Coverage for python/lsst/daf/butler/script/ingest_files.py: 24%
54 statements
« prev ^ index » next coverage.py v7.11.0, created at 2025-11-06 01:33 -0800
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/>.
27from __future__ import annotations
29__all__ = ("ingest_files",)
31import logging
32from collections import defaultdict
33from typing import TYPE_CHECKING, Any
35from astropy.table import Table
37from lsst.resources import ResourcePath
38from lsst.utils import doImport
40from .._butler import Butler
41from .._dataset_ref import DatasetIdGenEnum, DatasetRef
42from .._file_dataset import FileDataset
44if TYPE_CHECKING:
45 from .._dataset_type import DatasetType
46 from ..dimensions import DimensionUniverse
48log = logging.getLogger(__name__)
51def ingest_files(
52 repo: str,
53 dataset_type: str,
54 run: str,
55 table_file: str,
56 data_id: tuple[str, ...] = (),
57 formatter: str | None = None,
58 id_generation_mode: str = "UNIQUE",
59 prefix: str | None = None,
60 transfer: str = "auto",
61 track_file_attrs: bool = True,
62) -> None:
63 """Ingest files from a table.
65 Parameters
66 ----------
67 repo : `str`
68 URI string of the Butler repo to use.
69 dataset_type : `str`
70 The name of the dataset type for the files to be ingested. This
71 dataset type must exist.
72 run : `str`
73 The run in which the files should be ingested.
74 table_file : `str`
75 Path to a table file to read. This file can be in any format that
76 can be read by Astropy so long as Astropy can determine the format
77 itself.
78 data_id : `tuple` of `str`
79 Tuple of strings of the form ``keyword=value`` that can be used
80 to define dataId elements that are fixed for all ingested files
81 found in the table file. This allows those columns to be missing
82 from the table file. Dimensions given here override table columns.
83 formatter : `str`, optional
84 Fully-qualified python class name for the `Formatter` to use
85 to read the ingested files. If `None` the formatter is read from
86 datastore configuration based on the dataset type.
87 id_generation_mode : `str`, optional
88 Mode to use for generating IDs. Should map to `DatasetGenIdEnum`.
89 prefix : `str`, optional
90 Prefix to use when resolving relative paths in table files. The default
91 is to use the current working directory.
92 transfer : `str`, optional
93 Transfer mode to use for ingest.
94 track_file_attrs : `bool`, optional
95 Control whether file attributes such as the size or checksum should
96 be tracked by the datastore. Whether this parameter is honored
97 depends on the specific datastore implementation.
98 """
99 # Check that the formatter can be imported -- validate this as soon
100 # as possible before we read a potentially large table file.
101 if formatter:
102 doImport(formatter)
103 else:
104 formatter = None
106 # Force empty string prefix (from click) to None for API compatibility.
107 if not prefix:
108 prefix = None
110 # Convert the dataset ID gen mode string to enum.
111 id_gen_mode = DatasetIdGenEnum.__members__[id_generation_mode]
113 # Create the butler with the relevant run attached.
114 butler = Butler.from_config(repo, run=run)
116 datasetType = butler.get_dataset_type(dataset_type)
118 # Convert the k=v strings into a dataId dict.
119 universe = butler.dimensions
120 common_data_id = parse_data_id_tuple(data_id, universe)
122 # Read the table assuming that Astropy can work out the format.
123 uri = ResourcePath(table_file, forceAbsolute=False)
124 with uri.as_local() as local_file:
125 table = Table.read(local_file.ospath)
127 datasets = extract_datasets_from_table(
128 table, common_data_id, datasetType, run, formatter, prefix, id_gen_mode
129 )
131 butler.ingest(*datasets, transfer=transfer, record_validation_info=track_file_attrs)
134def extract_datasets_from_table(
135 table: Table,
136 common_data_id: dict,
137 datasetType: DatasetType,
138 run: str,
139 formatter: str | None = None,
140 prefix: str | None = None,
141 id_generation_mode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE,
142) -> list[FileDataset]:
143 """Extract datasets from the supplied table.
145 Parameters
146 ----------
147 table : `astropy.table.Table`
148 Table containing the datasets. The first column is assumed to be
149 the file URI and the remaining columns are dimensions.
150 common_data_id : `dict`
151 Data ID values that are common to every row in the table. These
152 take priority if a dimension in this dataId is also present as
153 a column in the table.
154 datasetType : `DatasetType`
155 The dataset type to be associated with the ingested data.
156 run : `str`
157 The name of the run that will be receiving these datasets.
158 formatter : `str`, optional
159 Fully-qualified python class name for the `Formatter` to use
160 to read the ingested files. If `None` the formatter is read from
161 datastore configuration based on the dataset type.
162 prefix : `str`, optional
163 Prefix to be used for relative paths. Can be `None` for current
164 working directory.
165 id_generation_mode : `DatasetIdGenEnum`, optional
166 The mode to use when creating the dataset IDs.
168 Returns
169 -------
170 datasets : `list` of `FileDataset`
171 The `FileDataset` objects corresponding to the rows in the table.
172 The number of elements in this list can be smaller than the number
173 of rows in the file because one file can appear in multiple rows
174 with different dataIds.
175 """
176 # The file is the first column and everything else is assumed to
177 # be dimensions so we need to know the name of that column.
178 file_column = table.colnames[0]
180 # Handle multiple dataIds per file by grouping by file.
181 refs_by_file = defaultdict(list)
182 n_dataset_refs = 0
183 for row in table:
184 # Convert the row to a dataId, remembering to extract the
185 # path column.
186 dataId = dict(row)
187 path = dataId.pop(file_column)
189 # The command line can override a column.
190 dataId.update(common_data_id)
192 # Create the dataset ref that is to be ingested.
193 ref = DatasetRef(datasetType, dataId, run=run, id_generation_mode=id_generation_mode) # type: ignore
195 # Convert path to absolute (because otherwise system will
196 # assume relative to datastore root and that is almost certainly
197 # never the right default here).
198 path_uri = ResourcePath(path, root=prefix, forceAbsolute=True)
200 refs_by_file[path_uri].append(ref)
201 n_dataset_refs += 1
203 datasets = [
204 FileDataset(
205 path=file_uri,
206 refs=refs,
207 formatter=formatter,
208 )
209 for file_uri, refs in refs_by_file.items()
210 ]
212 log.info("Ingesting %d dataset ref(s) from %d file(s)", n_dataset_refs, len(datasets))
214 return datasets
217def parse_data_id_tuple(data_ids: tuple[str, ...], universe: DimensionUniverse) -> dict[str, Any]:
218 """Convert any additional k=v strings in the dataId tuple to dict
219 form.
221 Parameters
222 ----------
223 data_ids : `tuple` of `str`
224 Strings of keyword=value pairs defining a data ID.
225 universe : `DimensionUniverse`
226 The relevant universe.
228 Returns
229 -------
230 data_id : `dict`
231 Data ID transformed from string into dictionary.
232 """
233 data_id: dict[str, Any] = {}
234 for id_str in data_ids:
235 dimension_str, value = id_str.split("=")
237 try:
238 dimension = universe.dimensions[dimension_str]
239 except KeyError:
240 raise ValueError(f"DataID dimension '{dimension_str}' is not known to this universe.") from None
242 # Cast the value to the right python type (since they will be
243 # strings at this point).
244 value = dimension.primaryKey.getPythonType()(value)
246 data_id[dimension_str] = value
247 return data_id