Mercurial > repos > yufei-luo > s_mart

comparison SMART/DiffExpAnal/countNumber_parallel_unSQL.py @ 18:94ab73e8a190

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author m-zytnicki
date Mon, 29 Apr 2013 03:20:15 -0400
parents
children
comparison
equal deleted inserted replaced
17:b0e8584489e6 18:94ab73e8a190
1 #! /usr/bin/env python
2
3
4 import optparse, os, sys, tarfile, random
5 from optparse import OptionParser
6 from commons.core.launcher.Launcher import Launcher
7 from commons.core.sql.TableJobAdaptatorFactory import TableJobAdaptatorFactory
8
9 def stop_err(msg):
10 sys.stderr.write('%s\n' % msg)
11 sys.exit()
12
13 def toTar(tarFileName, outCountNames):
14 dir = os.path.dirname(tarFileName)
15 tfile = tarfile.open(tarFileName + ".tmp.tar", "w")
16 currentPath = os.getcwd()
17 os.chdir(dir)
18 for file in outCountNames:
19 relativeFileName = os.path.basename(file)
20 tfile.add(relativeFileName)
21 os.system("mv %s %s" % (tarFileName + ".tmp.tar", tarFileName))
22 tfile.close()
23 os.chdir(currentPath)
24
25 def _map(iLauncher, cmd, cmdStart, cmdFinish ):
26 lCmds = []
27 lCmds.append(cmd)
28 lCmdStart = []
29 lCmdStart.append(cmdStart)
30 lCmdFinish = []
31 lCmdFinish.append(cmdFinish)
32 return(iLauncher.prepareCommands_withoutIndentation(lCmds, lCmdStart, lCmdFinish))
33
34 def _createCountNumberCommand(iLauncher, inputFile, outputFile):
35 lArgs = []
36 lArgs.append("%s" % inputFile)
37 lArgs.append("%s" % outputFile)
38 return iLauncher.getSystemCommand("perl %s/SMART/DiffExpAnal/countNumber.pl " % os.environ["REPET_PATH"], lArgs)
39
40 def __main__():
41 #Parse Command Line
42 parser = optparse.OptionParser()
43 parser.add_option("-i", "--input", dest="inputFile", help="input txt file, a list of overlapping results files.")
44 parser.add_option("-o", "--output", dest="outputFile", help="Out txt file.")
45 parser.add_option("-t", "--tar", dest="outputTar", default=None, help="output all count results in a tar file.")
46 (options, args) = parser.parse_args()
47
48 #Parse the input txt file and read a list of transcripts files.
49 file = open(options.inputFile, "r")
50 lines = file.readlines()
51 inputFileNames = []
52 outCountNames = []
53 outputName = options.outputFile
54 resDirName = os.path.dirname(outputName) + '/'
55
56 #Write output txt file and define all output count file names
57 out = open(outputName, "w")
58 out.write("label\tfiles\tgroup\n")
59 for line in lines:
60 tab = line.split()
61 inputFileNames.append(tab[1])
62 outCountName = resDirName + tab[0] + "_outCount_%s.csv" % random.randrange(0, 10000)
63 outCountNames.append(outCountName)
64 out.write(tab[0] + '\t' + outCountName + '\t' + tab[0][5] + '\n')
65 file.close()
66 out.close()
67
68 #Launch on nodes
69 acronym = "countNumber"
70 jobdb = TableJobAdaptatorFactory.createJobInstance()
71 iLauncher = Launcher(jobdb, os.getcwd(), "", "", os.getcwd(), os.getcwd(), "jobs", "", acronym, acronym, False, True)
72 lCmdsTuples = []
73 for i in range(len(inputFileNames)): #Construct the lines commands
74 inputFile = inputFileNames[i]
75 outputFile = outCountNames[i]
76 cmd2Launch = _createCountNumberCommand(iLauncher, inputFile, outputFile)
77 cmdStart = ""
78 cmdFinish = ""
79 lCmdsTuples.append(_map(iLauncher, cmd2Launch, cmdStart, cmdFinish))
80
81
82
83 iLauncher.runLauncherForMultipleJobs(acronym, lCmdsTuples, True)
84
85
86
87 if options.outputTar != None:
88 toTar(options.outputTar, outCountNames)
89
90
91 if __name__=="__main__":__main__()