Functions
def	searchClassDefXml ()

def	searchDuplicatePlugins ()

Variables
	action

	args

	default

	dest

	dumpGroup

	equivDict

	help

	ignoreEdmDP

	lostDefs

	options

	parser

	type

	typedefsDict

	xmlGroup

Function Documentation

def duplicateReflexLibrarySearch.searchClassDefXml ( )

Searches through the requested directory looking at
'classes_def.xml' files looking for duplicate Reflex definitions.

Definition at line 85 of file duplicateReflexLibrarySearch.py.

References cond::persistency.search().

 def searchClassDefXml ():
     """ Searches through the requested directory looking at
     'classes_def.xml' files looking for duplicate Reflex definitions."""
     # compile necessary RE statements
     classNameRE    = re.compile (r'class\s+name\s*=\s*"([^"]*)"')
     spacesRE       = re.compile (r'\s+')
     stdRE          = re.compile (r'std::')
     srcClassNameRE = re.compile (r'(\w+)/src/classes_def.*[.]xml')
     ignoreSrcRE    = re.compile (r'.*/FWCore/Skeletons/scripts/mkTemplates/.+')
     braketRE       = re.compile (r'<.+>')
     print "Searching for 'classes_def.xml' in '%s'." % os.path.join(os.environ.get('CMSSW_BASE'),'src')
     xmlFiles = []
     for srcDir in [os.environ.get('CMSSW_BASE'),os.environ.get('CMSSW_RELEASE_BASE')]:
       if not len(srcDir): continue
       for xml in commands.getoutput ('cd '+os.path.join(srcDir,'src')+'; find . -name "*classes_def*.xml" -follow -print').split ('\n'):
         if xml and (not xml in xmlFiles):
           xmlFiles.append(xml)
     if options.showXMLs:
         pprint.pprint (xmlFiles)
     # try and figure out the names of the packages
     xmlPackages = []
     packagesREs = {}
     equivREs    = {}
     explicitREs = []
     for item in equivDict:
         for pack in item:
             for equiv in item[pack]:
                 explicitREs.append( (re.compile(r'\b' + equiv + r'\b'),pack))
     if options.lostDefs:
         for filename in xmlFiles:
             if (not filename) or (ignoreSrcRE.match(filename)): continue
             match = srcClassNameRE.search (filename)
             if not match: continue
             packageName = match.group(1)
             xmlPackages.append (packageName)
             matchString = r'\b' + packageName + r'\b'
             packagesREs[packageName] = re.compile (matchString)
             equivList = equivREs.setdefault (packageName, [])
             for item in equivDict:
                 for equiv in item.get (packageName, []):
                     matchString = re.compile(r'\b' + equiv + r'\b')
                     equivList.append( (matchString, equiv) )
             equivList.append( (packagesREs[packageName], packageName) )
     classDict = {}
     ncdict = {'class' : 'className', 'function' : 'functionName'}
     for filename in xmlFiles:
         if (not filename) or (ignoreSrcRE.match(filename)): continue
         dupProblems     = ''
         exceptName      = ''
         regexList       = []
         localObjects    = []
         simpleObjectREs = []
         if options.lostDefs:
             lostMatch = srcClassNameRE.search (filename)
         if lostMatch:
                 exceptName = lostMatch.group (1)
                 regexList = equivREs[exceptName]
                 xcount = len(regexList)-1
                 if not regexList[xcount][0].search (exceptName):
                     print '%s not found in' % exceptName,
                     print regexList[xcount][0]
                     sys.exit()
             else: continue
     if options.verbose:
             print "filename", filename
         try:
             filepath = os.path.join(os.environ.get('CMSSW_BASE'),'src',filename)
             if not os.path.exists(filepath):
               filepath = os.path.join(os.environ.get('CMSSW_RELEASE_BASE'),'src',filename)
             xmlObj = xml2obj (filename = filepath,
                               filtering = True,
                               nameChangeDict = ncdict)
         except Exception as detail:
             print "File %s is malformed XML.  Please fix." % filename
             print "  ", detail
             continue
         try:
             classList = xmlObj.selection.className
         except:
             try:
                 classList = xmlObj.className
             except:
                 # this isn't a real classes_def.xml file.  Skip it
                 print "**** SKIPPING '%s' - Doesn't seem to have proper information." % filename
                 continue
         if not classList:
             classList = xmlObj.functionName
             if not classList:
                 print "**** SKIPPING '%s' - Dosen't seem to have proper information(not class/function)." % filename
                 continue
         for piece in classList:
             try:
                 className = spacesRE.sub ('', piece.name)
             except:
                 # must be one of these class pattern things.  Skip it
                 #print "     skipping %s" % filename, piece.__repr__()
                 continue
             className = stdRE.sub    ('', className)
             # print "  ", className
             # Now get rid of any typedefs
             for typedef, tdList in six.iteritems(typedefsDict):
                 for alias in tdList:
                     className = re.sub (alias, typedef, className)
             classDict.setdefault (className, set()).add (filename)
             # should we check for lost definitions?
             if not options.lostDefs:
                 continue
             localObjects.append (className)
             if options.lazyLostDefs and not braketRE.search (className):
                 #print "  ", className
                 matchString = r'\b' + className + r'\b'
                 simpleObjectREs.append( (re.compile (matchString), className ) )
         for className in localObjects:
             # if we see our name (or equivalent) here, then let's
             # skip complaining about this
             foundEquiv = False
             for equivRE in regexList:
                 #print "searching %s for %s" % (equivRE[1], className)
                 if equivRE[0].search (className):
                     foundEquiv = True
                     break
             for simpleRE in simpleObjectREs:
                 if simpleRE[0].search (className):
                     foundEquiv = True
                     if options.verbose and simpleRE[1] != className:
                         print "    Using %s to ignore %s" \
                               % (simpleRE[1], className)                    
                     break
             if foundEquiv: continue
             for exRes in explicitREs:
                 if exRes[0].search(className):
                     dupProblems += "  %s : %s\n" % (exRes[1], className)
                     foundEquiv = True
                     break
             if foundEquiv: continue
             for packageName in xmlPackages:
                 # don't bother looking for the name of this
                 # package in this package
                 if packagesREs[packageName].search (className):
                     dupProblems += "  %s : %s\n" % (packageName, className)
                     break
         # for piece
         if dupProblems:
             print '\n%s\n%s\n' % (filename, dupProblems)
     # for filename
     if options.dups:
         for name, fileSet in sorted( six.iteritems(classDict) ):
             if len (fileSet) < 2:
                 continue
             print name
             fileList = sorted (fileSet)
             for filename in fileList:
                 print "  ", filename
             print
         # for name, fileSet
     # if not noDups
     #pprint.pprint (classDict)
 
 

def duplicateReflexLibrarySearch.searchDuplicatePlugins ( )

Searches the edmpluginFile to find any duplicate
plugins.

Definition at line 244 of file duplicateReflexLibrarySearch.py.

References split.

 def searchDuplicatePlugins ():
     """ Searches the edmpluginFile to find any duplicate
     plugins."""
     edmpluginFile = ''
     libenv = 'LD_LIBRARY_PATH'
     if os.environ.get('SCRAM_ARCH').startswith('osx'): libenv = 'DYLD_FALLBACK_LIBRARY_PATH'
     biglib = '/biglib/'+os.environ.get('SCRAM_ARCH')
     for libdir in os.environ.get(libenv).split(':'):
       if libdir.endswith(biglib): continue
       if os.path.exists(libdir+'/.edmplugincache'): edmpluginFile = edmpluginFile + ' ' + libdir+'/.edmplugincache'
     if edmpluginFile == '': edmpluginFile = os.path.join(os.environ.get('CMSSW_BASE'),'lib',os.environ.get('SCRAM_ARCH'),'.edmplugincache')
     cmd = "cat %s | awk '{print $2\" \"$1}' | sort | uniq | awk '{print $1}' | sort | uniq -c | grep '2 ' | awk '{print $2}'" % edmpluginFile
     output = commands.getoutput (cmd).split('\n')
     for line in output:
       if line in ignoreEdmDP: continue
       line = line.replace("*","\*")
       cmd = "cat %s | grep ' %s ' | awk '{print $1}' | sort | uniq " % (edmpluginFile,line)
       out1 = commands.getoutput (cmd).split('\n')
       print line
       for plugin in out1:
         if plugin:
             print "   **"+plugin+"**"
       print
 