forked from cms-sw/cmssw
-
Notifications
You must be signed in to change notification settings - Fork 1
/
MatrixReader.py
533 lines (453 loc) · 20.6 KB
/
MatrixReader.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
from __future__ import print_function
import sys, os
from Configuration.PyReleaseValidation.WorkFlow import WorkFlow
from Configuration.PyReleaseValidation.MatrixUtil import InputInfo
# ================================================================================
class MatrixException(Exception):
def __init__(self, msg):
self.msg = msg
def __str__(self):
return self.msg
# ================================================================================
class MatrixReader(object):
def __init__(self, opt):
self.reset(opt.what)
self.wm=opt.wmcontrol
self.revertDqmio=opt.revertDqmio
self.addCommand=opt.command
self.apply=opt.apply
self.commandLineWf=opt.workflow
self.overWrite=opt.overWrite
self.noRun = opt.noRun
return
def reset(self, what='all'):
self.what = what
#a bunch of information, but not yet the WorkFlow object
self.workFlowSteps = {}
#the actual WorkFlow objects
self.workFlows = []
self.nameList = {}
self.filesPrefMap = {'relval_standard' : 'std-' ,
'relval_highstats': 'hi-' ,
'relval_pileup': 'PU-' ,
'relval_generator': 'gen-',
'relval_extendedgen': 'genExt-',
'relval_production': 'prod-' ,
'relval_ged': 'ged-',
'relval_upgrade':'upg-',
'relval_cleanedupgrade':'clnupg-',
'relval_gpu':'gpu-',
'relval_2017':'2017-',
'relval_2026':'2026-',
'relval_identity':'id-',
'relval_machine': 'mach-',
'relval_premix': 'premix-',
'relval_nano':'nano-'
}
self.files = ['relval_standard' ,
'relval_highstats',
'relval_pileup',
'relval_generator',
'relval_extendedgen',
'relval_production',
'relval_ged',
'relval_upgrade',
'relval_cleanedupgrade',
'relval_gpu',
'relval_2017',
'relval_2026',
'relval_identity',
'relval_machine',
'relval_premix',
'relval_nano'
]
self.filesDefault = {'relval_standard':True ,
'relval_highstats':True ,
'relval_pileup':True,
'relval_generator':True,
'relval_extendedgen':True,
'relval_production':True,
'relval_ged':True,
'relval_upgrade':False,
'relval_cleanedupgrade':False,
'relval_gpu':False,
'relval_2017':True,
'relval_2026':True,
'relval_identity':False,
'relval_machine':True,
'relval_premix':True,
'relval_nano':True
}
self.relvalModule = None
return
def makeCmd(self, step):
cmd = ''
cfg = None
input = None
for k,v in step.items():
if 'no_exec' in k : continue # we want to really run it ...
if k.lower() == 'cfg':
cfg = v
continue # do not append to cmd, return separately
if k.lower() == 'input':
input = v
continue # do not append to cmd, return separately
#chain the configs
#if k.lower() == '--python':
# v = 'step%d_%s'%(index,v)
cmd += ' ' + k + ' ' + str(v)
return cfg, input, cmd
def makeStep(self,step,overrides):
from Configuration.PyReleaseValidation.relval_steps import merge
if len(overrides) > 0:
copyStep=merge([overrides]+[step])
return copyStep
else:
return step
def readMatrix(self, fileNameIn, useInput=None, refRel=None, fromScratch=None):
prefix = self.filesPrefMap[fileNameIn]
print("processing", fileNameIn)
try:
_tmpMod = __import__( 'Configuration.PyReleaseValidation.'+fileNameIn )
self.relvalModule = sys.modules['Configuration.PyReleaseValidation.'+fileNameIn]
except Exception as e:
print("ERROR importing file ", fileNameIn, str(e))
return
if useInput is not None:
print("request for INPUT for ", useInput)
fromInput={}
if useInput:
for i in useInput:
if ':' in i:
(ik,il)=i.split(':')
if ik=='all':
for k in self.relvalModule.workflows.keys():
fromInput[float(k)]=int(il)
else:
fromInput[float(ik)]=int(il)
else:
if i=='all':
for k in self.relvalModule.workflows.keys():
fromInput[float(k)]=0
else:
fromInput[float(i)]=0
if fromScratch:
fromScratch=map(float,fromScratch)
for num in fromScratch:
if num in fromInput:
fromInput.pop(num)
#overwrite steps
if self.overWrite:
for p in self.overWrite:
self.relvalModule.steps.overwrite(p)
#change the origin of dataset on the fly
if refRel:
if ',' in refRel:
refRels=refRel.split(',')
if len(refRels)!=len(self.relvalModule.baseDataSetRelease):
return
self.relvalModule.changeRefRelease(
self.relvalModule.steps,
list(zip(self.relvalModule.baseDataSetRelease,refRels))
)
else:
self.relvalModule.changeRefRelease(
self.relvalModule.steps,
[(x,refRel) for x in self.relvalModule.baseDataSetRelease]
)
for num, wfInfo in self.relvalModule.workflows.items():
commands=[]
wfName = wfInfo[0]
stepList = wfInfo[1]
stepOverrides=wfInfo.overrides
# upgrade case: workflow has basic name, key[, suffix (only special workflows)]
wfKey = ""
wfSuffix = ""
if isinstance(wfName, list) and len(wfName)>1:
if len(wfName)>2: wfSuffix = wfName[2]
wfKey = wfName[1]
wfName = wfName[0]
# if no explicit name given for the workflow, use the name of step1
if wfName.strip() == '': wfName = stepList[0]
# option to specialize the wf as the third item in the WF list
addTo=None
addCom=None
if len(wfInfo)>=3:
addCom=wfInfo[2]
if not isinstance(addCom, list): addCom=[addCom]
#print 'added dict',addCom
if len(wfInfo)>=4:
addTo=wfInfo[3]
#pad with 0
while len(addTo)!=len(stepList):
addTo.append(0)
name=wfName
# separate suffixes by + because show() excludes first part of name
if len(wfKey)>0:
name = name+'+'+wfKey
if len(wfSuffix)>0: name = name+wfSuffix
stepIndex=0
ranStepList=[]
name_for_workflow = name
#first resolve INPUT possibilities
if num in fromInput:
ilevel=fromInput[num]
#print num,ilevel
for (stepIr,step) in enumerate(reversed(stepList)):
stepName=step
stepI=(len(stepList)-stepIr)-1
#print stepIr,step,stepI,ilevel
if stepI>ilevel:
#print "ignoring"
continue
if stepI!=0:
testName='__'.join(stepList[0:stepI+1])+'INPUT'
else:
testName=step+'INPUT'
#print "JR",stepI,stepIr,testName,stepList
if testName in self.relvalModule.steps:
#print "JR",stepI,stepIr
stepList[stepI]=testName
#pop the rest in the list
#print "\tmod prepop",stepList
for p in range(stepI):
stepList.pop(0)
#print "\t\tmod",stepList
break
for (stepI,step) in enumerate(stepList):
stepName=step
if self.relvalModule.steps[stepName] is None:
continue
if self.wm:
#cannot put a certain number of things in wm
if stepName in ['SKIMD','SKIMCOSD','SKIMDreHLT']:
continue
#replace stepName is needed
#if stepName in self.replaceStep
if len(name) > 0 : name += '+'
#any step can be mirrored with INPUT
## maybe we want too level deep input
"""
if num in fromInput:
if step+'INPUT' in self.relvalModule.steps.keys():
stepName = step+"INPUT"
stepList.remove(step)
stepList.insert(stepIndex,stepName)
"""
stepNameTmp = stepName
if len(wfKey)>0: stepNameTmp = stepNameTmp.replace('_'+wfKey,"")
if len(wfSuffix)>0: stepNameTmp = stepNameTmp.replace(wfSuffix,"")
name += stepNameTmp
if addCom and (not addTo or addTo[stepIndex]==1):
from Configuration.PyReleaseValidation.relval_steps import merge
copyStep=merge(addCom+[self.makeStep(self.relvalModule.steps[stepName],stepOverrides)])
cfg, input, opts = self.makeCmd(copyStep)
else:
cfg, input, opts = self.makeCmd(self.makeStep(self.relvalModule.steps[stepName],stepOverrides))
if input and cfg :
msg = "FATAL ERROR: found both cfg and input for workflow "+str(num)+' step '+stepName
raise MatrixException(msg)
if input:
cmd = input
if self.noRun:
cmd.run=[]
else:
if cfg:
cmd = 'cmsDriver.py '+cfg+' '+opts
else:
cmd = 'cmsDriver.py step'+str(stepIndex+1)+' '+opts
if self.wm:
cmd+=' --io %s.io --python %s.py'%(stepName,stepName)
if self.addCommand:
if self.apply:
if stepIndex in self.apply or stepName in self.apply:
cmd +=' '+self.addCommand
else:
cmd +=' '+self.addCommand
if self.wm and self.revertDqmio=='yes':
cmd=cmd.replace('DQMIO','DQM')
cmd=cmd.replace('--filetype DQM','')
commands.append(cmd)
ranStepList.append(stepName)
stepIndex+=1
self.workFlowSteps[(num,prefix)] = (num, name_for_workflow, commands, ranStepList)
return
def showRaw(self, useInput, refRel=None, fromScratch=None, what='all',step1Only=False,selected=None):
if selected:
selected=map(float,selected)
for matrixFile in self.files:
self.reset(what)
if self.what != 'all' and not any('_'+el in matrixFile for el in self.what.split(",")):
print("ignoring non-requested file",matrixFile)
continue
if self.what == 'all' and not self.filesDefault[matrixFile]:
print("ignoring file not used by default (enable with -w)",matrixFile)
continue
try:
self.readMatrix(matrixFile, useInput, refRel, fromScratch)
except Exception as e:
print("ERROR reading file:", matrixFile, str(e))
raise
if not self.workFlowSteps: continue
dataFileName = matrixFile.replace('relval_', 'cmsDriver_')+'_hlt.txt'
outFile = open(dataFileName,'w')
print("found ", len(self.workFlowSteps), ' workflows for ', dataFileName)
ids = sorted(self.workFlowSteps.keys())
indexAndSteps=[]
writtenWF=0
for key in ids:
if selected and not (key[0] in selected):
continue
#trick to skip the HImix IB test
if key[0]==203.1 or key[0]==204.1 or key[0]==205.1 or key[0]==4.51 or key[0]==4.52: continue
num, name, commands, stepList = self.workFlowSteps[key]
wfName,stepNames= name.split('+',1)
stepNames=stepNames.replace('+SKIMCOSD','')
stepNames=stepNames.replace('+SKIMD','')
if 'HARVEST' in stepNames:
#find out automatically what to remove
exactb=stepNames.index('+HARVEST')
exacte=stepNames.index('+',exactb+1) if ('+' in stepNames[exactb+1:]) else (len(stepNames))
stepNames=stepNames.replace(stepNames[exactb:exacte],'')
otherSteps = None
if '+' in stepNames:
step1,otherSteps = stepNames.split('+',1)
line = str(num) + ' ++ '+ wfName
if otherSteps and not step1Only:
line += ' ++ ' +otherSteps.replace('+',',')
else:
line += ' ++ none'
inputInfo=None
if not isinstance(commands[0],str):
inputInfo=commands[0]
if otherSteps:
for (i,c) in enumerate(otherSteps.split('+')):
#pad with set
for p in range(len(indexAndSteps),i+2):
indexAndSteps.append(set())
indexAndSteps[i+1].add((c,commands[i+1]))
if inputInfo :
#skip the samples from INPUT when step1Only is on
if step1Only: continue
line += ' ++ REALDATA: '+inputInfo.dataSet
if inputInfo.run!=[]: line += ', RUN:'+'|'.join(map(str,inputInfo.run))
line += ', FILES: ' +str(inputInfo.files)
line += ', EVENTS: '+str(inputInfo.events)
if inputInfo.label!='':
line += ', LABEL: ' +inputInfo.label
line += ', LOCATION:'+inputInfo.location
line += ' @@@'
else:
line += ' @@@ '+commands[0]
if self.revertDqmio=='yes':
line=line.replace('DQMIO','DQM')
writtenWF+=1
outFile.write(line+'\n')
outFile.write('\n'+'\n')
if step1Only: continue
for (index,s) in enumerate(indexAndSteps):
for (stepName,cmd) in s:
stepIndex=index+1
if 'dasquery.log' in cmd: continue
line = 'STEP%d ++ '%(stepIndex,) +stepName + ' @@@ '+cmd
if self.revertDqmio=='yes':
line=line.replace('DQMIO','DQM')
outFile.write(line+'\n')
outFile.write('\n'+'\n')
outFile.close()
print("wrote ",writtenWF, ' workflow'+('s' if (writtenWF!=1) else ''),' to ', outFile.name)
return
def workFlowsByLocation(self, cafVeto=True):
# Check if we are on CAF
onCAF = False
if 'cms/caf/cms' in os.environ['CMS_PATH']:
onCAF = True
workflows = []
for workflow in self.workFlows:
if isinstance(workflow.cmds[0], InputInfo):
if cafVeto and (workflow.cmds[0].location == 'CAF' and not onCAF):
continue
workflows.append(workflow)
return workflows
def showWorkFlows(self, selected=None, extended=True, cafVeto=True):
if selected: selected = list(map(float,selected))
wfs = self.workFlowsByLocation(cafVeto)
maxLen = 100 # for summary, limit width of output
fmt1 = "%-6s %-35s [1]: %s ..."
fmt2 = " %35s [%d]: %s ..."
print("\nfound a total of ", len(wfs), ' workflows:')
if selected:
print(" of which the following", len(selected), 'were selected:')
#-ap for now:
maxLen = -1 # for individual listing, no limit on width
fmt1 = "%-6s %-35s [1]: %s "
fmt2 = " %35s [%d]: %s"
N=[]
for wf in wfs:
if selected and float(wf.numId) not in selected: continue
if extended: print('')
#pad with zeros
for i in range(len(N),len(wf.cmds)): N.append(0)
N[len(wf.cmds)-1]+=1
wfName = wf.nameId
stepNames = '+'.join(wf.stepList)
for i,s in enumerate(wf.cmds):
if extended:
if i==0:
print(fmt1 % (wf.numId, stepNames, (str(s)+' ')[:maxLen]))
else:
print(fmt2 % ( ' ', i+1, (str(s)+' ')[:maxLen]))
else:
print("%-6s %-35s "% (wf.numId, stepNames))
break
print('')
for i,n in enumerate(N):
if n: print(n,'workflows with',i+1,'steps')
return
def createWorkFlows(self, fileNameIn):
prefixIn = self.filesPrefMap[fileNameIn]
# get through the list of items and update the requested workflows only
keyList = self.workFlowSteps.keys()
ids = []
for item in keyList:
id, pref = item
if pref != prefixIn : continue
ids.append(id)
ids.sort()
for key in ids:
val = self.workFlowSteps[(key,prefixIn)]
num, name, commands, stepList = val
nameId = str(num)+'_'+name
if nameId in self.nameList:
print("==> duplicate name found for ", nameId)
print(' keeping : ', self.nameList[nameId])
print(' ignoring : ', val)
else:
self.nameList[nameId] = val
self.workFlows.append(WorkFlow(num, name, commands=commands, stepList=stepList))
return
def prepare(self, useInput=None, refRel='', fromScratch=None):
for matrixFile in self.files:
if self.what != 'all' and not any('_'+el in matrixFile for el in self.what.split(",")):
print("ignoring non-requested file",matrixFile)
continue
if self.what == 'all' and not self.filesDefault[matrixFile]:
print("ignoring",matrixFile,"from default matrix")
continue
try:
self.readMatrix(matrixFile, useInput, refRel, fromScratch)
except Exception as e:
print("ERROR reading file:", matrixFile, str(e))
raise
try:
self.createWorkFlows(matrixFile)
except Exception as e:
print("ERROR creating workflows :", str(e))
raise
def show(self, selected=None, extended=True, cafVeto=True):
self.showWorkFlows(selected, extended, cafVeto)
print('\n','-'*80,'\n')
def updateDB(self):
import pickle
pickle.dump(self.workFlows, open('theMatrix.pkl', 'w') )
return