/**
* Prase downloaded uniSTS data source file
*
* @param file Name of the file to parse
* @throws FatalException Throws exception if error occurs during parsing
*/
public void parse(FileInfo file) throws FatalException {
/**
* UniStsParser parser deals with single file only. So just pick up the first file name from the
* list of files
*/
String fileName = (String) file.getFiles().firstElement();
System.out.println("UNISTS parsing started");
Logger.log(" UniSts::parsing started " + fileName, Logger.INFO);
boolean noAlias = false;
/**
* Based on the name of source file to parse detect whether the aliase or base unists file is
* getting parsed. This can be detected by checking whether the filename has "alias" keyword in
* its name. It indicates that the file name is having information about unists aliases and
* hence will populate unists_alias table and not the other uniSTS base tables. Below function
* returns true or false based on whether the file is for UniSTS Alias or not.
*/
noAlias = findFormat(fileName);
/**
* call initialise table method to set the metadata information about each of the used table.
* the information will be present with DBManager
*/
initializeTables(noAlias);
Logger.log("initilalise tables over", Logger.INFO);
/**
* Initialise file writers for each of the file which will correspond to each of the table being
* populated. The file writers will be stored in m_fileWriterHashTable present with the base
* class parser. Later on when writeRecordToDb method is called file writer for the required
* table is obtained from the HashTable and record is written there
*/
createFileWriters(noAlias);
Logger.log("Create file writers over", Logger.INFO);
/**
* Create required record objects based on what(UniSTS/UniSTS alias) file is getting parsed.
* These records will later hold the field information which is to be written to file for
* loading into the database
*/
createRecords(noAlias);
Logger.log("create records over", Logger.INFO);
/**
* This function will be called to write metadata information in each of the file where records
* will be written later. This file will be input to sqlloader and mysqlimport. sqlloader
* requires the table meat information to be present in the data file. This is done by the below
* function before we start writing records to the file
*/
writeMETADATA();
Logger.log("Write metadata over", Logger.INFO);
try {
while ((m_line = getNextRecord()) != null) {
if (false == m_line.startsWith("#")) {
parseLine(noAlias);
try {
if (true == noAlias) {
writeRecordToDb(Variables.uniStsBaseTableName, uniStsRecord);
String strAccNos = uniStsAccNoRecord.fields[2].toString();
StringTokenizer sTok = new StringTokenizer(strAccNos, ";");
while (sTok.hasMoreTokens()) {
String strAccNo = sTok.nextToken();
uniStsAccNoRecord.fields[2].setLength(0);
uniStsAccNoRecord.fields[2].append(strAccNo);
writeRecordToDb(Variables.uniStsAccessionTableName, uniStsAccNoRecord);
}
}
} catch (InsertException ie) {
Logger.log("Exception in UniSTS parser :" + ie.getMessage(), Logger.DEBUG);
}
/** current record is completed, reset various table records */
resetRecords(noAlias);
}
}
} catch (IOException ioex) {
/** io exception occured - remaining records will be skipped ? ? */
Logger.log("Exception in UniSTS parser :" + ioex.getMessage(), Logger.DEBUG);
throw new FatalException(ioex.getMessage());
} finally {
Logger.log(" UNISTS::parsing over. ", Logger.INFO);
}
/**
* set the variable representing the unists last modified value in the revision_history table
*/
Variables.UniSTSRevisionHistory = getFileRevisionHistory(fileName);
}
/**
* Method to parse file
*
* @param file Information of the file to parse
* @exception FatalException throws exception if error during parsing
*/
public void parse(FileInfo file) throws FatalException {
/**
* JDOM parser deals with single file only. So just pick up the first file name from the list of
* files
*/
String fileName = (String) file.getFiles().firstElement();
String oldFileName = fileName;
/** JDOM element scanner for the xml file */
ElementScanner xmlScanner;
Logger.log("Starting parsing for " + m_dbType + ": " + fileName, Logger.INFO);
ElementListener xmlListener = null;
try {
long startParsingTime = System.currentTimeMillis();
boolean history = false;
history = findFormat(fileName);
/** Check if file to be parsed is a Gene-History file */
if (false == history) {
xmlScanner = new ElementScanner();
xmlScanner.setValidation(false);
/** add the listener for the individual nodes according the the dbType specified. */
if (m_dbType.equalsIgnoreCase(Constants.ENTREZGENE)) {
String convertedFileName = null;
if (fileName.endsWith(".xml") == false) {
convertedFileName = ConvertToXml(fileName);
fileName = convertedFileName;
} else {
convertedFileName = fileName;
}
Logger.log("Converted .asn file to XML " + convertedFileName, Logger.WARNING);
m_parserDataObject = new EntrezParser(file, m_filesParsed, history);
/** Initialise the system_termtree and system_termdata tables. */
startSystemTreeDataFiles();
xmlListener =
new JDOMXMLListener(m_dbType, convertedFileName, ((EntrezParser) m_parserDataObject));
/** add a listener for node 'Entrezgene' in xml file */
xmlScanner.addElementListener(xmlListener, Constants.ENTREZNODE);
}
Logger.log(xmlScanner.toString(), Logger.INFO);
Logger.log("input to xmlscanner " + fileName, Logger.INFO);
/**
* scan the input file and call elementMatched function whenever the required node is found
* in the input file
*/
xmlScanner.parse(new InputSource(fileName));
/** close all the open fileWriter streams. */
m_parserDataObject.closeFileWriters();
deleteDownloadedFile(file);
}
/** Or if it is a Gene-History file */
else {
Logger.log("Parsing started for GENE_HISTORY table", Logger.INFO);
m_parserDataObject = new EntrezParser(file, m_filesParsed, history);
EntrezParser tempParser = (EntrezParser) m_parserDataObject;
while ((m_line = getNextRecord()) != null) {
StringTokenizer sTok = new StringTokenizer(m_line);
tempParser.parseGeneHistory(history, sTok);
}
m_parserDataObject.closeFileWriters();
}
Logger.log("Parsing over for " + m_dbType + ": " + fileName, Logger.INFO);
long endParsingTime = System.currentTimeMillis();
long parsingTime = endParsingTime - startParsingTime;
Logger.log("Parsing Time for: " + m_dbType + ": " + parsingTime, Logger.INFO);
} catch (NullPointerException nexcp) {
Variables.errorCount++;
Logger.log("NullPointerException has occured: " + nexcp.getMessage(), Logger.FATAL);
} catch (OutOfMemoryError outexcp) {
Variables.errorCount++;
Logger.log("OutOfMemoryError has occured: " + outexcp.getMessage(), Logger.FATAL);
} catch (JDOMException jexcp) {
Variables.errorCount++;
Logger.log("JDOMException has occured: " + jexcp.getMessage(), Logger.FATAL);
} catch (SAXException saxexcp) {
Variables.errorCount++;
Logger.log("SAXException has occured: " + saxexcp.getMessage(), Logger.FATAL);
} catch (IOException ioexcp) {
Variables.errorCount++;
Logger.log("IOException has occured: " + ioexcp.getMessage(), Logger.FATAL);
}
}