1 package compbio.statistic;
3 import java.text.ParseException;
4 import java.text.SimpleDateFormat;
5 import java.util.ArrayList;
6 import java.util.Calendar;
8 import java.util.HashMap;
12 import compbio.cassandra.DateBean;
13 import compbio.cassandra.ProteinBean;
14 import compbio.cassandra.CassandraNativeConnector;
15 import compbio.cassandra.CassandraReaderOld;
16 import compbio.cassandra.DataBase;
17 import compbio.cassandra.Pair;
18 import compbio.cassandra.JobBean;
19 import compbio.cassandra.Total;
20 import compbio.cassandra.TotalByCounterBean;
21 import compbio.cassandra.TotalJobsStatisticBean;
22 import compbio.cassandra.UserBean;
23 import compbio.engine.JobStatus;
25 public class CassandraRequester {
26 private CassandraReaderOld db = new CassandraReaderOld();
27 private ArrayList<DataBase> query;
28 private static long currentDate = 0;
29 private static long earlestDate = 0;
30 private final static SimpleDateFormat formatYYMMDD = new SimpleDateFormat("yyyy/MM/dd");
31 private final static SimpleDateFormat formatDDMMYY = new SimpleDateFormat("dd/MM/yyyy");
34 * query: execution time for the period from date1 till date2
36 public List<DataBase> extractExecutionTime(String date1, String date2) {
43 if (!isThisDateValid(date1, formatYYMMDD) || !isThisDateValid(date2, formatYYMMDD)) {
44 System.out.println("CassandraRequester.extractExecutionTime: wrong format for date1 " + date1 + "or date2 " + date2);
49 long dateStart = DateParsing(date1, formatYYMMDD);
50 long dateEnd = DateParsing(date2, formatYYMMDD);
51 if (dateEnd < earlestDate || dateStart > currentDate || dateStart > dateEnd)
53 if (dateStart < earlestDate)
54 dateStart = earlestDate;
55 if (dateEnd > currentDate)
56 dateStart = currentDate;
58 Calendar start = Calendar.getInstance();
59 start.setTime(new Date(dateStart));
60 Calendar end = Calendar.getInstance();
61 end.setTime(new Date(dateEnd));
62 query = new ArrayList<DataBase>();
63 List<Integer> totalTime = new ArrayList<Integer>();
64 for (int i = 0; i < nbins; i++)
66 List<Pair<String, String>> res = db.ReadProteinDataTable();
67 List<Pair<Date, Long>> numres = new ArrayList<Pair<Date, Long>>();
69 for (Pair<String, String> entry : res) {
70 SimpleDateFormat dateformatter = new SimpleDateFormat("yyyy/MM/dd");
72 Date jobstartdate = dateformatter.parse(entry.getElement0());
73 long date = jobstartdate.getTime();
74 if (dateStart <= date && date <= dateEnd) {
75 SimpleDateFormat datetimeformatter = new SimpleDateFormat("yyyy/MM/dd:H:m:s");
76 Date jobstarttime = datetimeformatter.parse(entry.getElement0());
77 Date jobendtime = datetimeformatter.parse(entry.getElement1());
78 long diff = (jobendtime.getTime() - jobstarttime.getTime()) / 1000;
79 Pair<Date, Long> pair = new Pair<Date, Long>(jobstartdate, Long.valueOf(diff));
82 } catch (ParseException e) {
87 for (Date date = start.getTime(); !start.after(end); start.add(Calendar.DATE, 1), date = start.getTime()) {
88 List<Integer> timeResult = new ArrayList<Integer>();
89 for (int i = 0; i < nbins; i++)
91 for (Pair<Date, Long> p : numres) {
92 if (date.equals(p.getElement0())) {
93 long lenResult = p.getElement1().longValue();
95 timeResult.set(0, timeResult.get(0) + 1);
96 else if (lenResult > 30 && lenResult <= 60)
97 timeResult.set(1, timeResult.get(1) + 1);
98 else if (lenResult > 60 && lenResult <= 120)
99 timeResult.set(2, timeResult.get(2) + 1);
100 else if (lenResult > 120 && lenResult <= 600)
101 timeResult.set(3, timeResult.get(3) + 1);
103 timeResult.set(4, timeResult.get(4) + 1);
107 for (int i = 0; i < nbins; i++)
108 totalTime.set(i, totalTime.get(i) + timeResult.get(i));
109 DataBase db = new DataBase();
110 db.setTimeRez(timeResult);
111 db.setDate(DateFormat(date.getTime()));
115 DataBase db = new DataBase();
116 db.setTimeTotalExec(totalTime);
122 * query: total number of jobs for the period from date1 till date2
124 public TotalJobsStatisticBean countJobs(String date1, String date2) {
125 /* if (null == date1) {
131 if (!isThisDateValid(date1, formatYYMMDD) || !isThisDateValid(date2, formatYYMMDD)) {
132 System.out.println("CassandraRequester.countJobs: wrong format for date1 " + date1 + "or date2 " + date2);
136 long dateStart = DateParsing(date1, formatYYMMDD);
137 long dateEnd = DateParsing(date2, formatYYMMDD);
138 /* if (dateEnd < earlestDate || dateStart > currentDate || dateStart > dateEnd)
140 if (dateStart < earlestDate)
141 dateStart = earlestDate;
142 if (dateEnd > currentDate)
143 dateStart = currentDate;*/
144 Calendar start = Calendar.getInstance();
145 start.setTime(new Date(dateStart));
146 Calendar end = Calendar.getInstance();
147 end.setTime(new Date(dateEnd));
148 TotalJobsStatisticBean query = new TotalJobsStatisticBean();
149 Total wholeTotal = new Total(0, 0, 0, 0, 0);
150 for (Date date = start.getTime(); !start.after(end); start.add(Calendar.DATE, 1), date = start.getTime()) {
151 Total res = db.ReadDateTable(date.getTime());
154 query.setDateTotal(DateFormat(date.getTime()), res);
155 wholeTotal.setTotal(res.getTotal() + wholeTotal.getTotal());
156 wholeTotal.setTotalOK(res.getTotalOK() + wholeTotal.getTotalOK());
157 wholeTotal.setTotalStopped(res.getTotalStopped() + wholeTotal.getTotalStopped());
158 wholeTotal.setTotalError(res.getTotalError() + wholeTotal.getTotalError());
159 wholeTotal.setTotalTimeOut(res.getTotalTimeOut() + wholeTotal.getTotalTimeOut());
161 query.setWholeTotal(wholeTotal);
166 * query: jobs and sequence at date
168 public DateBean readJobByDay(String date, JobStatus status) {
169 if (!isThisDateValid(date, formatDDMMYY)) {
170 System.out.println("CassandraRequester.readJobByDay: Wrong date format for " + date);
174 long day = DateParsing(date, formatDDMMYY);
175 if (day < earlestDate || day > currentDate)
178 if (status == JobStatus.OK) {
179 return db.ReadProteinData(day, date);
181 return db.ReadFailedJobs(day, date, status);
185 * query: protein sequence
187 public List<ProteinBean> readProteins(String protIn, String searchtype) {
188 List<ProteinBean> result;
189 if (searchtype.equals("whole"))
190 result = db.ReadWholeSequence(protIn);
192 result = db.ReadPartOfSequence(protIn);
196 if (searchtype.equals("partial")) {
197 for (ProteinBean entry : result) {
198 entry.setSubProt(CreateSubprot(entry.getSequence(), protIn));
205 * query: protein feature
207 public Map<String, String> readProteinsPrediction(String feature, int percent) {
208 Map<String, String> result = db.ReadProtein();;
211 Map<String, String> query = new HashMap<String, String>();
212 for (Map.Entry<String, String> entry : result.entrySet()) {
213 String pred = entry.getValue();
214 if (pred.replaceAll("[^"+feature+"]", "").length() > pred.length() * percent / 100 && (!entry.getKey().equals(""))) {
215 // if (!entry.getKey().equals(""))
216 query.put(entry.getKey(), pred);
223 * query protein sequences with number of jobs
225 public List<TotalByCounterBean> readProteinByCounter(int minimalcounter) {
226 List<TotalByCounterBean> query = new ArrayList<TotalByCounterBean>();
227 Map<String, Integer> map = db.ReadProteinSequenceByCounter();
230 for (Map.Entry<String, Integer> entry : map.entrySet()) {
231 if (entry.getValue() > minimalcounter) {
232 TotalByCounterBean bean = new TotalByCounterBean();
233 bean.setTotaljobs(entry.getValue());
234 bean.setName(entry.getKey());
242 * query ip with number of jobs
244 public List<TotalByCounterBean> readIpByCounter(Integer minimalcounter) {
245 List<TotalByCounterBean> query = new ArrayList<TotalByCounterBean>();
246 Map<String, Integer> map = db.ReadIpByCounter();
247 if (minimalcounter == null)
251 for (Map.Entry<String, Integer> entry : map.entrySet()) {
252 if (entry.getValue() > minimalcounter) {
253 TotalByCounterBean bean = new TotalByCounterBean();
254 bean.setTotaljobs(entry.getValue());
255 bean.setName(entry.getKey());
263 * query jobs log info
265 public JobBean readJobLog(String jobid) {
268 return db.ReadJobLog(jobid);
272 * query jobs by ipStructureJobLog
274 public UserBean readIp(String ip) {
277 Map<String, String[]> res = db.ReadIpWithJobs(ip);
280 UserBean query = new UserBean(ip);
281 query.setMainInfo(res);
286 * create list of parts of protein sequence;
288 private static List<String> CreateSubprot(String protein, String subprot) {
289 List<String> sub = new ArrayList<String>();
290 String subStr = protein;
291 while (subStr.length() > 0 && subStr.contains(subprot)) {
292 String first = subStr.substring(0, subStr.indexOf(subprot));
293 if (first.length() > 0)
296 subStr = subStr.substring(subStr.indexOf(subprot) + subprot.length(), subStr.length());
298 if (subStr.length() > 0)
304 * convert String date into long date (miliseconds since the epoch start)
306 private static long DateParsing(String datInput, SimpleDateFormat formatter) {
307 if (datInput == null) {
313 dateWorkSt = formatter.parse(datInput).getTime();
314 } catch (ParseException e) {
320 // convert long to date in string format
321 private static String DateFormat(long inDate) {
322 SimpleDateFormat datformat = new SimpleDateFormat("dd/MM/yyyy");
323 return datformat.format(new Date(inDate));
327 * set earlest date and current dates. earlestDate is static and should be
328 * set at the 1st call currentDate should be re-calculated every time
330 private static void SetDateRange() {
331 Calendar cal = Calendar.getInstance();
332 currentDate = DateParsing(cal.get(Calendar.YEAR) + "/" + (cal.get(Calendar.MONTH) + 1) + "/" + cal.get(Calendar.DAY_OF_MONTH),
334 if (0 == earlestDate) {
335 CassandraRequester cr = new CassandraRequester();
336 earlestDate = cr.earliestDate();
340 public boolean isThisDateValid(String dateToValidate, SimpleDateFormat sdf) {
341 if (dateToValidate == null || dateToValidate.equals("")) {
345 // if not valid, this will throw ParseException
346 sdf.setLenient(false);
347 Date date = sdf.parse(dateToValidate);
348 } catch (ParseException e) {
356 * find the earliest date in the database
358 public long earliestDate() {
359 earlestDate = CassandraNativeConnector.getEarliestDateInDB();