String tmpPath = stripQuotes(fromTree.getText());
URI fromURI = EximUtil.getValidatedURI(conf, tmpPath);
FileSystem fs = FileSystem.get(fromURI, conf);
String dbname = null;
CreateTableDesc tblDesc = null;
List<AddPartitionDesc> partitionDescs = new ArrayList<AddPartitionDesc>();
Path fromPath = new Path(fromURI.getScheme(), fromURI.getAuthority(),
fromURI.getPath());
boolean isLocal = FileUtils.isLocalFile(conf, fromURI);
inputs.add(new ReadEntity(fromPath, isLocal));
try {
Path metadataPath = new Path(fromPath, METADATA_NAME);
Map.Entry<org.apache.hadoop.hive.metastore.api.Table,
List<Partition>> rv = EximUtil.readMetaData(fs, metadataPath);
dbname = SessionState.get().getCurrentDatabase();
org.apache.hadoop.hive.metastore.api.Table table = rv.getKey();
tblDesc = new CreateTableDesc(
table.getTableName(),
false, // isExternal: set to false here, can be overwritten by the
// IMPORT stmt
table.getSd().getCols(),
table.getPartitionKeys(),
table.getSd().getBucketCols(),
table.getSd().getSortCols(),
table.getSd().getNumBuckets(),
null, null, null, null, null, // these 5 delims passed as serde params
null, // comment passed as table params
table.getSd().getInputFormat(),
table.getSd().getOutputFormat(),
null, // location: set to null here, can be
// overwritten by the IMPORT stmt
table.getSd().getSerdeInfo().getSerializationLib(),
null, // storagehandler passed as table params
table.getSd().getSerdeInfo().getParameters(),
table.getParameters(), false,
(null == table.getSd().getSkewedInfo()) ? null : table.getSd().getSkewedInfo()
.getSkewedColNames(),
(null == table.getSd().getSkewedInfo()) ? null : table.getSd().getSkewedInfo()
.getSkewedColValues());
tblDesc.setStoredAsSubDirectories(table.getSd().isStoredAsSubDirectories());
List<FieldSchema> partCols = tblDesc.getPartCols();
List<String> partColNames = new ArrayList<String>(partCols.size());
for (FieldSchema fsc : partCols) {
partColNames.add(fsc.getName());
}
List<Partition> partitions = rv.getValue();
for (Partition partition : partitions) {
// TODO: this should not create AddPartitionDesc per partition
AddPartitionDesc partsDesc = new AddPartitionDesc(dbname, tblDesc.getTableName(),
EximUtil.makePartSpec(tblDesc.getPartCols(), partition.getValues()),
partition.getSd().getLocation(), partition.getParameters());
AddPartitionDesc.OnePartitionDesc partDesc = partsDesc.getPartition(0);
partDesc.setInputFormat(partition.getSd().getInputFormat());
partDesc.setOutputFormat(partition.getSd().getOutputFormat());
partDesc.setNumBuckets(partition.getSd().getNumBuckets());
partDesc.setCols(partition.getSd().getCols());
partDesc.setSerializationLib(partition.getSd().getSerdeInfo().getSerializationLib());
partDesc.setSerdeParams(partition.getSd().getSerdeInfo().getParameters());
partDesc.setBucketCols(partition.getSd().getBucketCols());
partDesc.setSortCols(partition.getSd().getSortCols());
partDesc.setLocation(new Path(fromPath,
Warehouse.makePartName(tblDesc.getPartCols(), partition.getValues())).toString());
partitionDescs.add(partsDesc);
}
} catch (IOException e) {
throw new SemanticException(ErrorMsg.INVALID_PATH.getMsg(), e);
}
LOG.debug("metadata read and parsed");
for (int i = 1; i < ast.getChildCount(); ++i) {
ASTNode child = (ASTNode) ast.getChild(i);
switch (child.getToken().getType()) {
case HiveParser.KW_EXTERNAL:
tblDesc.setExternal(true);
break;
case HiveParser.TOK_TABLELOCATION:
String location = unescapeSQLString(child.getChild(0).getText());
location = EximUtil.relativeToAbsolutePath(conf, location);
tblDesc.setLocation(location);
break;
case HiveParser.TOK_TAB:
Tree tableTree = child.getChild(0);
// initialize destination table/partition
String tableName = getUnescapedName((ASTNode)tableTree);
tblDesc.setTableName(tableName);
// get partition metadata if partition specified
LinkedHashMap<String, String> partSpec = new LinkedHashMap<String, String>();
if (child.getChildCount() == 2) {
ASTNode partspec = (ASTNode) child.getChild(1);
// partSpec is a mapping from partition column name to its value.
for (int j = 0; j < partspec.getChildCount(); ++j) {
ASTNode partspec_val = (ASTNode) partspec.getChild(j);
String val = null;
String colName = unescapeIdentifier(partspec_val.getChild(0)
.getText().toLowerCase());
if (partspec_val.getChildCount() < 2) { // DP in the form of T
// partition (ds, hr)
throw new SemanticException(
ErrorMsg.INVALID_PARTITION
.getMsg(" - Dynamic partitions not allowed"));
} else { // in the form of T partition (ds="2010-03-03")
val = stripQuotes(partspec_val.getChild(1).getText());
}
partSpec.put(colName, val);
}
boolean found = false;
for (Iterator<AddPartitionDesc> partnIter = partitionDescs
.listIterator(); partnIter.hasNext();) {
AddPartitionDesc addPartitionDesc = partnIter.next();
if (!found && addPartitionDesc.getPartition(0).getPartSpec().equals(partSpec)) {
found = true;
} else {
partnIter.remove();
}
}
if (!found) {
throw new SemanticException(
ErrorMsg.INVALID_PARTITION
.getMsg(" - Specified partition not found in import directory"));
}
}
}
}
if (tblDesc.getTableName() == null) {
throw new SemanticException(ErrorMsg.NEED_TABLE_SPECIFICATION.getMsg());
} else {
conf.set("import.destination.table", tblDesc.getTableName());
for (AddPartitionDesc addPartitionDesc : partitionDescs) {
addPartitionDesc.setTableName(tblDesc.getTableName());
}
}
Warehouse wh = new Warehouse(conf);
try {
Table table = db.getTable(tblDesc.getTableName());
checkTable(table, tblDesc);
LOG.debug("table " + tblDesc.getTableName()
+ " exists: metadata checked");
tableExists = true;
conf.set("import.destination.dir", table.getDataLocation().toString());
if (table.isPartitioned()) {
LOG.debug("table partitioned");
for (AddPartitionDesc addPartitionDesc : partitionDescs) {
Map<String, String> partSpec = addPartitionDesc.getPartition(0).getPartSpec();
if (db.getPartition(table, partSpec, false) == null) {
rootTasks.add(addSinglePartition(fromURI, fs, tblDesc, table, wh, addPartitionDesc));
} else {
throw new SemanticException(
ErrorMsg.PARTITION_EXISTS.getMsg(partSpecToString(partSpec)));
}
}
} else {
LOG.debug("table non-partitioned");
checkTargetLocationEmpty(fs, new Path(table.getDataLocation()
.toString()));
loadTable(fromURI, table);
}
// Set this to read because we can't overwrite any existing partitions
outputs.add(new WriteEntity(table, WriteEntity.WriteType.DDL_NO_LOCK));
} catch (InvalidTableException e) {
LOG.debug("table " + tblDesc.getTableName() + " does not exist");
Task<?> t = TaskFactory.get(new DDLWork(getInputs(), getOutputs(),
tblDesc), conf);
Table table = new Table(dbname, tblDesc.getTableName());
String currentDb = SessionState.get().getCurrentDatabase();
conf.set("import.destination.dir",
wh.getTablePath(db.getDatabaseCurrent(),
tblDesc.getTableName()).toString());
if ((tblDesc.getPartCols() != null) && (tblDesc.getPartCols().size() != 0)) {
for (AddPartitionDesc addPartitionDesc : partitionDescs) {
t.addDependentTask(
addSinglePartition(fromURI, fs, tblDesc, table, wh, addPartitionDesc));
}
} else {
LOG.debug("adding dependent CopyWork/MoveWork for table");
if (tblDesc.isExternal() && (tblDesc.getLocation() == null)) {
LOG.debug("Importing in place, no emptiness check, no copying/loading");
Path dataPath = new Path(fromURI.toString(), "data");
tblDesc.setLocation(dataPath.toString());
} else {
Path tablePath = null;
if (tblDesc.getLocation() != null) {
tablePath = new Path(tblDesc.getLocation());
} else {
tablePath = wh.getTablePath(db.getDatabaseCurrent(), tblDesc.getTableName());
}
checkTargetLocationEmpty(fs, tablePath);
t.addDependentTask(loadTable(fromURI, table));
}
}