public IDataWriter getDataWriter(
String sourceNodeId,
ISymmetricDialect symmetricDialect,
TransformWriter transformWriter,
List<IDatabaseWriterFilter> filters,
List<IDatabaseWriterErrorHandler> errorHandlers,
List<? extends Conflict> conflictSettings,
List<ResolvedData> resolvedData) {
int maxRowsBeforeFlush =
parameterService.getInt("mssql.bulk.load.max.rows.before.flush", 100000);
boolean fireTriggers = parameterService.is("mssql.bulk.load.fire.triggers", false);
String uncPath = parameterService.getString("mssql.bulk.load.unc.path");
String rowTerminator =
StringEscapeUtils.unescapeJava(
parameterService.getString("mssql.bulk.load.row.terminator", "\\r\\n"));
String fieldTerminator =
StringEscapeUtils.unescapeJava(
parameterService.getString("mssql.bulk.load.field.terminator", "||"));
return new MsSqlBulkDatabaseWriter(
symmetricDialect.getPlatform(),
stagingManager,
jdbcExtractor,
maxRowsBeforeFlush,
fireTriggers,
uncPath,
fieldTerminator,
rowTerminator);
}
public static boolean isValidFileName(String fileName) {
if (StringUtils.isBlank(fileName)) {
return false;
}
for (String blacklistChar : PropsValues.SYNC_FILE_BLACKLIST_CHARS) {
if (fileName.contains(blacklistChar)) {
return false;
}
}
for (String blacklistLastChar : PropsValues.SYNC_FILE_BLACKLIST_CHARS_LAST) {
if (blacklistLastChar.startsWith("\\u")) {
blacklistLastChar = StringEscapeUtils.unescapeJava(blacklistLastChar);
}
if (fileName.endsWith(blacklistLastChar)) {
return false;
}
}
String nameWithoutExtension = FilenameUtils.removeExtension(fileName);
for (String blacklistName : PropsValues.SYNC_FILE_BLACKLIST_NAMES) {
if (nameWithoutExtension.equalsIgnoreCase(blacklistName)) {
return false;
}
}
return true;
}
@Override
public String getLabel() {
return StringEscapeUtils.unescapeJava(
MessageFormat.format(
ResourceBundle.getBundle("bundles/nexu")
.getString("product.selection.detected.card.button.label"),
this.getTerminalIndex(),
this.getTerminalLabel(),
this.getAtr()));
}
/**
* Unescapes any Java literals found in the <code>String</code>. For example, it will turn a
* sequence of <code>'\'</code> and <code>'n'</code> into a newline character, unless the <code>
* '\'</code> is preceded by another <code>'\'</code>.
*
* @param str the <code>String</code> to unescape, may be null
* @return a new unescaped <code>String</code>, <code>null</code> if null string input
*/
public static String unescapeJava(String str) {
if (str == null) {
return null;
}
try {
StringWriter writer = new StringWriter(str.length());
unescapeJava(writer, str);
return writer.toString();
} catch (IOException ioe) {
// this should never ever happen while writing to a StringWriter
throw new UnhandledException(ioe);
}
}
/**
* Unescapes any Java literals found in the <code>String</code>. For example, it will turn a
* sequence of <code>'\'</code> and <code>'n'</code> into a newline character, unless the <code>
* '\'</code> is preceded by another <code>'\'</code>.
*
* @param str the <code>String</code> to unescape, may be null
* @return a new unescaped <code>String</code>, <code>null</code> if null string input
*/
public static String unescapeJava(String str) {
if (str == null) {
return null;
}
try {
StringPrintWriter writer = new StringPrintWriter(str.length());
unescapeJava(writer, str);
return writer.getString();
} catch (IOException ioe) {
// this should never ever happen while writing to a StringWriter
ioe.printStackTrace();
return null;
}
}
@Override
public void process(JCas aJCas) throws AnalysisEngineProcessException {
modelProvider.configure(aJCas.getCas());
modelProvider.getResource();
String documentText = aJCas.getDocumentText();
// Unescape text, if Chinese, Arabic and stuff are thrown in as UTF-8 escaped sequence it
// will lead to an increased error rate
String unescapedDocumentText = StringEscapeUtils.unescapeJava(documentText);
String language = detectLanguage(unescapedDocumentText);
aJCas.setDocumentLanguage(language);
}
public static String recodeForDbp38(String uriString) {
String ret;
URI uri = null;
if (uriString.startsWith("http://") || uriString.startsWith("https://")) {
// First try to parse the string as an URI so that any superfluous
// percent-encodings can get decoded later
try {
uri = new URI(uriString);
} catch (Exception ex) {
throw new GateRuntimeException("Could not parse URI " + uriString, ex);
}
// now use this constructor to-recode only the necessary parts
try {
String path = uri.getPath();
path = path.trim();
path = path.replaceAll(" +", "_");
uri =
new URI(
uri.getScheme(), null, uri.getHost(), -1, path, uri.getQuery(), uri.getFragment());
} catch (Exception ex) {
throw new GateRuntimeException("Could not re-construct URI: " + uri);
}
ret = uri.toString();
} else {
if (uriString.contains("\\u")) {
uriString = StringEscapeUtils.unescapeJava(uriString);
}
uriString = uriString.trim();
uriString = uriString.replaceAll(" +", "_");
// We need to %-encode colons, otherwise the getPath() method will return
// null ...
uriString = uriString.replaceAll(":", "%3A");
try {
uri = new URI(uriString);
// decode and prepare for minimal percent encoding
uriString = uri.getPath();
} catch (URISyntaxException ex) {
// do nothing: the uriString must already be ready for percent-encoding
}
uriString = uriString.replaceAll(" +", "_");
try {
uri = new URI(null, null, null, -1, "/" + uriString, null, null);
} catch (Exception ex) {
throw new GateRuntimeException("Could not re-construct URI part: " + uriString);
}
ret = uri.toString().substring(1);
}
return ret;
}
public static String replaceAll(String s) {
YAMLConfig config = HoloAPI.getInstance().getConfig(HoloAPI.ConfigType.MAIN);
ConfigurationSection cs = config.getConfigurationSection("specialCharacters");
if (cs != null) {
for (String key : cs.getKeys(false)) {
if (s.contains(key)) {
s =
s.replace(
key,
StringEscapeUtils.unescapeJava(
"\\u" + config.getString("specialCharacters." + key)));
}
}
}
return s;
}
public JsonObject getCrmAccount(CrmSettings settings) {
WebResource resource =
getAmoCrmWebResource(
settings.getLink()
+ GET_CRM_ACCOUNT_PATH
+ "?USER_LOGIN="******"&USER_HASH="
+ settings.getToken());
ClientResponse response = resource.get(ClientResponse.class);
String output = StringEscapeUtils.unescapeJava(response.getEntity(String.class));
JsonParser parser = new JsonParser();
return parser.parse(output).getAsJsonObject();
}
public static String getSanitizedFileName(String fileName, String extension) {
for (String blacklistChar : PropsValues.SYNC_FILE_BLACKLIST_CHARS) {
fileName = fileName.replace(blacklistChar, "_");
}
for (String blacklistCharLast : PropsValues.SYNC_FILE_BLACKLIST_CHARS_LAST) {
if (blacklistCharLast.startsWith("\\u")) {
blacklistCharLast = StringEscapeUtils.unescapeJava(blacklistCharLast);
}
if (fileName.endsWith(blacklistCharLast)) {
fileName = fileName.substring(0, fileName.length() - 1);
}
}
if ((extension != null) && !extension.isEmpty()) {
int x = fileName.lastIndexOf(".");
if ((x == -1) || !extension.equalsIgnoreCase(fileName.substring(x + 1))) {
fileName += "." + extension;
}
}
if (fileName.length() > 255) {
int x = fileName.length() - 1;
if ((extension != null) && !extension.isEmpty()) {
x = fileName.lastIndexOf(".");
}
int y = x - (fileName.length() - 255);
fileName = fileName.substring(0, y) + fileName.substring(x);
}
return fileName;
}
@Override
public RecordWriter<WritableComparable, Tuple> getRecordWriter(TaskAttemptContext context)
throws IOException, InterruptedException {
Configuration conf = context.getConfiguration();
PigTextOutputFormat outputFormat;
if (conf.get(CubertStrings.TEXT_OUTPUT_SEPARATOR) == null) {
outputFormat = new PigTextOutputFormat(defaultDelimiter);
} else {
String str = conf.get(CubertStrings.TEXT_OUTPUT_SEPARATOR);
str = StringEscapeUtils.unescapeJava(str);
byte[] bytes = str.getBytes("UTF-8");
if (bytes.length > 1)
throw new RuntimeException(
String.format("Invalid separator in text output format %s", str));
outputFormat = new PigTextOutputFormat(bytes[0]);
}
return outputFormat.getRecordWriter(context);
}
/**
* Unescapes any JavaScript literals found in the <code>String</code> to a <code>Writer</code>.
*
* <p>For example, it will turn a sequence of <code>'\'</code> and <code>'n'</code> into a newline
* character, unless the <code>'\'</code> is preceded by another <code>'\'</code>.
*
* <p>A <code>null</code> string input has no effect.
*
* @see #unescapeJava(Writer,String)
* @param out the <code>Writer</code> used to output unescaped characters
* @param str the <code>String</code> to unescape, may be null
* @throws IllegalArgumentException if the Writer is <code>null</code>
* @throws IOException if error occurs on underlying Writer
*/
public static void unescapeJavaScript(Writer out, String str) throws IOException {
unescapeJava(out, str);
}
@Override
public final void createTable(final CatalogProtos.TableDescProto tableDescProto)
throws CatalogException {
HiveCatalogStoreClientPool.HiveCatalogStoreClient client = null;
TableDesc tableDesc = new TableDesc(tableDescProto);
String[] splitted = CatalogUtil.splitFQTableName(tableDesc.getName());
String databaseName = splitted[0];
String tableName = splitted[1];
try {
client = clientPool.getClient();
org.apache.hadoop.hive.metastore.api.Table table =
new org.apache.hadoop.hive.metastore.api.Table();
table.setDbName(databaseName);
table.setTableName(tableName);
table.setParameters(
new HashMap<String, String>(tableDesc.getMeta().getOptions().getAllKeyValus()));
// TODO: set owner
// table.setOwner();
StorageDescriptor sd = new StorageDescriptor();
sd.setSerdeInfo(new SerDeInfo());
sd.getSerdeInfo().setParameters(new HashMap<String, String>());
sd.getSerdeInfo().setName(table.getTableName());
// if tajo set location method, thrift client make exception as follows:
// Caused by: MetaException(message:java.lang.NullPointerException)
// If you want to modify table path, you have to modify on Hive cli.
if (tableDesc.isExternal()) {
table.setTableType(TableType.EXTERNAL_TABLE.name());
table.putToParameters("EXTERNAL", "TRUE");
Path tablePath = new Path(tableDesc.getUri());
FileSystem fs = tablePath.getFileSystem(conf);
if (fs.isFile(tablePath)) {
LOG.warn("A table path is a file, but HiveCatalogStore does not allow a file path.");
sd.setLocation(tablePath.getParent().toString());
} else {
sd.setLocation(tablePath.toString());
}
}
// set column information
List<Column> columns = tableDesc.getSchema().getRootColumns();
ArrayList<FieldSchema> cols = new ArrayList<FieldSchema>(columns.size());
for (Column eachField : columns) {
cols.add(
new FieldSchema(
eachField.getSimpleName(),
HiveCatalogUtil.getHiveFieldType(eachField.getDataType()),
""));
}
sd.setCols(cols);
// set partition keys
if (tableDesc.hasPartition()
&& tableDesc.getPartitionMethod().getPartitionType().equals(PartitionType.COLUMN)) {
List<FieldSchema> partitionKeys = new ArrayList<FieldSchema>();
for (Column eachPartitionKey :
tableDesc.getPartitionMethod().getExpressionSchema().getRootColumns()) {
partitionKeys.add(
new FieldSchema(
eachPartitionKey.getSimpleName(),
HiveCatalogUtil.getHiveFieldType(eachPartitionKey.getDataType()),
""));
}
table.setPartitionKeys(partitionKeys);
}
if (tableDesc.getMeta().getStoreType().equalsIgnoreCase(BuiltinStorages.RCFILE)) {
String serde = tableDesc.getMeta().getOption(StorageConstants.RCFILE_SERDE);
sd.setInputFormat(org.apache.hadoop.hive.ql.io.RCFileInputFormat.class.getName());
sd.setOutputFormat(org.apache.hadoop.hive.ql.io.RCFileOutputFormat.class.getName());
if (StorageConstants.DEFAULT_TEXT_SERDE.equals(serde)) {
sd.getSerdeInfo()
.setSerializationLib(
org.apache.hadoop.hive.serde2.columnar.ColumnarSerDe.class.getName());
} else {
sd.getSerdeInfo()
.setSerializationLib(
org.apache.hadoop.hive.serde2.columnar.LazyBinaryColumnarSerDe.class.getName());
}
if (tableDesc.getMeta().getOptions().containsKey(StorageConstants.RCFILE_NULL)) {
table.putToParameters(
serdeConstants.SERIALIZATION_NULL_FORMAT,
StringEscapeUtils.unescapeJava(
tableDesc.getMeta().getOption(StorageConstants.RCFILE_NULL)));
}
} else if (tableDesc.getMeta().getStoreType().equals(BuiltinStorages.TEXT)) {
sd.getSerdeInfo()
.setSerializationLib(
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.class.getName());
sd.setInputFormat(org.apache.hadoop.mapred.TextInputFormat.class.getName());
sd.setOutputFormat(
org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat.class.getName());
String fieldDelimiter =
tableDesc
.getMeta()
.getOption(
StorageConstants.TEXT_DELIMITER, StorageConstants.DEFAULT_FIELD_DELIMITER);
// User can use an unicode for filed delimiter such as \u0001, \001.
// In this case, java console will convert this value into "\\u001".
// And hive will un-espace this value again.
// As a result, user can use right field delimiter.
// So, we have to un-escape this value.
sd.getSerdeInfo()
.putToParameters(
serdeConstants.SERIALIZATION_FORMAT,
StringEscapeUtils.unescapeJava(fieldDelimiter));
sd.getSerdeInfo()
.putToParameters(
serdeConstants.FIELD_DELIM, StringEscapeUtils.unescapeJava(fieldDelimiter));
table.getParameters().remove(StorageConstants.TEXT_DELIMITER);
if (tableDesc.getMeta().containsOption(StorageConstants.TEXT_NULL)) {
table.putToParameters(
serdeConstants.SERIALIZATION_NULL_FORMAT,
StringEscapeUtils.unescapeJava(
tableDesc.getMeta().getOption(StorageConstants.TEXT_NULL)));
table.getParameters().remove(StorageConstants.TEXT_NULL);
}
} else if (tableDesc
.getMeta()
.getStoreType()
.equalsIgnoreCase(BuiltinStorages.SEQUENCE_FILE)) {
String serde = tableDesc.getMeta().getOption(StorageConstants.SEQUENCEFILE_SERDE);
sd.setInputFormat(org.apache.hadoop.mapred.SequenceFileInputFormat.class.getName());
sd.setOutputFormat(
org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat.class.getName());
if (StorageConstants.DEFAULT_TEXT_SERDE.equals(serde)) {
sd.getSerdeInfo()
.setSerializationLib(
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.class.getName());
String fieldDelimiter =
tableDesc
.getMeta()
.getOption(
StorageConstants.SEQUENCEFILE_DELIMITER,
StorageConstants.DEFAULT_FIELD_DELIMITER);
// User can use an unicode for filed delimiter such as \u0001, \001.
// In this case, java console will convert this value into "\\u001".
// And hive will un-espace this value again.
// As a result, user can use right field delimiter.
// So, we have to un-escape this value.
sd.getSerdeInfo()
.putToParameters(
serdeConstants.SERIALIZATION_FORMAT,
StringEscapeUtils.unescapeJava(fieldDelimiter));
sd.getSerdeInfo()
.putToParameters(
serdeConstants.FIELD_DELIM, StringEscapeUtils.unescapeJava(fieldDelimiter));
table.getParameters().remove(StorageConstants.SEQUENCEFILE_DELIMITER);
} else {
sd.getSerdeInfo()
.setSerializationLib(
org.apache.hadoop.hive.serde2.lazybinary.LazyBinarySerDe.class.getName());
}
if (tableDesc.getMeta().containsOption(StorageConstants.SEQUENCEFILE_NULL)) {
table.putToParameters(
serdeConstants.SERIALIZATION_NULL_FORMAT,
StringEscapeUtils.unescapeJava(
tableDesc.getMeta().getOption(StorageConstants.SEQUENCEFILE_NULL)));
table.getParameters().remove(StorageConstants.SEQUENCEFILE_NULL);
}
} else {
if (tableDesc.getMeta().getStoreType().equalsIgnoreCase(BuiltinStorages.PARQUET)) {
sd.setInputFormat(parquet.hive.DeprecatedParquetInputFormat.class.getName());
sd.setOutputFormat(parquet.hive.DeprecatedParquetOutputFormat.class.getName());
sd.getSerdeInfo()
.setSerializationLib(parquet.hive.serde.ParquetHiveSerDe.class.getName());
} else {
throw new UnsupportedException(
tableDesc.getMeta().getStoreType() + " in HivecatalogStore");
}
}
sd.setSortCols(new ArrayList<Order>());
table.setSd(sd);
client.getHiveClient().createTable(table);
} catch (Throwable t) {
throw new TajoInternalError(t);
} finally {
if (client != null) client.release();
}
}
public static String unescapeJava(String string) {
return string == null ? null : StringEscapeUtils.unescapeJava(string);
}
