Coverage for tests/test_simpleButler.py: 11%
255 statements
« prev ^ index » next coverage.py v7.3.2, created at 2023-12-06 10:53 +0000
« prev ^ index » next coverage.py v7.3.2, created at 2023-12-06 10:53 +0000
1# This file is part of daf_butler.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (http://www.lsst.org).
6# See the COPYRIGHT file at the top-level directory of this distribution
7# for details of code ownership.
8#
9# This software is dual licensed under the GNU General Public License and also
10# under a 3-clause BSD license. Recipients may choose which of these licenses
11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
12# respectively. If you choose the GPL option then the following text applies
13# (but note that there is still no warranty even if you opt for BSD instead):
14#
15# This program is free software: you can redistribute it and/or modify
16# it under the terms of the GNU General Public License as published by
17# the Free Software Foundation, either version 3 of the License, or
18# (at your option) any later version.
19#
20# This program is distributed in the hope that it will be useful,
21# but WITHOUT ANY WARRANTY; without even the implied warranty of
22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23# GNU General Public License for more details.
24#
25# You should have received a copy of the GNU General Public License
26# along with this program. If not, see <http://www.gnu.org/licenses/>.
28from __future__ import annotations
30import json
31import os
32import re
33import tempfile
34import unittest
35from typing import Any
37try:
38 import numpy as np
39except ImportError:
40 np = None
42import astropy.time
43from lsst.daf.butler import Butler, ButlerConfig, CollectionType, DatasetId, DatasetRef, DatasetType, Timespan
44from lsst.daf.butler.registry import RegistryConfig, RegistryDefaults, _RegistryFactory
45from lsst.daf.butler.tests import DatastoreMock
46from lsst.daf.butler.tests.utils import makeTestTempDir, removeTestTempDir
48TESTDIR = os.path.abspath(os.path.dirname(__file__))
51class SimpleButlerTestCase(unittest.TestCase):
52 """Tests for butler (including import/export functionality) that should not
53 depend on the Registry Database backend or Datastore implementation, and
54 can instead utilize an in-memory SQLite Registry and a mocked Datastore.
55 """
57 datasetsManager = (
58 "lsst.daf.butler.registry.datasets.byDimensions.ByDimensionsDatasetRecordStorageManagerUUID"
59 )
60 datasetsImportFile = "datasets-uuid.yaml"
62 def setUp(self):
63 self.root = makeTestTempDir(TESTDIR)
65 def tearDown(self):
66 removeTestTempDir(self.root)
68 def makeButler(self, **kwargs: Any) -> Butler:
69 """Return new Butler instance on each call."""
70 config = ButlerConfig()
72 # make separate temporary directory for registry of this instance
73 tmpdir = tempfile.mkdtemp(dir=self.root)
74 config["registry", "db"] = f"sqlite:///{tmpdir}/gen3.sqlite3"
75 config["registry", "managers", "datasets"] = self.datasetsManager
76 config["root"] = self.root
78 # have to make a registry first
79 registryConfig = RegistryConfig(config.get("registry"))
80 _RegistryFactory(registryConfig).create_from_config()
82 butler = Butler.from_config(config, **kwargs)
83 DatastoreMock.apply(butler)
84 return butler
86 def comparableRef(self, ref: DatasetRef) -> DatasetRef:
87 """Return a DatasetRef that can be compared to a DatasetRef from
88 other repository.
90 For repositories that do not support round-trip of ID values this
91 method returns unresolved DatasetRef, for round-trip-safe repos it
92 returns unchanged ref.
93 """
94 return ref
96 def testReadBackwardsCompatibility(self):
97 """Test that we can read an export file written by a previous version
98 and commit to the daf_butler git repo.
100 Notes
101 -----
102 At present this export file includes only dimension data, not datasets,
103 which greatly limits the usefulness of this test. We should address
104 this at some point, but I think it's best to wait for the changes to
105 the export format required for CALIBRATION collections to land.
106 """
107 butler = self.makeButler(writeable=True)
108 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "hsc-rc2-subset.yaml"))
109 # Spot-check a few things, but the most important test is just that
110 # the above does not raise.
111 self.assertGreaterEqual(
112 {record.id for record in butler.registry.queryDimensionRecords("detector", instrument="HSC")},
113 set(range(104)), # should have all science CCDs; may have some focus ones.
114 )
115 self.assertGreaterEqual(
116 {
117 (record.id, record.physical_filter)
118 for record in butler.registry.queryDimensionRecords("visit", instrument="HSC")
119 },
120 {
121 (27136, "HSC-Z"),
122 (11694, "HSC-G"),
123 (23910, "HSC-R"),
124 (11720, "HSC-Y"),
125 (23900, "HSC-R"),
126 (22646, "HSC-Y"),
127 (1248, "HSC-I"),
128 (19680, "HSC-I"),
129 (1240, "HSC-I"),
130 (424, "HSC-Y"),
131 (19658, "HSC-I"),
132 (344, "HSC-Y"),
133 (1218, "HSC-R"),
134 (1190, "HSC-Z"),
135 (23718, "HSC-R"),
136 (11700, "HSC-G"),
137 (26036, "HSC-G"),
138 (23872, "HSC-R"),
139 (1170, "HSC-Z"),
140 (1876, "HSC-Y"),
141 },
142 )
144 def testDatasetTransfers(self):
145 """Test exporting all datasets from a repo and then importing them all
146 back in again.
147 """
148 # Import data to play with.
149 butler1 = self.makeButler(writeable=True)
150 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
151 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
152 with tempfile.NamedTemporaryFile(mode="w", suffix=".yaml") as file:
153 # Export all datasets.
154 with butler1.export(filename=file.name) as exporter:
155 exporter.saveDatasets(butler1.registry.queryDatasets(..., collections=...))
156 # Import it all again.
157 butler2 = self.makeButler(writeable=True)
158 butler2.import_(filename=file.name)
159 datasets1 = list(butler1.registry.queryDatasets(..., collections=...))
160 datasets2 = list(butler2.registry.queryDatasets(..., collections=...))
161 self.assertTrue(all(isinstance(ref.id, DatasetId) for ref in datasets1))
162 self.assertTrue(all(isinstance(ref.id, DatasetId) for ref in datasets2))
163 self.assertCountEqual(
164 [self.comparableRef(ref) for ref in datasets1],
165 [self.comparableRef(ref) for ref in datasets2],
166 )
168 def testImportTwice(self):
169 """Test exporting dimension records and datasets from a repo and then
170 importing them all back in again twice.
171 """
172 # Import data to play with.
173 butler1 = self.makeButler(writeable=True)
174 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
175 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
176 with tempfile.NamedTemporaryFile(mode="w", suffix=".yaml", delete=False) as file:
177 # Export all datasets.
178 with butler1.export(filename=file.name) as exporter:
179 exporter.saveDatasets(butler1.registry.queryDatasets(..., collections=...))
180 butler2 = self.makeButler(writeable=True)
181 # Import it once.
182 butler2.import_(filename=file.name)
183 # Import it again
184 butler2.import_(filename=file.name)
185 datasets1 = list(butler1.registry.queryDatasets(..., collections=...))
186 datasets2 = list(butler2.registry.queryDatasets(..., collections=...))
187 self.assertTrue(all(isinstance(ref.id, DatasetId) for ref in datasets1))
188 self.assertTrue(all(isinstance(ref.id, DatasetId) for ref in datasets2))
189 self.assertCountEqual(
190 [self.comparableRef(ref) for ref in datasets1],
191 [self.comparableRef(ref) for ref in datasets2],
192 )
194 def testCollectionTransfers(self):
195 """Test exporting and then importing collections of various types."""
196 # Populate a registry with some datasets.
197 butler1 = self.makeButler(writeable=True)
198 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
199 butler1.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
200 registry1 = butler1.registry
201 # Add some more collections.
202 registry1.registerRun("run1")
203 registry1.registerCollection("tag1", CollectionType.TAGGED)
204 registry1.registerCollection("calibration1", CollectionType.CALIBRATION)
205 registry1.registerCollection("chain1", CollectionType.CHAINED)
206 registry1.registerCollection("chain2", CollectionType.CHAINED)
207 registry1.setCollectionChain("chain1", ["tag1", "run1", "chain2"])
208 registry1.setCollectionChain("chain2", ["calibration1", "run1"])
209 # Associate some datasets into the TAGGED and CALIBRATION collections.
210 flats1 = list(registry1.queryDatasets("flat", collections=...))
211 registry1.associate("tag1", flats1)
212 t1 = astropy.time.Time("2020-01-01T01:00:00", format="isot", scale="tai")
213 t2 = astropy.time.Time("2020-01-01T02:00:00", format="isot", scale="tai")
214 t3 = astropy.time.Time("2020-01-01T03:00:00", format="isot", scale="tai")
215 bias1a = registry1.findDataset("bias", instrument="Cam1", detector=1, collections="imported_g")
216 bias2a = registry1.findDataset("bias", instrument="Cam1", detector=2, collections="imported_g")
217 bias3a = registry1.findDataset("bias", instrument="Cam1", detector=3, collections="imported_g")
218 bias2b = registry1.findDataset("bias", instrument="Cam1", detector=2, collections="imported_r")
219 bias3b = registry1.findDataset("bias", instrument="Cam1", detector=3, collections="imported_r")
220 registry1.certify("calibration1", [bias2a, bias3a], Timespan(t1, t2))
221 registry1.certify("calibration1", [bias2b], Timespan(t2, None))
222 registry1.certify("calibration1", [bias3b], Timespan(t2, t3))
223 registry1.certify("calibration1", [bias1a], Timespan.makeEmpty())
225 with tempfile.NamedTemporaryFile(mode="w", suffix=".yaml") as file:
226 # Export all collections, and some datasets.
227 with butler1.export(filename=file.name) as exporter:
228 # Sort results to put chain1 before chain2, which is
229 # intentionally not topological order.
230 for collection in sorted(registry1.queryCollections()):
231 exporter.saveCollection(collection)
232 exporter.saveDatasets(flats1)
233 exporter.saveDatasets([bias1a, bias2a, bias2b, bias3a, bias3b])
234 # Import them into a new registry.
235 butler2 = self.makeButler(writeable=True)
236 butler2.import_(filename=file.name)
237 registry2 = butler2.registry
238 # Check that it all round-tripped, starting with the collections
239 # themselves.
240 self.assertIs(registry2.getCollectionType("run1"), CollectionType.RUN)
241 self.assertIs(registry2.getCollectionType("tag1"), CollectionType.TAGGED)
242 self.assertIs(registry2.getCollectionType("calibration1"), CollectionType.CALIBRATION)
243 self.assertIs(registry2.getCollectionType("chain1"), CollectionType.CHAINED)
244 self.assertIs(registry2.getCollectionType("chain2"), CollectionType.CHAINED)
245 self.assertEqual(
246 list(registry2.getCollectionChain("chain1")),
247 ["tag1", "run1", "chain2"],
248 )
249 self.assertEqual(
250 list(registry2.getCollectionChain("chain2")),
251 ["calibration1", "run1"],
252 )
253 # Check that tag collection contents are the same.
254 self.maxDiff = None
255 self.assertCountEqual(
256 [self.comparableRef(ref) for ref in registry1.queryDatasets(..., collections="tag1")],
257 [self.comparableRef(ref) for ref in registry2.queryDatasets(..., collections="tag1")],
258 )
259 # Check that calibration collection contents are the same.
260 self.assertCountEqual(
261 [
262 (self.comparableRef(assoc.ref), assoc.timespan)
263 for assoc in registry1.queryDatasetAssociations("bias", collections="calibration1")
264 ],
265 [
266 (self.comparableRef(assoc.ref), assoc.timespan)
267 for assoc in registry2.queryDatasetAssociations("bias", collections="calibration1")
268 ],
269 )
271 def testButlerGet(self):
272 """Test that butler.get can work with different variants."""
273 # Import data to play with.
274 butler = self.makeButler(writeable=True)
275 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
276 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
278 # Find the DatasetRef for a flat
279 coll = "imported_g"
280 flat2g = butler.find_dataset(
281 "flat", instrument="Cam1", full_name="Ab", physical_filter="Cam1-G", collections=coll
282 )
284 # Create a numpy integer to check that works fine
285 detector_np = np.int64(2) if np else 2
287 # Try to get it using different variations of dataId + keyword
288 # arguments
289 # Note that instrument.class_name does not work
290 variants = (
291 (None, {"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G"}),
292 (None, {"instrument": "Cam1", "detector": detector_np, "physical_filter": "Cam1-G"}),
293 ({"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G"}, {}),
294 ({"instrument": "Cam1", "detector": detector_np, "physical_filter": "Cam1-G"}, {}),
295 ({"instrument": "Cam1", "detector": 2}, {"physical_filter": "Cam1-G"}),
296 ({"detector.full_name": "Ab"}, {"instrument": "Cam1", "physical_filter": "Cam1-G"}),
297 ({"full_name": "Ab"}, {"instrument": "Cam1", "physical_filter": "Cam1-G"}),
298 (None, {"full_name": "Ab", "instrument": "Cam1", "physical_filter": "Cam1-G"}),
299 (None, {"detector": "Ab", "instrument": "Cam1", "physical_filter": "Cam1-G"}),
300 ({"name_in_raft": "b", "raft": "A"}, {"instrument": "Cam1", "physical_filter": "Cam1-G"}),
301 ({"name_in_raft": "b"}, {"raft": "A", "instrument": "Cam1", "physical_filter": "Cam1-G"}),
302 (None, {"name_in_raft": "b", "raft": "A", "instrument": "Cam1", "physical_filter": "Cam1-G"}),
303 (
304 {"detector.name_in_raft": "b", "detector.raft": "A"},
305 {"instrument": "Cam1", "physical_filter": "Cam1-G"},
306 ),
307 (
308 {
309 "detector.name_in_raft": "b",
310 "detector.raft": "A",
311 "instrument": "Cam1",
312 "physical_filter": "Cam1-G",
313 },
314 {},
315 ),
316 # Duplicate (but valid) information.
317 (None, {"instrument": "Cam1", "detector": 2, "raft": "A", "physical_filter": "Cam1-G"}),
318 ({"detector": 2}, {"instrument": "Cam1", "raft": "A", "physical_filter": "Cam1-G"}),
319 ({"raft": "A"}, {"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G"}),
320 ({"raft": "A"}, {"instrument": "Cam1", "detector": "Ab", "physical_filter": "Cam1-G"}),
321 )
323 for dataId, kwds in variants:
324 try:
325 flat_id, _ = butler.get("flat", dataId=dataId, collections=coll, **kwds)
326 except Exception as e:
327 raise type(e)(f"{str(e)}: dataId={dataId}, kwds={kwds}") from e
328 self.assertEqual(flat_id, flat2g.id, msg=f"DataId: {dataId}, kwds: {kwds}")
330 # Check that bad combinations raise.
331 variants = (
332 # Inconsistent detector information.
333 (None, {"instrument": "Cam1", "detector": 2, "raft": "B", "physical_filter": "Cam1-G"}),
334 ({"detector": 2}, {"instrument": "Cam1", "raft": "B", "physical_filter": "Cam1-G"}),
335 ({"detector": 12}, {"instrument": "Cam1", "raft": "B", "physical_filter": "Cam1-G"}),
336 ({"raft": "B"}, {"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G"}),
337 ({"raft": "B"}, {"instrument": "Cam1", "detector": "Ab", "physical_filter": "Cam1-G"}),
338 # Under-specified.
339 ({"raft": "B"}, {"instrument": "Cam1", "physical_filter": "Cam1-G"}),
340 # Spurious kwargs.
341 (None, {"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G", "x": "y"}),
342 ({"x": "y"}, {"instrument": "Cam1", "detector": 2, "physical_filter": "Cam1-G"}),
343 )
344 for dataId, kwds in variants:
345 with self.assertRaises((ValueError, LookupError)):
346 butler.get("flat", dataId=dataId, collections=coll, **kwds)
348 def testGetCalibration(self):
349 """Test that `Butler.get` can be used to fetch from
350 `~CollectionType.CALIBRATION` collections if the data ID includes
351 extra dimensions with temporal information.
352 """
353 # Import data to play with.
354 butler = self.makeButler(writeable=True)
355 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
356 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
357 # Certify some biases into a CALIBRATION collection.
358 registry = butler.registry
359 registry.registerCollection("calibs", CollectionType.CALIBRATION)
360 t1 = astropy.time.Time("2020-01-01T01:00:00", format="isot", scale="tai")
361 t2 = astropy.time.Time("2020-01-01T02:00:00", format="isot", scale="tai")
362 t3 = astropy.time.Time("2020-01-01T03:00:00", format="isot", scale="tai")
363 bias2a = registry.findDataset("bias", instrument="Cam1", detector=2, collections="imported_g")
364 bias3a = registry.findDataset("bias", instrument="Cam1", detector=3, collections="imported_g")
365 bias2b = registry.findDataset("bias", instrument="Cam1", detector=2, collections="imported_r")
366 bias3b = registry.findDataset("bias", instrument="Cam1", detector=3, collections="imported_r")
367 registry.certify("calibs", [bias2a, bias3a], Timespan(t1, t2))
368 registry.certify("calibs", [bias2b], Timespan(t2, None))
369 registry.certify("calibs", [bias3b], Timespan(t2, t3))
370 # Insert some exposure dimension data.
371 registry.insertDimensionData(
372 "exposure",
373 {
374 "instrument": "Cam1",
375 "id": 3,
376 "obs_id": "three",
377 "timespan": Timespan(t1, t2),
378 "physical_filter": "Cam1-G",
379 "day_obs": 20201114,
380 "seq_num": 55,
381 },
382 {
383 "instrument": "Cam1",
384 "id": 4,
385 "obs_id": "four",
386 "timespan": Timespan(t2, t3),
387 "physical_filter": "Cam1-G",
388 "day_obs": 20211114,
389 "seq_num": 42,
390 },
391 )
392 # Get some biases from raw-like data IDs.
393 bias2a_id, _ = butler.get(
394 "bias", {"instrument": "Cam1", "exposure": 3, "detector": 2}, collections="calibs"
395 )
396 self.assertEqual(bias2a_id, bias2a.id)
397 bias3b_id, _ = butler.get(
398 "bias", {"instrument": "Cam1", "exposure": 4, "detector": 3}, collections="calibs"
399 )
400 self.assertEqual(bias3b_id, bias3b.id)
402 # Get using the kwarg form
403 bias3b_id, _ = butler.get("bias", instrument="Cam1", exposure=4, detector=3, collections="calibs")
404 self.assertEqual(bias3b_id, bias3b.id)
406 # Do it again but using the record information
407 bias2a_id, _ = butler.get(
408 "bias",
409 {"instrument": "Cam1", "exposure.obs_id": "three", "detector.full_name": "Ab"},
410 collections="calibs",
411 )
412 self.assertEqual(bias2a_id, bias2a.id)
413 bias3b_id, _ = butler.get(
414 "bias",
415 {"exposure.obs_id": "four", "detector.full_name": "Ba"},
416 collections="calibs",
417 instrument="Cam1",
418 )
419 self.assertEqual(bias3b_id, bias3b.id)
421 # And again but this time using the alternate value rather than
422 # the primary.
423 bias3b_id, _ = butler.get(
424 "bias", {"exposure": "four", "detector": "Ba"}, collections="calibs", instrument="Cam1"
425 )
426 self.assertEqual(bias3b_id, bias3b.id)
428 # And again but this time using the alternate value rather than
429 # the primary and do it in the keyword arguments.
430 bias3b_id, _ = butler.get(
431 "bias", exposure="four", detector="Ba", collections="calibs", instrument="Cam1"
432 )
433 self.assertEqual(bias3b_id, bias3b.id)
435 # Now with implied record columns
436 bias3b_id, _ = butler.get(
437 "bias",
438 day_obs=20211114,
439 seq_num=42,
440 raft="B",
441 name_in_raft="a",
442 collections="calibs",
443 instrument="Cam1",
444 )
445 self.assertEqual(bias3b_id, bias3b.id)
447 # Allow a fully-specified dataId and unnecessary extra information
448 # that comes from the record.
449 bias3b_id, _ = butler.get(
450 "bias",
451 dataId=dict(
452 exposure=4,
453 day_obs=20211114,
454 seq_num=42,
455 detector=3,
456 instrument="Cam1",
457 ),
458 collections="calibs",
459 )
460 self.assertEqual(bias3b_id, bias3b.id)
462 # Extra but inconsistent record values are a problem.
463 with self.assertRaises(ValueError):
464 bias3b_id, _ = butler.get(
465 "bias",
466 exposure=3,
467 day_obs=20211114,
468 seq_num=42,
469 detector=3,
470 collections="calibs",
471 instrument="Cam1",
472 )
474 # Ensure that spurious kwargs cause an exception.
475 with self.assertRaises(ValueError):
476 butler.get(
477 "bias",
478 {"exposure.obs_id": "four", "immediate": True, "detector.full_name": "Ba"},
479 collections="calibs",
480 instrument="Cam1",
481 )
483 with self.assertRaises(ValueError):
484 butler.get(
485 "bias",
486 day_obs=20211114,
487 seq_num=42,
488 raft="B",
489 name_in_raft="a",
490 collections="calibs",
491 instrument="Cam1",
492 immediate=True,
493 )
495 def testRegistryDefaults(self):
496 """Test that we can default the collections and some data ID keys when
497 constructing a butler.
499 Many tests that use default run already exist in ``test_butler.py``, so
500 that isn't tested here. And while most of this functionality is
501 implemented in `Registry`, we test it here instead of
502 ``daf/butler/tests/registry.py`` because it shouldn't depend on the
503 database backend at all.
504 """
505 butler = self.makeButler(writeable=True)
506 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
507 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
508 # Need to actually set defaults later, not at construction, because
509 # we need to import the instrument before we can use it as a default.
510 # Don't set a default instrument value for data IDs, because 'Cam1'
511 # should be inferred by virtue of that being the only value in the
512 # input collections.
513 butler.registry.defaults = RegistryDefaults(collections=["imported_g"])
514 # Use findDataset without collections or instrument.
515 ref = butler.find_dataset("flat", detector=2, physical_filter="Cam1-G")
516 # Do the same with Butler.get; this should ultimately invoke a lot of
517 # the same code, so it's a bit circular, but mostly we're checking that
518 # it works at all.
519 dataset_id, _ = butler.get("flat", detector=2, physical_filter="Cam1-G")
520 self.assertEqual(ref.id, dataset_id)
521 # Query for datasets. Test defaulting the data ID in both kwargs and
522 # in the WHERE expression.
523 queried_refs_1 = set(butler.registry.queryDatasets("flat", detector=2, physical_filter="Cam1-G"))
524 self.assertEqual({ref}, queried_refs_1)
525 queried_refs_2 = set(
526 butler.registry.queryDatasets("flat", where="detector=2 AND physical_filter='Cam1-G'")
527 )
528 self.assertEqual({ref}, queried_refs_2)
529 # Query for data IDs with a dataset constraint.
530 queried_data_ids = set(
531 butler.registry.queryDataIds(
532 {"instrument", "detector", "physical_filter"},
533 datasets={"flat"},
534 detector=2,
535 physical_filter="Cam1-G",
536 )
537 )
538 self.assertEqual({ref.dataId}, queried_data_ids)
539 # Add another instrument to the repo, and a dataset that uses it to
540 # the `imported_g` collection.
541 butler.registry.insertDimensionData("instrument", {"name": "Cam2"})
542 camera = DatasetType(
543 "camera",
544 dimensions=butler.dimensions["instrument"].graph,
545 storageClass="Camera",
546 )
547 butler.registry.registerDatasetType(camera)
548 butler.registry.insertDatasets(camera, [{"instrument": "Cam2"}], run="imported_g")
549 # Initialize a new butler with `imported_g` as its default run.
550 # This should not have a default instrument, because there are two.
551 # Pass run instead of collections; this should set both.
552 butler2 = Butler.from_config(butler=butler, run="imported_g")
553 self.assertEqual(list(butler2.registry.defaults.collections), ["imported_g"])
554 self.assertEqual(butler2.registry.defaults.run, "imported_g")
555 self.assertFalse(butler2.registry.defaults.dataId)
556 # Initialize a new butler with an instrument default explicitly given.
557 # Set collections instead of run, which should then be None.
558 butler3 = Butler.from_config(butler=butler, collections=["imported_g"], instrument="Cam2")
559 self.assertEqual(list(butler3.registry.defaults.collections), ["imported_g"])
560 self.assertIsNone(butler3.registry.defaults.run, None)
561 self.assertEqual(butler3.registry.defaults.dataId.required, {"instrument": "Cam2"})
563 # Check that repr() does not fail.
564 defaults = RegistryDefaults(collections=["imported_g"], run="test")
565 r = repr(defaults)
566 self.assertIn("collections=('imported_g',)", r)
567 self.assertIn("run='test'", r)
569 defaults = RegistryDefaults(run="test", instrument="DummyCam", skypix="pix")
570 r = repr(defaults)
571 self.assertIn("skypix='pix'", r)
572 self.assertIn("instrument='DummyCam'", r)
574 def testJson(self):
575 """Test JSON serialization mediated by registry."""
576 butler = self.makeButler(writeable=True)
577 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
578 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", self.datasetsImportFile))
579 # Need to actually set defaults later, not at construction, because
580 # we need to import the instrument before we can use it as a default.
581 # Don't set a default instrument value for data IDs, because 'Cam1'
582 # should be inferred by virtue of that being the only value in the
583 # input collections.
584 butler.registry.defaults = RegistryDefaults(collections=["imported_g"])
585 # Use findDataset without collections or instrument.
586 ref = butler.find_dataset("flat", detector=2, physical_filter="Cam1-G")
588 # Transform the ref and dataset type to and from JSON
589 # and check that it can be reconstructed properly
591 # Do it with the ref and a component ref in minimal and standard form
592 compRef = ref.makeComponentRef("wcs")
594 for test_item in (ref, ref.datasetType, compRef, compRef.datasetType):
595 for minimal in (False, True):
596 json_str = test_item.to_json(minimal=minimal)
597 from_json = type(test_item).from_json(json_str, registry=butler.registry)
598 self.assertEqual(from_json, test_item, msg=f"From JSON '{json_str}' using registry")
600 # for minimal=False case also do a test without registry
601 if not minimal:
602 from_json = type(test_item).from_json(json_str, universe=butler.dimensions)
603 self.assertEqual(from_json, test_item, msg=f"From JSON '{json_str}' using universe")
605 def testJsonDimensionRecordsAndHtmlRepresentation(self):
606 # Dimension Records
607 butler = self.makeButler(writeable=True)
608 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "hsc-rc2-subset.yaml"))
610 for dimension in ("detector", "visit"):
611 records = butler.registry.queryDimensionRecords(dimension, instrument="HSC")
612 for r in records:
613 for minimal in (True, False):
614 json_str = r.to_json(minimal=minimal)
615 r_json = type(r).from_json(json_str, registry=butler.registry)
616 self.assertEqual(r_json, r)
617 # check with direct method
618 simple = r.to_simple()
619 fromDirect = type(simple).direct(**json.loads(json_str))
620 self.assertEqual(simple, fromDirect)
621 # Also check equality of each of the components as dicts
622 self.assertEqual(r_json.toDict(), r.toDict())
624 # check the html representation of records
625 r_html = r._repr_html_()
626 self.assertTrue(isinstance(r_html, str))
627 self.assertIn(dimension, r_html)
629 def testWildcardQueries(self):
630 """Test that different collection type queries work."""
631 # Import data to play with.
632 butler = self.makeButler(writeable=True)
633 butler.import_(filename=os.path.join(TESTDIR, "data", "registry", "base.yaml"))
635 # Create some collections
636 created = {"collection", "u/user/test", "coll3"}
637 for collection in created:
638 butler.registry.registerCollection(collection, type=CollectionType.RUN)
640 collections = butler.registry.queryCollections()
641 self.assertEqual(set(collections), created)
643 expressions = (
644 ("collection", {"collection"}),
645 (..., created),
646 ("*", created),
647 (("collection", "*"), created),
648 ("u/*", {"u/user/test"}),
649 (re.compile("u.*"), {"u/user/test"}),
650 (re.compile(".*oll.*"), {"collection", "coll3"}),
651 ("*oll*", {"collection", "coll3"}),
652 ((re.compile(r".*\d$"), "u/user/test"), {"coll3", "u/user/test"}),
653 ("*[0-9]", {"coll3"}),
654 )
655 for expression, expected in expressions:
656 result = butler.registry.queryCollections(expression)
657 self.assertEqual(set(result), expected)
660class SimpleButlerMixedUUIDTestCase(SimpleButlerTestCase):
661 """Same as SimpleButlerTestCase but uses UUID-based datasets manager and
662 loads datasets from YAML file with integer IDs.
663 """
665 datasetsImportFile = "datasets.yaml"
668if __name__ == "__main__":
669 unittest.main()