Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

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

# This file is part of pipe_base. 

# 

# Developed for the LSST Data Management System. 

# This product includes software developed by the LSST Project 

# (http://www.lsst.org). 

# See the COPYRIGHT file at the top-level directory of this distribution 

# for details of code ownership. 

# 

# This program is free software: you can redistribute it and/or modify 

# it under the terms of the GNU General Public License as published by 

# the Free Software Foundation, either version 3 of the License, or 

# (at your option) any later version. 

# 

# This program is distributed in the hope that it will be useful, 

# but WITHOUT ANY WARRANTY; without even the implied warranty of 

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 

# GNU General Public License for more details. 

# 

# You should have received a copy of the GNU General Public License 

# along with this program. If not, see <http://www.gnu.org/licenses/>. 

 

"""Simple unit test for Pipeline. 

""" 

 

from collections import namedtuple 

import unittest 

 

from lsst.pipe.base import (PipelineTask, PipelineTaskConfig, 

InputDatasetField, OutputDatasetField, 

DatasetTypeDescriptor, Pipeline, 

TaskDef, pipeTools) 

import lsst.utils.tests 

 

# mock for actual dataset type 

DS = namedtuple("DS", "name dimensions") 

 

 

# This method is used by PipelineTask to instanciate DatasetType, normally this 

# should come from some other module but we have not defined that yet, so I 

# stick a trivial (mock) implementation here. 

def makeDatasetTypeDescr(dsConfig): 

datasetType = DS(name=dsConfig.name, dimensions=dsConfig.dimensions) 

return DatasetTypeDescriptor(datasetType, scalar=False) 

 

 

class ExamplePipelineTaskConfig(PipelineTaskConfig): 

input1 = InputDatasetField(name="", 

dimensions=[], 

storageClass="example", 

doc="Input for this task") 

input2 = InputDatasetField(name="", 

dimensions=[], 

storageClass="example", 

doc="Input for this task") 

output1 = OutputDatasetField(name="", 

dimensions=[], 

storageClass="example", 

doc="Output for this task") 

output2 = OutputDatasetField(name="", 

dimensions=[], 

storageClass="example", 

doc="Output for this task") 

 

 

def _makeConfig(inputName, outputName): 

"""Factory method for config instances 

 

inputName and outputName can be either string or tuple of strings 

with two items max. 

""" 

config = ExamplePipelineTaskConfig() 

if isinstance(inputName, tuple): 

config.input1.name = inputName[0] 

config.input2.name = inputName[1] if len(inputName) > 1 else "" 

else: 

config.input1.name = inputName 

 

if isinstance(outputName, tuple): 

config.output1.name = outputName[0] 

config.output2.name = outputName[1] if len(outputName) > 1 else "" 

else: 

config.output1.name = outputName 

 

dimensions = ["Visit", "Detector"] 

config.input1.dimensions = dimensions 

config.input2.dimensions = dimensions 

config.output1.dimensions = dimensions 

config.output2.dimensions = dimensions 

 

return config 

 

 

class ExamplePipelineTask(PipelineTask): 

ConfigClass = ExamplePipelineTaskConfig 

 

@classmethod 

def getInputDatasetTypes(cls, config): 

types = {"input1": makeDatasetTypeDescr(config.input1)} 

if config.input2.name: 

types["input2"] = makeDatasetTypeDescr(config.input2) 

return types 

 

@classmethod 

def getOutputDatasetTypes(cls, config): 

types = {"output1": makeDatasetTypeDescr(config.output1)} 

if config.output2.name: 

types["output2"] = makeDatasetTypeDescr(config.output2) 

return types 

 

 

def _makePipeline(tasks): 

"""Generate Pipeline instance. 

 

Parameters 

---------- 

tasks : list of tuples 

Each tuple in the list has 3 or 4 items: 

- input DatasetType name(s), string or tuple of strings 

- output DatasetType name(s), string or tuple of strings 

- task label, string 

- optional task class object, can be None 

 

Returns 

------- 

Pipeline instance 

""" 

pipe = Pipeline() 

for task in tasks: 

inputs = task[0] 

outputs = task[1] 

label = task[2] 

klass = task[3] if len(task) > 3 else ExamplePipelineTask 

config = _makeConfig(inputs, outputs) 

pipe.append(TaskDef("ExamplePipelineTask", config, klass, label)) 

return pipe 

 

 

class PipelineToolsTestCase(unittest.TestCase): 

"""A test case for pipelineTools 

""" 

 

def setUp(self): 

pass 

 

def tearDown(self): 

pass 

 

def testIsOrdered(self): 

"""Tests for pipeTools.isPipelineOrdered method 

""" 

pipeline = _makePipeline([("A", "B", "task1"), 

("B", "C", "task2")]) 

self.assertTrue(pipeTools.isPipelineOrdered(pipeline)) 

 

pipeline = _makePipeline([("B", "C", "task2"), 

("A", "B", "task1")]) 

self.assertFalse(pipeTools.isPipelineOrdered(pipeline)) 

 

pipeline = _makePipeline([("A", ("B", "C"), "task1"), 

("B", "D", "task2"), 

("C", "E", "task3"), 

(("D", "E"), "F", "task4")]) 

self.assertTrue(pipeTools.isPipelineOrdered(pipeline)) 

 

pipeline = _makePipeline([("A", ("B", "C"), "task1"), 

("C", "E", "task2"), 

("B", "D", "task3"), 

(("D", "E"), "F", "task4")]) 

self.assertTrue(pipeTools.isPipelineOrdered(pipeline)) 

 

pipeline = _makePipeline([(("D", "E"), "F", "task4"), 

("B", "D", "task2"), 

("C", "E", "task3"), 

("A", ("B", "C"), "task1")]) 

self.assertFalse(pipeTools.isPipelineOrdered(pipeline)) 

 

def testIsOrderedExceptions(self): 

"""Tests for pipeTools.isPipelineOrdered method exceptions 

""" 

# two producers should throw ValueError 

pipeline = _makePipeline([("A", "B", "task1"), 

("B", "C", "task2"), 

("A", "C", "task3"), 

]) 

with self.assertRaises(pipeTools.DuplicateOutputError): 

pipeTools.isPipelineOrdered(pipeline) 

 

# missing factory should throw ValueError 

pipeline = _makePipeline([("A", "B", "task1", None), 

("B", "C", "task2", None)]) 

with self.assertRaises(pipeTools.MissingTaskFactoryError): 

pipeTools.isPipelineOrdered(pipeline) 

 

def testOrderPipeline(self): 

"""Tests for pipeTools.orderPipeline method 

""" 

pipeline = _makePipeline([("A", "B", "task1"), 

("B", "C", "task2")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 2) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task2") 

 

pipeline = _makePipeline([("B", "C", "task2"), 

("A", "B", "task1")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 2) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task2") 

 

pipeline = _makePipeline([("A", ("B", "C"), "task1"), 

("B", "D", "task2"), 

("C", "E", "task3"), 

(("D", "E"), "F", "task4")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 4) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task2") 

self.assertEqual(pipeline[2].label, "task3") 

self.assertEqual(pipeline[3].label, "task4") 

 

pipeline = _makePipeline([("A", ("B", "C"), "task1"), 

("C", "E", "task3"), 

("B", "D", "task2"), 

(("D", "E"), "F", "task4")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 4) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task3") 

self.assertEqual(pipeline[2].label, "task2") 

self.assertEqual(pipeline[3].label, "task4") 

 

pipeline = _makePipeline([(("D", "E"), "F", "task4"), 

("B", "D", "task2"), 

("C", "E", "task3"), 

("A", ("B", "C"), "task1")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 4) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task2") 

self.assertEqual(pipeline[2].label, "task3") 

self.assertEqual(pipeline[3].label, "task4") 

 

pipeline = _makePipeline([(("D", "E"), "F", "task4"), 

("C", "E", "task3"), 

("B", "D", "task2"), 

("A", ("B", "C"), "task1")]) 

pipeline = pipeTools.orderPipeline(pipeline) 

self.assertEqual(len(pipeline), 4) 

self.assertEqual(pipeline[0].label, "task1") 

self.assertEqual(pipeline[1].label, "task3") 

self.assertEqual(pipeline[2].label, "task2") 

self.assertEqual(pipeline[3].label, "task4") 

 

def testOrderPipelineExceptions(self): 

"""Tests for pipeTools.orderPipeline method exceptions 

""" 

# two producers should throw ValueError 

pipeline = _makePipeline([("A", "B", "task1"), 

("B", "C", "task2"), 

("A", "C", "task3"), 

]) 

with self.assertRaises(pipeTools.DuplicateOutputError): 

pipeline = pipeTools.orderPipeline(pipeline) 

 

# missing factory should throw ValueError 

pipeline = _makePipeline([("A", "B", "task1", None), 

("B", "C", "task2", None)]) 

with self.assertRaises(pipeTools.MissingTaskFactoryError): 

pipeline = pipeTools.orderPipeline(pipeline) 

 

# cycle in a graph should throw ValueError 

pipeline = _makePipeline([("A", ("A", "B"), "task1")]) 

with self.assertRaises(pipeTools.PipelineDataCycleError): 

pipeline = pipeTools.orderPipeline(pipeline) 

 

# another kind of cycle in a graph 

pipeline = _makePipeline([("A", "B", "task1"), 

("B", "C", "task2"), 

("C", "D", "task3"), 

("D", "A", "task4")]) 

with self.assertRaises(pipeTools.PipelineDataCycleError): 

pipeline = pipeTools.orderPipeline(pipeline) 

 

 

class MyMemoryTestCase(lsst.utils.tests.MemoryTestCase): 

pass 

 

 

def setup_module(module): 

lsst.utils.tests.init() 

 

 

294 ↛ 295line 294 didn't jump to line 295, because the condition on line 294 was never trueif __name__ == "__main__": 

lsst.utils.tests.init() 

unittest.main()