VALIDATE, col.getValidate());
command.add(RELY, col.getRely());
constraints.add(command.render());
}
}
}
}
Aggregations