private static StorageDescriptor makeStorageDescriptor(
String tableName, List<Column> columns, Storage storage) {
if (storage.isSorted() || storage.isSkewed()) {
throw new IllegalArgumentException(
"Writing to sorted and/or skewed table/partition is not supported");
}
SerDeInfo serdeInfo = new SerDeInfo();
serdeInfo.setName(tableName);
serdeInfo.setSerializationLib(storage.getStorageFormat().getSerDeNullable());
serdeInfo.setParameters(storage.getSerdeParameters());
StorageDescriptor sd = new StorageDescriptor();
sd.setLocation(emptyToNull(storage.getLocation()));
sd.setCols(columns.stream().map(MetastoreUtil::toMetastoreApiFieldSchema).collect(toList()));
sd.setSerdeInfo(serdeInfo);
sd.setInputFormat(storage.getStorageFormat().getInputFormatNullable());
sd.setOutputFormat(storage.getStorageFormat().getOutputFormatNullable());
sd.setParameters(ImmutableMap.of());
Optional<HiveBucketProperty> bucketProperty = storage.getBucketProperty();
if (bucketProperty.isPresent()) {
sd.setNumBuckets(bucketProperty.get().getBucketCount());
sd.setBucketCols(bucketProperty.get().getBucketedBy());
}
return sd;
}
private static Properties getHiveSchema(
Storage sd,
List<Column> dataColumns,
List<Column> tableDataColumns,
Map<String, String> parameters,
String databaseName,
String tableName,
List<Column> partitionKeys) {
// Mimics function in Hive:
// MetaStoreUtils.getSchema(StorageDescriptor, StorageDescriptor, Map<String, String>, String,
// String, List<FieldSchema>)
Properties schema = new Properties();
schema.setProperty(FILE_INPUT_FORMAT, sd.getStorageFormat().getInputFormat());
schema.setProperty(FILE_OUTPUT_FORMAT, sd.getStorageFormat().getOutputFormat());
schema.setProperty(META_TABLE_NAME, databaseName + "." + tableName);
schema.setProperty(META_TABLE_LOCATION, sd.getLocation());
if (sd.getBucketProperty().isPresent()) {
schema.setProperty(BUCKET_FIELD_NAME, sd.getBucketProperty().get().getBucketedBy().get(0));
schema.setProperty(
BUCKET_COUNT, Integer.toString(sd.getBucketProperty().get().getBucketCount()));
} else {
schema.setProperty(BUCKET_COUNT, "0");
}
for (Map.Entry<String, String> param : sd.getSerdeParameters().entrySet()) {
schema.setProperty(param.getKey(), (param.getValue() != null) ? param.getValue() : "");
}
schema.setProperty(SERIALIZATION_LIB, sd.getStorageFormat().getSerDe());
StringBuilder columnNameBuilder = new StringBuilder();
StringBuilder columnTypeBuilder = new StringBuilder();
StringBuilder columnCommentBuilder = new StringBuilder();
boolean first = true;
for (Column column : tableDataColumns) {
if (!first) {
columnNameBuilder.append(",");
columnTypeBuilder.append(":");
columnCommentBuilder.append('\0');
}
columnNameBuilder.append(column.getName());
columnTypeBuilder.append(column.getType());
columnCommentBuilder.append(column.getComment().orElse(""));
first = false;
}
String columnNames = columnNameBuilder.toString();
String columnTypes = columnTypeBuilder.toString();
schema.setProperty(META_TABLE_COLUMNS, columnNames);
schema.setProperty(META_TABLE_COLUMN_TYPES, columnTypes);
schema.setProperty("columns.comments", columnCommentBuilder.toString());
schema.setProperty(SERIALIZATION_DDL, toThriftDdl(tableName, dataColumns));
String partString = "";
String partStringSep = "";
String partTypesString = "";
String partTypesStringSep = "";
for (Column partKey : partitionKeys) {
partString += partStringSep;
partString += partKey.getName();
partTypesString += partTypesStringSep;
partTypesString += partKey.getType().getHiveTypeName();
if (partStringSep.length() == 0) {
partStringSep = "/";
partTypesStringSep = ":";
}
}
if (partString.length() > 0) {
schema.setProperty(META_TABLE_PARTITION_COLUMNS, partString);
schema.setProperty(META_TABLE_PARTITION_COLUMN_TYPES, partTypesString);
}
if (parameters != null) {
for (Map.Entry<String, String> entry : parameters.entrySet()) {
// add non-null parameters to the schema
if (entry.getValue() != null) {
schema.setProperty(entry.getKey(), entry.getValue());
}
}
}
return schema;
}