UserCode/RateMonShiftTool_dev/MenuAnalyzer.py

#!/usr/bin/env python

import os
import sys
from DatabaseParser import ConnectDB
import re
import cx_Oracle

try:  ## set is builtin in python 2.6.4 and sets is deprecated
    set
except NameError:
    from sets import Set

class MenuAnalyzer:
    def __init__(self,name):
        ##default ranges
        self.maxModuleNameLength = 300
        self.maxModulesPerPath   = 50
        self.maxPaths = 500
        self.maxEndPaths = 30

        ##required streams
        self.requiredStreamsAndPDs = { 'Calibration' : ['TestEnablesEcalHcalDT'],'EcalCalibration' : ['EcalLaser'],
                                  'TrackerCalibration' : ['TestEnablesTracker']}
        self.ExpressStreamName = 'Express'
        self.expressPDs      = { 'ExpressPhysics' : 'Collisions',
                                 'ExpressCosmics' : 'Cosmics' }

        self.expressType = ''

        self.menuName = name
        self.endPathList = set()
        self.perPathModuleList={}
        self.perModuleTypeList={}
        self.perStreamPDList={}
        self.perPDPathList={}
        self.Results={}
        self.ModuleList=[]
        self.AnalysisList=[]
        

        ## statically define the analysis map: new analyses must be registered here
        self.AnalysisMap = {
            'moduleLength' : self.checkModuleLength,
            'numberOfPaths' : self.checkNumPaths,
            'numberOfEndPaths' : self.checkNumEndPaths,
            'reqStreamsAndPDs' : self.reqStreamsAndPDs,
            'checkExpress' : self.checkExpress,
            'checkNameFormats' :self.checkNameFormats
            }

        self.T0REGEXP = { ## These are the regexps that T0 uses to access things
            # Regexp for save file name paths. We don't process anything else.
            'RXSAFEPATH' : re.compile("^[-A-Za-z0-9_]+$"),
            # Regexp for valid dataset names.
            'RXDATASET' : re.compile("^[-A-Za-z0-9_]+$"),
            # Regexp for valid RelVal dataset names.
            'RXRELVALMC' : re.compile("^/RelVal[^/]+/(CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)[-_].*$"),
            'RXRELVALDATA' : re.compile("^/[^/]+/(CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)[-_].*$"),
            # Regexp for online DQM files.
            'RXONLINE' : re.compile("^(?:.*/)?DQM_V(\d+)(_[A-Za-z0-9]+)?_R(\d+)\.root$"),
            # Regexp for offline DQM files.
            'RXOFFLINE' : re.compile("^(?:.*/)?DQM_V(\d+)_R(\d+)((?:__[-A-Za-z0-9_]+){3})\.root$"),
            # Regexp for acquisition era part of the processed dataset name.
            'RXERA' : re.compile("^([A-Za-z]+\d+|CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)")
            }

    def AddAnalysis(self,name): 
        self.AnalysisList.append(name)

    def AddAllAnalyses(self):
        for name in self.AnalysisMap.iterkeys():
            self.AddAnalysis(name)

    def Analyze(self):
        cursor = ConnectDB('hlt')
        self.GetModules(cursor)
        self.GetStreamsPathsPDs(cursor)
        for analysis in self.AnalysisList: 
            if not self.AnalysisMap.has_key(analysis):
                print "ERROR: Analysis %s not defined" % (analysis,)
                continue
            self.AnalysisMap[analysis]()
            
        
    def checkModuleLength(self):
        self.Results['moduleLength'] = []
        for modName,type in self.perModuleTypeList.iteritems():
            if len(modName) > self.maxModuleNameLength: self.Results['moduleLength'].append(modName)
        
    def checkNumPaths(self):
        if len(self.perPathModuleList) > self.maxPaths:
            self.Results['numberOfPaths'] = len(self.perPathModuleList) 
        else:
            self.Results['numberOfPaths'] = 0
    def checkNumEndPaths(self):
        if len(self.endPathList) > self.maxEndPaths:
            self.Results['numberOfEndPaths'] = len(self.endPathList)
        else:
            self.Results['numberOfEndPaths'] = 0
    def reqStreamsAndPDs(self):
        self.Results['reqStreamsAndPDs'] = []
        for stream,PDList in self.requiredStreamsAndPDs.iteritems():
            if not self.perStreamPDList.has_key(stream): self.Results['reqStreamsAndPDs'].append(stream)
            for PD in PDList:
                if not PD in self.requiredStreamsAndPDs[stream]: self.Results['reqStreamsAndPDs'].append(stream+'::'+PD)
    
    def checkExpress(self):
        self.Results['checkExpress'] = []
        if not self.perStreamPDList.has_key(self.ExpressStreamName):
            self.Results['checkExpress'].append(self.ExpressStreamName)
            return

        if len(self.perStreamPDList[self.ExpressStreamName]) >1:
            self.Results['checkExpress'].append("MULTIPLE_PDS")
        if len(self.perStreamPDList[self.ExpressStreamName]) <1:
            self.Results['checkExpress'].append("NO_PDS")

        for PD in self.perStreamPDList[self.ExpressStreamName]:
            if not self.expressPDs.has_key(PD):
                self.Results['checkExpress'].append(self.ExpressStreamName+"::"+PD)
            else:
                self.expressType = self.expressPDs[PD]
                
        
    def checkNameFormats(self):
        self.Results['checkNameFormats']=[]
        for PD in self.perPDPathList.iterkeys():
            if not self.T0REGEXP['RXDATASET'].match(PD):
                self.Results['checkNameFormats'].append(PD)
        for path in self.perPathModuleList.iterkeys():
            if not self.T0REGEXP['RXSAFEPATH'].match(path):
                self.Results['checkNameFormats'].append(path)
        
    def GetModules(self,cursor):
        sqlquery ="""  
        SELECT I.NAME,E.NAME,D.NAME,I.ISENDPATH
        FROM
        CMS_HLT.PARAMETERS B,
        CMS_HLT.SUPERIDPARAMETERASSOC C,
        CMS_HLT.MODULETEMPLATES D,
        CMS_HLT.MODULES E,
        CMS_HLT.PATHMODULEASSOC F,
        CMS_HLT.CONFIGURATIONPATHASSOC G,
        CMS_HLT.CONFIGURATIONS H,
        CMS_HLT.PATHS I
        WHERE
        B.PARAMID = C.PARAMID AND
        C.SUPERID = F.MODULEID AND
        E.TEMPLATEID = D.SUPERID AND
        F.MODULEID = E.SUPERID AND
        F.PATHID=G.PATHID AND
        I.PATHID=G.PATHID AND
        G.CONFIGID=H.CONFIGID AND
        H.CONFIGDESCRIPTOR='%s' 
        """ % (self.menuName,)
        
        cursor.execute(sqlquery)
        for PathName,ModuleName,ModuleType,endPath in cursor.fetchall():
            if not self.perPathModuleList.has_key(PathName): self.perPathModuleList[PathName] = []
            self.perPathModuleList[PathName].append(ModuleName)
            self.perModuleTypeList[ModuleName] = ModuleType
            if endPath: self.endPathList.add(PathName)


    def GetStreamsPathsPDs(self,cursor):
        sqlquery= """
        SELECT A.STREAMLABEL,E.NAME,F.DATASETLABEL
        FROM
        CMS_HLT.STREAMS A,
        CMS_HLT.CONFIGURATIONS B,
        CMS_HLT.CONFIGURATIONPATHASSOC C,
        CMS_HLT.PATHSTREAMDATASETASSOC D,
        CMS_HLT.PATHS E,
        CMS_HLT.PRIMARYDATASETS F
        WHERE
        B.CONFIGDESCRIPTOR='%s' AND
        C.CONFIGID=B.CONFIGID AND
        D.PATHID=C.PATHID AND
        A.STREAMID=D.STREAMID AND
        E.PATHID = C.PATHID AND
        F.DATASETID = D.DATASETID
        """ % (self.menuName,)
        
        cursor.execute(sqlquery)
        for StreamName,PathName,PDName in cursor.fetchall():
            if not self.perStreamPDList.has_key(StreamName): self.perStreamPDList[StreamName] = []
            if not PDName in self.perStreamPDList[StreamName]: self.perStreamPDList[StreamName].append(PDName)
            if not self.perPDPathList.has_key(PDName): self.perPDPathList[PDName] = []
            self.perPDPathList[PDName].append(PathName)

Revision:	1.1
Committed:	Thu Jul 12 16:04:24 2012 UTC (12 years, 9 months ago) by amott
Content type:	text/x-python
Branch:	MAIN
Log Message:	initial versions of tools to check menu consistencies
#	User	Rev	Content
1	amott	1.1	#!/usr/bin/env python
2
3			import os
4			import sys
5			from DatabaseParser import ConnectDB
6			import re
7			import cx_Oracle
8
9			try: ## set is builtin in python 2.6.4 and sets is deprecated
10			set
11			except NameError:
12			from sets import Set
13
14			class MenuAnalyzer:
15			def __init__(self,name):
16			##default ranges
17			self.maxModuleNameLength = 300
18			self.maxModulesPerPath = 50
19			self.maxPaths = 500
20			self.maxEndPaths = 30
21
22			##required streams
23			self.requiredStreamsAndPDs = { 'Calibration' : ['TestEnablesEcalHcalDT'],'EcalCalibration' : ['EcalLaser'],
24			'TrackerCalibration' : ['TestEnablesTracker']}
25			self.ExpressStreamName = 'Express'
26			self.expressPDs = { 'ExpressPhysics' : 'Collisions',
27			'ExpressCosmics' : 'Cosmics' }
28
29			self.expressType = ''
30
31			self.menuName = name
32			self.endPathList = set()
33			self.perPathModuleList={}
34			self.perModuleTypeList={}
35			self.perStreamPDList={}
36			self.perPDPathList={}
37			self.Results={}
38			self.ModuleList=[]
39			self.AnalysisList=[]
40
41
42			## statically define the analysis map: new analyses must be registered here
43			self.AnalysisMap = {
44			'moduleLength' : self.checkModuleLength,
45			'numberOfPaths' : self.checkNumPaths,
46			'numberOfEndPaths' : self.checkNumEndPaths,
47			'reqStreamsAndPDs' : self.reqStreamsAndPDs,
48			'checkExpress' : self.checkExpress,
49			'checkNameFormats' :self.checkNameFormats
50			}
51
52			self.T0REGEXP = { ## These are the regexps that T0 uses to access things
53			# Regexp for save file name paths. We don't process anything else.
54			'RXSAFEPATH' : re.compile("^[-A-Za-z0-9_]+$"),
55			# Regexp for valid dataset names.
56			'RXDATASET' : re.compile("^[-A-Za-z0-9_]+$"),
57			# Regexp for valid RelVal dataset names.
58			'RXRELVALMC' : re.compile("^/RelVal[^/]+/(CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)[-_].*$"),
59			'RXRELVALDATA' : re.compile("^/[^/]+/(CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)[-_].*$"),
60			# Regexp for online DQM files.
61			'RXONLINE' : re.compile("^(?:.*/)?DQM_V(\d+)(_[A-Za-z0-9]+)?_R(\d+)\.root$"),
62			# Regexp for offline DQM files.
63			'RXOFFLINE' : re.compile("^(?:.*/)?DQM_V(\d+)_R(\d+)((?:__[-A-Za-z0-9_]+){3})\.root$"),
64			# Regexp for acquisition era part of the processed dataset name.
65			'RXERA' : re.compile("^([A-Za-z]+\d+\|CMSSW(?:_[0-9]+)+(?:_pre[0-9]+)?)")
66			}
67
68			def AddAnalysis(self,name):
69			self.AnalysisList.append(name)
70
71			def AddAllAnalyses(self):
72			for name in self.AnalysisMap.iterkeys():
73			self.AddAnalysis(name)
74
75			def Analyze(self):
76			cursor = ConnectDB('hlt')
77			self.GetModules(cursor)
78			self.GetStreamsPathsPDs(cursor)
79			for analysis in self.AnalysisList:
80			if not self.AnalysisMap.has_key(analysis):
81			print "ERROR: Analysis %s not defined" % (analysis,)
82			continue
83			self.AnalysisMap[analysis]()
84
85
86
87			def checkModuleLength(self):
88			self.Results['moduleLength'] = []
89			for modName,type in self.perModuleTypeList.iteritems():
90			if len(modName) > self.maxModuleNameLength: self.Results['moduleLength'].append(modName)
91
92			def checkNumPaths(self):
93			if len(self.perPathModuleList) > self.maxPaths:
94			self.Results['numberOfPaths'] = len(self.perPathModuleList)
95			else:
96			self.Results['numberOfPaths'] = 0
97			def checkNumEndPaths(self):
98			if len(self.endPathList) > self.maxEndPaths:
99			self.Results['numberOfEndPaths'] = len(self.endPathList)
100			else:
101			self.Results['numberOfEndPaths'] = 0
102			def reqStreamsAndPDs(self):
103			self.Results['reqStreamsAndPDs'] = []
104			for stream,PDList in self.requiredStreamsAndPDs.iteritems():
105			if not self.perStreamPDList.has_key(stream): self.Results['reqStreamsAndPDs'].append(stream)
106			for PD in PDList:
107			if not PD in self.requiredStreamsAndPDs[stream]: self.Results['reqStreamsAndPDs'].append(stream+'::'+PD)
108
109			def checkExpress(self):
110			self.Results['checkExpress'] = []
111			if not self.perStreamPDList.has_key(self.ExpressStreamName):
112			self.Results['checkExpress'].append(self.ExpressStreamName)
113			return
114
115			if len(self.perStreamPDList[self.ExpressStreamName]) >1:
116			self.Results['checkExpress'].append("MULTIPLE_PDS")
117			if len(self.perStreamPDList[self.ExpressStreamName]) <1:
118			self.Results['checkExpress'].append("NO_PDS")
119
120			for PD in self.perStreamPDList[self.ExpressStreamName]:
121			if not self.expressPDs.has_key(PD):
122			self.Results['checkExpress'].append(self.ExpressStreamName+"::"+PD)
123			else:
124			self.expressType = self.expressPDs[PD]
125
126
127
128			def checkNameFormats(self):
129			self.Results['checkNameFormats']=[]
130			for PD in self.perPDPathList.iterkeys():
131			if not self.T0REGEXP['RXDATASET'].match(PD):
132			self.Results['checkNameFormats'].append(PD)
133			for path in self.perPathModuleList.iterkeys():
134			if not self.T0REGEXP['RXSAFEPATH'].match(path):
135			self.Results['checkNameFormats'].append(path)
136
137			def GetModules(self,cursor):
138			sqlquery ="""
139			SELECT I.NAME,E.NAME,D.NAME,I.ISENDPATH
140			FROM
141			CMS_HLT.PARAMETERS B,
142			CMS_HLT.SUPERIDPARAMETERASSOC C,
143			CMS_HLT.MODULETEMPLATES D,
144			CMS_HLT.MODULES E,
145			CMS_HLT.PATHMODULEASSOC F,
146			CMS_HLT.CONFIGURATIONPATHASSOC G,
147			CMS_HLT.CONFIGURATIONS H,
148			CMS_HLT.PATHS I
149			WHERE
150			B.PARAMID = C.PARAMID AND
151			C.SUPERID = F.MODULEID AND
152			E.TEMPLATEID = D.SUPERID AND
153			F.MODULEID = E.SUPERID AND
154			F.PATHID=G.PATHID AND
155			I.PATHID=G.PATHID AND
156			G.CONFIGID=H.CONFIGID AND
157			H.CONFIGDESCRIPTOR='%s'
158			""" % (self.menuName,)
159
160			cursor.execute(sqlquery)
161			for PathName,ModuleName,ModuleType,endPath in cursor.fetchall():
162			if not self.perPathModuleList.has_key(PathName): self.perPathModuleList[PathName] = []
163			self.perPathModuleList[PathName].append(ModuleName)
164			self.perModuleTypeList[ModuleName] = ModuleType
165			if endPath: self.endPathList.add(PathName)
166
167
168			def GetStreamsPathsPDs(self,cursor):
169			sqlquery= """
170			SELECT A.STREAMLABEL,E.NAME,F.DATASETLABEL
171			FROM
172			CMS_HLT.STREAMS A,
173			CMS_HLT.CONFIGURATIONS B,
174			CMS_HLT.CONFIGURATIONPATHASSOC C,
175			CMS_HLT.PATHSTREAMDATASETASSOC D,
176			CMS_HLT.PATHS E,
177			CMS_HLT.PRIMARYDATASETS F
178			WHERE
179			B.CONFIGDESCRIPTOR='%s' AND
180			C.CONFIGID=B.CONFIGID AND
181			D.PATHID=C.PATHID AND
182			A.STREAMID=D.STREAMID AND
183			E.PATHID = C.PATHID AND
184			F.DATASETID = D.DATASETID
185			""" % (self.menuName,)
186
187			cursor.execute(sqlquery)
188			for StreamName,PathName,PDName in cursor.fetchall():
189			if not self.perStreamPDList.has_key(StreamName): self.perStreamPDList[StreamName] = []
190			if not PDName in self.perStreamPDList[StreamName]: self.perStreamPDList[StreamName].append(PDName)
191			if not self.perPDPathList.has_key(PDName): self.perPDPathList[PDName] = []
192			self.perPDPathList[PDName].append(PathName)
193