compressed-lists 0.2.0__tar.gz → 0.3.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.pre-commit-config.yaml +1 -1
- compressed_lists-0.3.0/CHANGELOG.md +17 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/PKG-INFO +1 -1
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/tutorial.md +1 -1
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/__init__.py +1 -1
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/base.py +37 -12
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/biocframe_list.py +64 -10
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/bool_list.py +2 -2
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/float_list.py +2 -2
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/integer_list.py +2 -2
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/numpy_list.py +14 -8
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/partition.py +10 -6
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/split_generic.py +40 -10
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists/string_list.py +3 -3
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/PKG-INFO +1 -1
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/SOURCES.txt +3 -1
- compressed_lists-0.3.0/tests/test_base.py +132 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_biocframe.py +19 -5
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_int.py +2 -2
- compressed_lists-0.3.0/tests/test_generics.py +86 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_partitioning.py +38 -0
- compressed_lists-0.3.0/tests/test_type_validations.py +32 -0
- compressed_lists-0.2.0/CHANGELOG.md +0 -10
- compressed_lists-0.2.0/tests/test_generics.py +0 -24
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.coveragerc +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.github/workflows/publish-pypi.yml +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.github/workflows/run-tests.yml +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.gitignore +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/.readthedocs.yml +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/AUTHORS.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/CONTRIBUTING.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/LICENSE.txt +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/README.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/Makefile +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/_static/.gitignore +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/authors.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/changelog.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/conf.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/contributing.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/index.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/license.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/readme.md +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/docs/requirements.txt +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/pyproject.toml +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/setup.cfg +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/setup.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/dependency_links.txt +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/not-zip-safe +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/requires.txt +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/top_level.txt +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/conftest.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_bool.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_custom.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_float.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_numpy.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tests/test_comp_str.py +0 -0
- {compressed_lists-0.2.0 → compressed_lists-0.3.0}/tox.ini +0 -0
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
# Changelog
|
|
2
|
+
|
|
3
|
+
## Version 0.3.0
|
|
4
|
+
|
|
5
|
+
- Renamed `CompressedBiocFrameList` to `CompressedSplitBiocFrameList` since the current implementation can only handle dataframes with the same number and names of columns.
|
|
6
|
+
- Correctly link some of the annoying typehint errors.
|
|
7
|
+
- `to_list` and `unlist` now perform same operation as the R implementation.
|
|
8
|
+
- `splitAsCompressedList` has a more abstract implementation if no generic is implemented. If it fails, returns an error.
|
|
9
|
+
|
|
10
|
+
## Version 0.2.0
|
|
11
|
+
|
|
12
|
+
- Major changes to the package; Switch to typed lists from the biocutils package.
|
|
13
|
+
|
|
14
|
+
## Version 0.1.0 - 0.1.1
|
|
15
|
+
|
|
16
|
+
- Initial implementation of various classes - Partitioning and CompressedLists.
|
|
17
|
+
- Udpate documentation, tutorial and tests.
|
|
@@ -22,5 +22,5 @@ from .string_list import CompressedStringList, CompressedCharacterList
|
|
|
22
22
|
from .bool_list import CompressedBooleanList
|
|
23
23
|
from .float_list import CompressedFloatList
|
|
24
24
|
from .numpy_list import CompressedNumpyList
|
|
25
|
-
from .biocframe_list import
|
|
25
|
+
from .biocframe_list import CompressedSplitBiocFrameList
|
|
26
26
|
from .split_generic import splitAsCompressedList
|
|
@@ -144,7 +144,13 @@ class CompressedList:
|
|
|
144
144
|
output = f"{type(self).__name__}(number_of_elements={len(self)}"
|
|
145
145
|
output += ", unlist_data=" + ut.print_truncated_list(self._unlist_data)
|
|
146
146
|
output += ", partitioning=" + self._partitioning.__repr__()
|
|
147
|
-
|
|
147
|
+
|
|
148
|
+
_etype_name = "__unknown_class__"
|
|
149
|
+
if isinstance(self._element_type, str):
|
|
150
|
+
_etype_name = self._element_type
|
|
151
|
+
elif hasattr(self._element_type, "__name__"):
|
|
152
|
+
_etype_name = self._element_type.__name__
|
|
153
|
+
output += ", element_type=" + _etype_name
|
|
148
154
|
|
|
149
155
|
if len(self._element_metadata) > 0:
|
|
150
156
|
output += ", element_metadata=" + ut.print_truncated_dict(self._element_metadata)
|
|
@@ -162,7 +168,13 @@ class CompressedList:
|
|
|
162
168
|
"""
|
|
163
169
|
output = f"class: {type(self).__name__}\n"
|
|
164
170
|
|
|
165
|
-
|
|
171
|
+
_etype_name = "__unknown_class__"
|
|
172
|
+
if isinstance(self._element_type, str):
|
|
173
|
+
_etype_name = self._element_type
|
|
174
|
+
elif hasattr(self._element_type, "__name__"):
|
|
175
|
+
_etype_name = self._element_type.__name__
|
|
176
|
+
|
|
177
|
+
output += f"number of elements: ({len(self)}) of type: {_etype_name}\n"
|
|
166
178
|
|
|
167
179
|
output += f"unlist_data: {ut.print_truncated_list(self._unlist_data)}\n"
|
|
168
180
|
|
|
@@ -203,11 +215,11 @@ class CompressedList:
|
|
|
203
215
|
######>> names <<######
|
|
204
216
|
#######################
|
|
205
217
|
|
|
206
|
-
def get_names(self) -> Optional[ut.
|
|
218
|
+
def get_names(self) -> Optional[ut.Names]:
|
|
207
219
|
"""Get the names of list elements."""
|
|
208
220
|
return self._partitioning.get_names()
|
|
209
221
|
|
|
210
|
-
def set_names(self, names:
|
|
222
|
+
def set_names(self, names: Sequence[str], in_place: bool = False) -> "CompressedList":
|
|
211
223
|
"""Set the names of list elements.
|
|
212
224
|
|
|
213
225
|
names:
|
|
@@ -227,7 +239,7 @@ class CompressedList:
|
|
|
227
239
|
return output
|
|
228
240
|
|
|
229
241
|
@property
|
|
230
|
-
def names(self) -> Optional[ut.
|
|
242
|
+
def names(self) -> Optional[ut.Names]:
|
|
231
243
|
"""Alias for :py:attr:`~get_names`."""
|
|
232
244
|
return self._partitioning.get_names()
|
|
233
245
|
|
|
@@ -460,7 +472,7 @@ class CompressedList:
|
|
|
460
472
|
|
|
461
473
|
@classmethod
|
|
462
474
|
def from_list(
|
|
463
|
-
cls, lst:
|
|
475
|
+
cls, lst: Any, names: Optional[Union[ut.Names, Sequence[str]]] = None, metadata: Optional[dict] = None
|
|
464
476
|
) -> "CompressedList":
|
|
465
477
|
"""Create a CompressedList from a regular list.
|
|
466
478
|
|
|
@@ -497,27 +509,40 @@ class CompressedList:
|
|
|
497
509
|
######>> coercions <<######
|
|
498
510
|
###########################
|
|
499
511
|
|
|
500
|
-
def to_list(self) -> List[Any]:
|
|
512
|
+
def to_list(self) -> List[List[Any]]:
|
|
501
513
|
"""Convert to a regular Python list.
|
|
502
514
|
|
|
503
515
|
Returns:
|
|
504
516
|
A regular Python list with all elements.
|
|
505
517
|
"""
|
|
506
|
-
|
|
518
|
+
result = []
|
|
519
|
+
for i in range(len(self)):
|
|
520
|
+
_subset = list(self[i])
|
|
521
|
+
if len(_subset) == 0:
|
|
522
|
+
_subset = [None]
|
|
523
|
+
result.append(_subset)
|
|
524
|
+
|
|
525
|
+
return result
|
|
507
526
|
|
|
508
|
-
def
|
|
527
|
+
def as_list(self) -> List[List[Any]]:
|
|
528
|
+
"""Alias to :py:meth:`~to_list`"""
|
|
529
|
+
return self.to_list()
|
|
530
|
+
|
|
531
|
+
def unlist(self, use_names: bool = False) -> Any:
|
|
509
532
|
"""Get the underlying unlisted data.
|
|
510
533
|
|
|
511
534
|
Args:
|
|
512
535
|
use_names:
|
|
513
536
|
Whether to include names in the result if applicable.
|
|
514
537
|
|
|
515
|
-
Currently not used.
|
|
516
|
-
|
|
517
538
|
Returns:
|
|
518
539
|
The unlisted data.
|
|
519
540
|
"""
|
|
520
|
-
return
|
|
541
|
+
return (
|
|
542
|
+
self._unlist_data
|
|
543
|
+
if use_names is False
|
|
544
|
+
else self._unlist_data.set_names(self.get_partitioning().get_names(), in_place=False)
|
|
545
|
+
)
|
|
521
546
|
|
|
522
547
|
def relist(self, unlist_data: Any) -> "CompressedList":
|
|
523
548
|
"""Create a new `CompressedList` with the same partitioning but different data.
|
|
@@ -12,8 +12,10 @@ __copyright__ = "Jayaram Kancherla"
|
|
|
12
12
|
__license__ = "MIT"
|
|
13
13
|
|
|
14
14
|
|
|
15
|
-
class
|
|
16
|
-
"""CompressedList for BiocFrames.
|
|
15
|
+
class CompressedSplitBiocFrameList(CompressedList):
|
|
16
|
+
"""CompressedList for BiocFrames.
|
|
17
|
+
|
|
18
|
+
All BiocFrames are expected to have the same number and names of columns."""
|
|
17
19
|
|
|
18
20
|
def __init__(
|
|
19
21
|
self,
|
|
@@ -23,7 +25,7 @@ class CompressedBiocFrameList(CompressedList):
|
|
|
23
25
|
metadata: Optional[dict] = None,
|
|
24
26
|
**kwargs,
|
|
25
27
|
):
|
|
26
|
-
"""Initialize a
|
|
28
|
+
"""Initialize a CompressedSplitBiocFrameList.
|
|
27
29
|
|
|
28
30
|
Args:
|
|
29
31
|
unlist_data:
|
|
@@ -50,9 +52,12 @@ class CompressedBiocFrameList(CompressedList):
|
|
|
50
52
|
|
|
51
53
|
@classmethod
|
|
52
54
|
def from_list(
|
|
53
|
-
cls,
|
|
54
|
-
|
|
55
|
-
|
|
55
|
+
cls,
|
|
56
|
+
lst: List[BiocFrame],
|
|
57
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
58
|
+
metadata: Optional[dict] = None,
|
|
59
|
+
) -> "CompressedSplitBiocFrameList":
|
|
60
|
+
"""Create a `CompressedSplitBiocFrameList` from a regular list.
|
|
56
61
|
|
|
57
62
|
This concatenates the list of `BiocFrame` objects.
|
|
58
63
|
|
|
@@ -60,6 +65,8 @@ class CompressedBiocFrameList(CompressedList):
|
|
|
60
65
|
lst:
|
|
61
66
|
List of `BiocFrame` objects.
|
|
62
67
|
|
|
68
|
+
Must have the same number and names of columns.
|
|
69
|
+
|
|
63
70
|
names:
|
|
64
71
|
Optional names for list elements.
|
|
65
72
|
|
|
@@ -106,21 +113,68 @@ class CompressedBiocFrameList(CompressedList):
|
|
|
106
113
|
"Custom classes should implement their own `extract_range` method for slice operations"
|
|
107
114
|
) from e
|
|
108
115
|
|
|
116
|
+
##########################
|
|
117
|
+
######>> Printing <<######
|
|
118
|
+
##########################
|
|
119
|
+
|
|
120
|
+
def __repr__(self) -> str:
|
|
121
|
+
"""
|
|
122
|
+
Returns:
|
|
123
|
+
A string representation.
|
|
124
|
+
"""
|
|
125
|
+
output = f"{type(self).__name__}(number_of_elements={len(self)}"
|
|
126
|
+
output += ", unlist_data=" + self._unlist_data.__repr__()
|
|
127
|
+
output += ", partitioning=" + self._partitioning.__repr__()
|
|
128
|
+
output += (
|
|
129
|
+
", element_type=" + self._element_type.__name__
|
|
130
|
+
if not isinstance(self._element_type, str)
|
|
131
|
+
else self._element_type
|
|
132
|
+
)
|
|
133
|
+
|
|
134
|
+
if len(self._element_metadata) > 0:
|
|
135
|
+
output += ", element_metadata=" + ut.print_truncated_dict(self._element_metadata)
|
|
136
|
+
|
|
137
|
+
if len(self._metadata) > 0:
|
|
138
|
+
output += ", metadata=" + ut.print_truncated_dict(self._metadata)
|
|
139
|
+
|
|
140
|
+
output += ")"
|
|
141
|
+
return output
|
|
142
|
+
|
|
143
|
+
def __str__(self) -> str:
|
|
144
|
+
"""
|
|
145
|
+
Returns:
|
|
146
|
+
A pretty-printed string containing the contents of this object.
|
|
147
|
+
"""
|
|
148
|
+
output = f"class: {type(self).__name__}\n"
|
|
149
|
+
|
|
150
|
+
output += f"number of elements: ({len(self)}) of type: {self._element_type.__name__ if not isinstance(self._element_type, str) else self._element_type}\n"
|
|
151
|
+
|
|
152
|
+
output += f"unlist_data: {ut.print_truncated_list(self._unlist_data.get_column_names())}\n"
|
|
153
|
+
|
|
154
|
+
output += f"partitioning: {ut.print_truncated_list(self._partitioning)}\n"
|
|
155
|
+
|
|
156
|
+
output += f"element_metadata({str(len(self._element_metadata))}): {ut.print_truncated_list(list(self._element_metadata.keys()), sep=' ', include_brackets=False, transform=lambda y: y)}\n"
|
|
157
|
+
output += f"metadata({str(len(self._metadata))}): {ut.print_truncated_list(list(self._metadata.keys()), sep=' ', include_brackets=False, transform=lambda y: y)}\n"
|
|
158
|
+
|
|
159
|
+
return output
|
|
160
|
+
|
|
109
161
|
|
|
110
162
|
@splitAsCompressedList.register
|
|
111
163
|
def _(
|
|
112
164
|
data: BiocFrame,
|
|
113
165
|
groups_or_partitions: Union[list, Partitioning],
|
|
114
|
-
names: Optional[Sequence[str]] = None,
|
|
166
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
115
167
|
metadata: Optional[dict] = None,
|
|
116
|
-
) ->
|
|
168
|
+
) -> CompressedSplitBiocFrameList:
|
|
117
169
|
"""Handle lists of BiocFrame objects."""
|
|
118
170
|
|
|
119
171
|
partitioned_data, groups_or_partitions = _generic_register_helper(
|
|
120
172
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
121
173
|
)
|
|
122
174
|
|
|
123
|
-
if not isinstance(partitioned_data, BiocFrame):
|
|
175
|
+
if not isinstance(partitioned_data, BiocFrame) and len(partitioned_data) != 0:
|
|
124
176
|
partitioned_data = ut.relaxed_combine_rows(*partitioned_data)
|
|
125
177
|
|
|
126
|
-
return
|
|
178
|
+
return CompressedSplitBiocFrameList(
|
|
179
|
+
unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata
|
|
180
|
+
)
|
|
@@ -58,7 +58,7 @@ class CompressedBooleanList(CompressedList):
|
|
|
58
58
|
def _(
|
|
59
59
|
data: ut.BooleanList,
|
|
60
60
|
groups_or_partitions: Union[list, Partitioning],
|
|
61
|
-
names: Optional[Sequence[str]] = None,
|
|
61
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
62
62
|
metadata: Optional[dict] = None,
|
|
63
63
|
) -> CompressedBooleanList:
|
|
64
64
|
"""Handle lists of booleans."""
|
|
@@ -67,7 +67,7 @@ def _(
|
|
|
67
67
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
68
68
|
)
|
|
69
69
|
|
|
70
|
-
if not isinstance(partitioned_data, ut.BooleanList):
|
|
70
|
+
if not isinstance(partitioned_data, ut.BooleanList) and len(partitioned_data) != 0:
|
|
71
71
|
partitioned_data = ut.combine_sequences(*partitioned_data)
|
|
72
72
|
|
|
73
73
|
return CompressedBooleanList(unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata)
|
|
@@ -58,7 +58,7 @@ class CompressedFloatList(CompressedList):
|
|
|
58
58
|
def _(
|
|
59
59
|
data: ut.FloatList,
|
|
60
60
|
groups_or_partitions: Union[list, Partitioning],
|
|
61
|
-
names: Optional[Sequence[str]] = None,
|
|
61
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
62
62
|
metadata: Optional[dict] = None,
|
|
63
63
|
) -> CompressedFloatList:
|
|
64
64
|
"""Handle lists of floats."""
|
|
@@ -67,7 +67,7 @@ def _(
|
|
|
67
67
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
68
68
|
)
|
|
69
69
|
|
|
70
|
-
if not isinstance(partitioned_data, ut.FloatList):
|
|
70
|
+
if not isinstance(partitioned_data, ut.FloatList) and len(partitioned_data) != 0:
|
|
71
71
|
partitioned_data = ut.combine_sequences(*partitioned_data)
|
|
72
72
|
|
|
73
73
|
return CompressedFloatList(unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata)
|
|
@@ -58,7 +58,7 @@ class CompressedIntegerList(CompressedList):
|
|
|
58
58
|
def _(
|
|
59
59
|
data: ut.IntegerList,
|
|
60
60
|
groups_or_partitions: Union[list, Partitioning],
|
|
61
|
-
names: Optional[Sequence[str]] = None,
|
|
61
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
62
62
|
metadata: Optional[dict] = None,
|
|
63
63
|
) -> CompressedIntegerList:
|
|
64
64
|
"""Handle lists of integers."""
|
|
@@ -67,7 +67,7 @@ def _(
|
|
|
67
67
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
68
68
|
)
|
|
69
69
|
|
|
70
|
-
if not isinstance(partitioned_data, ut.IntegerList):
|
|
70
|
+
if not isinstance(partitioned_data, ut.IntegerList) and len(partitioned_data) != 0:
|
|
71
71
|
partitioned_data = ut.combine_sequences(*partitioned_data)
|
|
72
72
|
|
|
73
73
|
return CompressedIntegerList(unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata)
|
|
@@ -44,11 +44,14 @@ class CompressedNumpyList(CompressedList):
|
|
|
44
44
|
"""
|
|
45
45
|
|
|
46
46
|
if not isinstance(unlist_data, np.ndarray):
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
|
|
51
|
-
|
|
47
|
+
if len(unlist_data) == 0:
|
|
48
|
+
unlist_data = np.asarray([])
|
|
49
|
+
else:
|
|
50
|
+
try:
|
|
51
|
+
warn("trying to concatenate/coerce 'unlist_data' to a `np.ndarray`..")
|
|
52
|
+
unlist_data = np.concatenate(unlist_data)
|
|
53
|
+
except Exception as e:
|
|
54
|
+
raise TypeError("'unlist_data' must be an `np.ndarray`, provided ", type(unlist_data)) from e
|
|
52
55
|
|
|
53
56
|
super().__init__(
|
|
54
57
|
unlist_data, partitioning, element_type=np.ndarray, element_metadata=element_metadata, metadata=metadata
|
|
@@ -56,7 +59,10 @@ class CompressedNumpyList(CompressedList):
|
|
|
56
59
|
|
|
57
60
|
@classmethod
|
|
58
61
|
def from_list(
|
|
59
|
-
cls,
|
|
62
|
+
cls,
|
|
63
|
+
lst: List[np.ndarray],
|
|
64
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
65
|
+
metadata: Optional[dict] = None,
|
|
60
66
|
) -> "CompressedNumpyList":
|
|
61
67
|
"""
|
|
62
68
|
Create a `CompressedNumpyList` from a list of NumPy vectors.
|
|
@@ -88,7 +94,7 @@ class CompressedNumpyList(CompressedList):
|
|
|
88
94
|
def _(
|
|
89
95
|
data: np.ndarray,
|
|
90
96
|
groups_or_partitions: Union[list, Partitioning],
|
|
91
|
-
names: Optional[Sequence[str]] = None,
|
|
97
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
92
98
|
metadata: Optional[dict] = None,
|
|
93
99
|
) -> CompressedNumpyList:
|
|
94
100
|
"""Handle NumPy arrays."""
|
|
@@ -97,7 +103,7 @@ def _(
|
|
|
97
103
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
98
104
|
)
|
|
99
105
|
|
|
100
|
-
if not isinstance(partitioned_data, np.ndarray):
|
|
106
|
+
if not isinstance(partitioned_data, np.ndarray) and len(partitioned_data) != 0:
|
|
101
107
|
partitioned_data = ut.combine_sequences(*partitioned_data)
|
|
102
108
|
|
|
103
109
|
return CompressedNumpyList(unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata)
|
|
@@ -24,7 +24,9 @@ class Partitioning:
|
|
|
24
24
|
It keeps track of where each element begins and ends in the unlisted data.
|
|
25
25
|
"""
|
|
26
26
|
|
|
27
|
-
def __init__(
|
|
27
|
+
def __init__(
|
|
28
|
+
self, ends: Sequence[int], names: Optional[Union[ut.Names, Sequence[str]]] = None, validate: bool = True
|
|
29
|
+
):
|
|
28
30
|
"""Initialize a Partitioning object.
|
|
29
31
|
|
|
30
32
|
Args:
|
|
@@ -52,7 +54,9 @@ class Partitioning:
|
|
|
52
54
|
_validate_names(names, len(ends))
|
|
53
55
|
|
|
54
56
|
@classmethod
|
|
55
|
-
def from_lengths(
|
|
57
|
+
def from_lengths(
|
|
58
|
+
cls, lengths: Sequence[int], names: Optional[Union[ut.Names, Sequence[str]]] = None
|
|
59
|
+
) -> "Partitioning":
|
|
56
60
|
"""Create a Partitioning from a sequence of lengths.
|
|
57
61
|
|
|
58
62
|
Args:
|
|
@@ -69,7 +73,7 @@ class Partitioning:
|
|
|
69
73
|
return cls(ends, names)
|
|
70
74
|
|
|
71
75
|
@classmethod
|
|
72
|
-
def from_list(cls, lst: List, names: Optional[Sequence[str]] = None) -> "Partitioning":
|
|
76
|
+
def from_list(cls, lst: List, names: Optional[Union[ut.Names, Sequence[str]]] = None) -> "Partitioning":
|
|
73
77
|
"""Create a Partitioning from a list by using the lengths of each element.
|
|
74
78
|
|
|
75
79
|
Args:
|
|
@@ -146,7 +150,7 @@ class Partitioning:
|
|
|
146
150
|
"""Return the lengths of each partition."""
|
|
147
151
|
return self._ends - self._starts
|
|
148
152
|
|
|
149
|
-
def element_lengths(self) ->
|
|
153
|
+
def element_lengths(self) -> np.ndarray:
|
|
150
154
|
"""Alias for :py:attr:`~get_element_lengths`."""
|
|
151
155
|
return self.get_element_lengths()
|
|
152
156
|
|
|
@@ -216,7 +220,7 @@ class Partitioning:
|
|
|
216
220
|
"""Return the names of each partition."""
|
|
217
221
|
return self._names
|
|
218
222
|
|
|
219
|
-
def set_names(self, names: Optional[Sequence[str]], in_place: bool = False) -> "Partitioning":
|
|
223
|
+
def set_names(self, names: Optional[Union[ut.Names, Sequence[str]]], in_place: bool = False) -> "Partitioning":
|
|
220
224
|
"""Set the names of list elements.
|
|
221
225
|
|
|
222
226
|
Args:
|
|
@@ -247,7 +251,7 @@ class Partitioning:
|
|
|
247
251
|
return self.get_names()
|
|
248
252
|
|
|
249
253
|
@names.setter
|
|
250
|
-
def names(self, names: Optional[Sequence[str]]):
|
|
254
|
+
def names(self, names: Optional[Union[ut.Names, Sequence[str]]]):
|
|
251
255
|
"""Alias for :py:meth:`~set_names` with ``in_place = True``.
|
|
252
256
|
|
|
253
257
|
As this mutates the original object, a warning is raised.
|
|
@@ -2,9 +2,9 @@ from collections import defaultdict
|
|
|
2
2
|
from functools import singledispatch
|
|
3
3
|
from typing import Any, List, Optional, Sequence, Tuple, Union
|
|
4
4
|
|
|
5
|
+
import biocutils as ut
|
|
5
6
|
import numpy as np
|
|
6
7
|
|
|
7
|
-
from .base import CompressedList
|
|
8
8
|
from .partition import Partitioning
|
|
9
9
|
|
|
10
10
|
__author__ = "Jayaram Kancherla"
|
|
@@ -13,7 +13,7 @@ __license__ = "MIT"
|
|
|
13
13
|
|
|
14
14
|
|
|
15
15
|
def groups_to_partition(
|
|
16
|
-
data: Any, groups: list, names: Optional[Sequence[str]] = None
|
|
16
|
+
data: Any, groups: list, names: Optional[Union[ut.Names, Sequence[str]]] = None
|
|
17
17
|
) -> Tuple[List[Any], Partitioning]:
|
|
18
18
|
"""Convert group membership vector to partitioned data and Partitioning object.
|
|
19
19
|
|
|
@@ -33,9 +33,13 @@ def groups_to_partition(
|
|
|
33
33
|
if len(data) != len(groups):
|
|
34
34
|
raise ValueError(f"Length of data ({len(data)}) must match length of groups ({len(groups)})")
|
|
35
35
|
|
|
36
|
+
group_idx_dict = defaultdict(list)
|
|
37
|
+
for idx, group in enumerate(groups):
|
|
38
|
+
group_idx_dict[group].append(idx)
|
|
39
|
+
|
|
36
40
|
group_dict = defaultdict(list)
|
|
37
|
-
for
|
|
38
|
-
group_dict[
|
|
41
|
+
for key, val in group_idx_dict.items():
|
|
42
|
+
group_dict[key] = ut.subset(data, val)
|
|
39
43
|
|
|
40
44
|
sorted_groups = sorted(group_dict.keys())
|
|
41
45
|
partitioned_data = [group_dict[group] for group in sorted_groups]
|
|
@@ -50,7 +54,6 @@ def groups_to_partition(
|
|
|
50
54
|
group_names = names
|
|
51
55
|
|
|
52
56
|
partitioning = Partitioning.from_list(partitioned_data, group_names)
|
|
53
|
-
|
|
54
57
|
return partitioned_data, partitioning
|
|
55
58
|
|
|
56
59
|
|
|
@@ -58,9 +61,9 @@ def groups_to_partition(
|
|
|
58
61
|
def splitAsCompressedList(
|
|
59
62
|
data: Any,
|
|
60
63
|
groups_or_partitions: Union[list, Partitioning],
|
|
61
|
-
names: Optional[Sequence[str]] = None,
|
|
64
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
62
65
|
metadata: Optional[dict] = None,
|
|
63
|
-
) ->
|
|
66
|
+
) -> Any:
|
|
64
67
|
"""Generic function to split data into an appropriate `CompressedList` subclass.
|
|
65
68
|
|
|
66
69
|
This function can work in two modes:
|
|
@@ -85,14 +88,38 @@ def splitAsCompressedList(
|
|
|
85
88
|
An appropriate `CompressedList` subclass instance.
|
|
86
89
|
"""
|
|
87
90
|
element_type = type(data)
|
|
88
|
-
|
|
91
|
+
try:
|
|
92
|
+
if isinstance(groups_or_partitions, Partitioning):
|
|
93
|
+
partitioned_data = []
|
|
94
|
+
for i in range(len(groups_or_partitions)):
|
|
95
|
+
start, end = groups_or_partitions.get_partition_range(i)
|
|
96
|
+
_part = data[start:end]
|
|
97
|
+
partitioned_data.append(_part)
|
|
98
|
+
else:
|
|
99
|
+
if isinstance(groups_or_partitions, np.ndarray):
|
|
100
|
+
groups_or_partitions = groups_or_partitions.tolist()
|
|
101
|
+
|
|
102
|
+
group_idx_dict = defaultdict(list)
|
|
103
|
+
for idx, group in enumerate(groups_or_partitions):
|
|
104
|
+
group_idx_dict[group].append(idx)
|
|
105
|
+
|
|
106
|
+
group_dict = defaultdict(list)
|
|
107
|
+
for key, val in group_idx_dict.items():
|
|
108
|
+
group_dict[key] = ut.subset(data, val)
|
|
109
|
+
|
|
110
|
+
sorted_groups = sorted(group_dict.keys())
|
|
111
|
+
partitioned_data = [group_dict[group] for group in sorted_groups]
|
|
112
|
+
|
|
113
|
+
return partitioned_data
|
|
114
|
+
except Exception as e:
|
|
115
|
+
raise NotImplementedError(f"No `splitAsCompressedList` dispatcher found for element type {element_type}") from e
|
|
89
116
|
|
|
90
117
|
|
|
91
118
|
def _generic_register_helper(data, groups_or_partitions, names=None):
|
|
92
119
|
if groups_or_partitions is None:
|
|
93
120
|
raise ValueError("'groups_or_paritions' cannot be 'None'.")
|
|
94
121
|
|
|
95
|
-
if
|
|
122
|
+
if data is None:
|
|
96
123
|
raise ValueError("'data' cannot be empty.")
|
|
97
124
|
|
|
98
125
|
if isinstance(groups_or_partitions, Partitioning):
|
|
@@ -107,9 +134,12 @@ def _generic_register_helper(data, groups_or_partitions, names=None):
|
|
|
107
134
|
# partitioned_data.append(data[start:end])
|
|
108
135
|
partitioned_data = data
|
|
109
136
|
elif isinstance(groups_or_partitions, (list, np.ndarray)):
|
|
137
|
+
if isinstance(groups_or_partitions, np.ndarray):
|
|
138
|
+
groups_or_partitions = groups_or_partitions.tolist()
|
|
139
|
+
|
|
110
140
|
partitioned_data, groups_or_partitions = groups_to_partition(data, groups=groups_or_partitions, names=names)
|
|
111
141
|
|
|
112
|
-
if
|
|
142
|
+
if partitioned_data is None:
|
|
113
143
|
raise ValueError("No data after grouping")
|
|
114
144
|
else:
|
|
115
145
|
raise ValueError("'groups_or_paritions' must be a group vector or a Partition object.")
|
|
@@ -61,16 +61,16 @@ class CompressedCharacterList(CompressedStringList):
|
|
|
61
61
|
def _(
|
|
62
62
|
data: ut.StringList,
|
|
63
63
|
groups_or_partitions: Union[list, Partitioning],
|
|
64
|
-
names: Optional[Sequence[str]] = None,
|
|
64
|
+
names: Optional[Union[ut.Names, Sequence[str]]] = None,
|
|
65
65
|
metadata: Optional[dict] = None,
|
|
66
66
|
) -> CompressedStringList:
|
|
67
|
-
"""Handle lists of
|
|
67
|
+
"""Handle lists of strings."""
|
|
68
68
|
|
|
69
69
|
partitioned_data, groups_or_partitions = _generic_register_helper(
|
|
70
70
|
data=data, groups_or_partitions=groups_or_partitions, names=names
|
|
71
71
|
)
|
|
72
72
|
|
|
73
|
-
if not isinstance(partitioned_data, ut.StringList):
|
|
73
|
+
if not isinstance(partitioned_data, ut.StringList) and len(partitioned_data) != 0:
|
|
74
74
|
partitioned_data = ut.combine_sequences(*partitioned_data)
|
|
75
75
|
|
|
76
76
|
return CompressedStringList(unlist_data=partitioned_data, partitioning=groups_or_partitions, metadata=metadata)
|
|
@@ -41,6 +41,7 @@ src/compressed_lists.egg-info/not-zip-safe
|
|
|
41
41
|
src/compressed_lists.egg-info/requires.txt
|
|
42
42
|
src/compressed_lists.egg-info/top_level.txt
|
|
43
43
|
tests/conftest.py
|
|
44
|
+
tests/test_base.py
|
|
44
45
|
tests/test_comp_biocframe.py
|
|
45
46
|
tests/test_comp_bool.py
|
|
46
47
|
tests/test_comp_custom.py
|
|
@@ -49,4 +50,5 @@ tests/test_comp_int.py
|
|
|
49
50
|
tests/test_comp_numpy.py
|
|
50
51
|
tests/test_comp_str.py
|
|
51
52
|
tests/test_generics.py
|
|
52
|
-
tests/test_partitioning.py
|
|
53
|
+
tests/test_partitioning.py
|
|
54
|
+
tests/test_type_validations.py
|
|
@@ -0,0 +1,132 @@
|
|
|
1
|
+
from copy import copy, deepcopy
|
|
2
|
+
|
|
3
|
+
import pytest
|
|
4
|
+
|
|
5
|
+
from compressed_lists import CompressedList
|
|
6
|
+
|
|
7
|
+
__author__ = "Jayaram Kancherla"
|
|
8
|
+
__copyright__ = "Jayaram Kancherla"
|
|
9
|
+
__license__ = "MIT"
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
@pytest.fixture
|
|
13
|
+
def base_list_data():
|
|
14
|
+
return [[1, 2], [3], [4, 5, 6]]
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
@pytest.fixture
|
|
18
|
+
def base_list(base_list_data):
|
|
19
|
+
names = ["A", "B", "C"]
|
|
20
|
+
return CompressedList.from_list(base_list_data, names)
|
|
21
|
+
|
|
22
|
+
|
|
23
|
+
def test_base_empty_creation():
|
|
24
|
+
cl_empty = CompressedList.from_list([])
|
|
25
|
+
assert len(cl_empty) == 0
|
|
26
|
+
assert len(cl_empty.unlist_data) == 0
|
|
27
|
+
assert list(cl_empty.get_element_lengths()) == []
|
|
28
|
+
assert cl_empty.to_list() == []
|
|
29
|
+
|
|
30
|
+
|
|
31
|
+
def test_base_list_of_empty():
|
|
32
|
+
cl_list_empty = CompressedList.from_list([[], [], []])
|
|
33
|
+
assert len(cl_list_empty) == 3
|
|
34
|
+
assert len(cl_list_empty.unlist_data) == 0
|
|
35
|
+
assert list(cl_list_empty.get_element_lengths()) == [0, 0, 0]
|
|
36
|
+
assert list(cl_list_empty[0]) == []
|
|
37
|
+
assert list(cl_list_empty[1]) == []
|
|
38
|
+
assert list(cl_list_empty[2]) == []
|
|
39
|
+
|
|
40
|
+
|
|
41
|
+
def test_base_set_names(base_list):
|
|
42
|
+
new_names = ["X", "Y", "Z"]
|
|
43
|
+
|
|
44
|
+
cl_new_names = base_list.set_names(new_names, in_place=False)
|
|
45
|
+
assert list(base_list.names) == ["A", "B", "C"] # Original is unchanged
|
|
46
|
+
assert list(cl_new_names.names) == new_names
|
|
47
|
+
|
|
48
|
+
with pytest.warns(UserWarning, match="Setting property 'names'"):
|
|
49
|
+
base_list.names = new_names
|
|
50
|
+
assert list(base_list.names) == new_names
|
|
51
|
+
|
|
52
|
+
|
|
53
|
+
def test_base_set_unlist_data(base_list):
|
|
54
|
+
new_data = [10, 20, 30, 40, 50, 60]
|
|
55
|
+
assert len(new_data) == len(base_list.unlist_data)
|
|
56
|
+
|
|
57
|
+
cl_new_data = base_list.set_unlist_data(new_data, in_place=False)
|
|
58
|
+
assert base_list.unlist_data == [1, 2, 3, 4, 5, 6] # Original is unchanged
|
|
59
|
+
assert cl_new_data.unlist_data == new_data
|
|
60
|
+
assert list(cl_new_data[0]) == [10, 20]
|
|
61
|
+
|
|
62
|
+
with pytest.warns(UserWarning, match="Setting property 'unlist_data'"):
|
|
63
|
+
base_list.unlist_data = new_data
|
|
64
|
+
assert base_list.unlist_data == new_data
|
|
65
|
+
assert list(base_list[1]) == [30]
|
|
66
|
+
|
|
67
|
+
|
|
68
|
+
def test_base_set_unlist_data_error(base_list):
|
|
69
|
+
with pytest.raises(ValueError, match="Length of 'unlist_data'"):
|
|
70
|
+
base_list.set_unlist_data([1, 2, 3], in_place=False)
|
|
71
|
+
|
|
72
|
+
|
|
73
|
+
def test_base_metadata(base_list):
|
|
74
|
+
meta = {"source": "test"}
|
|
75
|
+
cl_meta = base_list.set_metadata(meta, in_place=False)
|
|
76
|
+
assert base_list.metadata == {}
|
|
77
|
+
assert cl_meta.metadata == {"source": "test"}
|
|
78
|
+
|
|
79
|
+
with pytest.warns(UserWarning, match="Setting property 'metadata'"):
|
|
80
|
+
base_list.metadata = meta
|
|
81
|
+
assert base_list.metadata == {"source": "test"}
|
|
82
|
+
|
|
83
|
+
el_meta = {"info": "details"}
|
|
84
|
+
cl_el_meta = base_list.set_element_metadata(el_meta, in_place=False)
|
|
85
|
+
assert base_list.element_metadata == {}
|
|
86
|
+
assert cl_el_meta.element_metadata == {"info": "details"}
|
|
87
|
+
|
|
88
|
+
with pytest.warns(UserWarning, match="Setting property 'element_metadata'"):
|
|
89
|
+
base_list.element_metadata = el_meta
|
|
90
|
+
assert base_list.element_metadata == {"info": "details"}
|
|
91
|
+
|
|
92
|
+
|
|
93
|
+
def test_base_copying(base_list):
|
|
94
|
+
cl_copy = copy(base_list)
|
|
95
|
+
assert cl_copy is not base_list
|
|
96
|
+
assert cl_copy.unlist_data is base_list.unlist_data
|
|
97
|
+
assert cl_copy.get_partitioning() is base_list.get_partitioning()
|
|
98
|
+
|
|
99
|
+
cl_deepcopy = deepcopy(base_list)
|
|
100
|
+
assert cl_deepcopy is not base_list
|
|
101
|
+
assert cl_deepcopy.unlist_data is not base_list.unlist_data
|
|
102
|
+
assert cl_deepcopy.unlist_data == base_list.unlist_data
|
|
103
|
+
assert cl_deepcopy.get_partitioning() is not base_list.get_partitioning()
|
|
104
|
+
assert list(cl_deepcopy.get_partitioning().get_ends()) == list(base_list.get_partitioning().get_ends())
|
|
105
|
+
|
|
106
|
+
|
|
107
|
+
def test_base_repr_str(base_list):
|
|
108
|
+
assert isinstance(repr(base_list), str)
|
|
109
|
+
assert "CompressedList" in repr(base_list)
|
|
110
|
+
assert "number_of_elements=3" in repr(base_list)
|
|
111
|
+
|
|
112
|
+
assert isinstance(str(base_list), str)
|
|
113
|
+
assert "class: CompressedList" in str(base_list)
|
|
114
|
+
assert "number of elements: (3)" in str(base_list)
|
|
115
|
+
|
|
116
|
+
|
|
117
|
+
def test_base_extract_subset_edge_cases(base_list, base_list_data):
|
|
118
|
+
sub_empty = base_list.extract_subset([])
|
|
119
|
+
assert len(sub_empty) == 0
|
|
120
|
+
assert len(sub_empty.unlist_data) == 0
|
|
121
|
+
|
|
122
|
+
sub_dup = base_list.extract_subset([0, 1, 0])
|
|
123
|
+
assert len(sub_dup) == 3
|
|
124
|
+
assert list(sub_dup.names) == ["A", "B", "A"]
|
|
125
|
+
assert sub_dup.to_list() == [base_list_data[0], base_list_data[1], base_list_data[0]]
|
|
126
|
+
assert sub_dup.unlist_data == [1, 2, 3, 1, 2]
|
|
127
|
+
|
|
128
|
+
sub_order = base_list.extract_subset([2, 0])
|
|
129
|
+
assert len(sub_order) == 2
|
|
130
|
+
assert list(sub_order.names) == ["C", "A"]
|
|
131
|
+
assert sub_order.to_list() == [base_list_data[2], base_list_data[0]]
|
|
132
|
+
assert sub_order.unlist_data == [4, 5, 6, 1, 2]
|
|
@@ -1,8 +1,8 @@
|
|
|
1
|
+
import biocutils as ut
|
|
1
2
|
import pytest
|
|
2
3
|
from biocframe import BiocFrame
|
|
3
|
-
import biocutils as ut
|
|
4
4
|
|
|
5
|
-
from compressed_lists import
|
|
5
|
+
from compressed_lists import CompressedSplitBiocFrameList, CompressedStringList, Partitioning, splitAsCompressedList
|
|
6
6
|
|
|
7
7
|
__author__ = "Jayaram Kancherla"
|
|
8
8
|
__copyright__ = "Jayaram Kancherla"
|
|
@@ -20,9 +20,9 @@ def frame_data():
|
|
|
20
20
|
|
|
21
21
|
|
|
22
22
|
def test_creation(frame_data):
|
|
23
|
-
frame_list =
|
|
23
|
+
frame_list = CompressedSplitBiocFrameList(frame_data, partitioning=Partitioning.from_lengths([1, 2]))
|
|
24
24
|
|
|
25
|
-
assert isinstance(frame_list,
|
|
25
|
+
assert isinstance(frame_list, CompressedSplitBiocFrameList)
|
|
26
26
|
assert len(frame_list) == 2
|
|
27
27
|
assert isinstance(frame_list.unlist_data, BiocFrame)
|
|
28
28
|
assert len(frame_list.get_unlist_data()) == 3
|
|
@@ -37,8 +37,22 @@ def test_bframe_typed_list_column():
|
|
|
37
37
|
"symbol": ["MAP1A", "BIN1", "ESR1"],
|
|
38
38
|
}
|
|
39
39
|
)
|
|
40
|
-
frame_list =
|
|
40
|
+
frame_list = CompressedSplitBiocFrameList(bframe, partitioning=Partitioning.from_lengths([1, 2]))
|
|
41
41
|
|
|
42
42
|
ens_col = frame_list["ensembl"]
|
|
43
43
|
assert isinstance(ens_col, CompressedStringList)
|
|
44
44
|
assert len(ens_col) == 2
|
|
45
|
+
|
|
46
|
+
|
|
47
|
+
def test_split_biocframe(frame_data):
|
|
48
|
+
frame_data.set_column("groups", [0, 0, 1], in_place=True)
|
|
49
|
+
print(frame_data)
|
|
50
|
+
clist = splitAsCompressedList(frame_data, groups_or_partitions=frame_data.get_column("groups"))
|
|
51
|
+
|
|
52
|
+
assert isinstance(clist, CompressedSplitBiocFrameList)
|
|
53
|
+
|
|
54
|
+
val = clist.__repr__()
|
|
55
|
+
assert isinstance(val, str)
|
|
56
|
+
|
|
57
|
+
val = clist.__str__()
|
|
58
|
+
assert isinstance(val, str)
|
|
@@ -88,13 +88,13 @@ def test_getitem_by_slice(int_list):
|
|
|
88
88
|
def test_iteration(int_list, int_data):
|
|
89
89
|
items = list(int_list)
|
|
90
90
|
for i, lst in enumerate(items):
|
|
91
|
-
assert np.allclose(
|
|
91
|
+
assert np.allclose(lst, int_data[i])
|
|
92
92
|
|
|
93
93
|
|
|
94
94
|
def test_to_list(int_list, int_data):
|
|
95
95
|
regular_list = int_list.to_list()
|
|
96
96
|
for i, lst in enumerate(regular_list):
|
|
97
|
-
assert np.allclose(
|
|
97
|
+
assert np.allclose(lst, int_data[i])
|
|
98
98
|
|
|
99
99
|
|
|
100
100
|
def test_unlist(int_list):
|
|
@@ -0,0 +1,86 @@
|
|
|
1
|
+
import biocutils as ut
|
|
2
|
+
import numpy as np
|
|
3
|
+
import pytest
|
|
4
|
+
|
|
5
|
+
from compressed_lists import (
|
|
6
|
+
CompressedFloatList,
|
|
7
|
+
CompressedIntegerList,
|
|
8
|
+
CompressedNumpyList,
|
|
9
|
+
Partitioning,
|
|
10
|
+
splitAsCompressedList,
|
|
11
|
+
)
|
|
12
|
+
from compressed_lists.split_generic import _generic_register_helper, groups_to_partition
|
|
13
|
+
|
|
14
|
+
__author__ = "Jayaram Kancherla"
|
|
15
|
+
__copyright__ = "Jayaram Kancherla"
|
|
16
|
+
__license__ = "MIT"
|
|
17
|
+
|
|
18
|
+
|
|
19
|
+
def test_groups():
|
|
20
|
+
float_vec = ut.FloatList([1.1, 1.2, 2.1, 2.2, 2.3, 3.0])
|
|
21
|
+
groups = [1, 2, 3, 1, 2, 3]
|
|
22
|
+
|
|
23
|
+
clist = splitAsCompressedList(float_vec, groups_or_partitions=groups)
|
|
24
|
+
|
|
25
|
+
assert isinstance(clist, CompressedFloatList)
|
|
26
|
+
|
|
27
|
+
|
|
28
|
+
def test_partitions():
|
|
29
|
+
int_list = splitAsCompressedList(
|
|
30
|
+
ut.IntegerList([1, 2, 3, 4, 5, 6, 7, 8, 9]), groups_or_partitions=Partitioning(ends=[3, 5, 9])
|
|
31
|
+
)
|
|
32
|
+
|
|
33
|
+
assert isinstance(int_list, CompressedIntegerList)
|
|
34
|
+
|
|
35
|
+
|
|
36
|
+
def test_groups_to_partition_value_error():
|
|
37
|
+
with pytest.raises(ValueError, match="must match length of groups"):
|
|
38
|
+
groups_to_partition([1, 2, 3], [1, 2], None)
|
|
39
|
+
|
|
40
|
+
|
|
41
|
+
def test_groups_to_partition_empty():
|
|
42
|
+
data = []
|
|
43
|
+
groups = []
|
|
44
|
+
part_data, part = groups_to_partition(data, groups)
|
|
45
|
+
assert part_data == []
|
|
46
|
+
assert len(part) == 0
|
|
47
|
+
|
|
48
|
+
|
|
49
|
+
def test_splitAsCompressedList_unsupported_type():
|
|
50
|
+
class UnregisteredType:
|
|
51
|
+
def __init__(self):
|
|
52
|
+
self.data = [1, 2]
|
|
53
|
+
|
|
54
|
+
def __len__(self):
|
|
55
|
+
return len(self.data)
|
|
56
|
+
|
|
57
|
+
def __getitem__(self, i):
|
|
58
|
+
return self.data[i]
|
|
59
|
+
|
|
60
|
+
data = UnregisteredType()
|
|
61
|
+
with pytest.raises(NotImplementedError, match="No `splitAsCompressedList` dispatcher found"):
|
|
62
|
+
splitAsCompressedList(data, groups_or_partitions=[0, 0])
|
|
63
|
+
|
|
64
|
+
|
|
65
|
+
def test_splitAsCompressedList_empty():
|
|
66
|
+
data_int = ut.IntegerList([])
|
|
67
|
+
groups_int = []
|
|
68
|
+
clist_int = splitAsCompressedList(data_int, groups_int)
|
|
69
|
+
assert isinstance(clist_int, CompressedIntegerList)
|
|
70
|
+
assert len(clist_int) == 0
|
|
71
|
+
assert len(clist_int.unlist_data) == 0
|
|
72
|
+
|
|
73
|
+
data_np = np.array([])
|
|
74
|
+
groups_np = []
|
|
75
|
+
clist_np = splitAsCompressedList(data_np, groups_np)
|
|
76
|
+
assert isinstance(clist_np, CompressedNumpyList)
|
|
77
|
+
assert len(clist_np) == 0
|
|
78
|
+
assert len(clist_np.unlist_data) == 0
|
|
79
|
+
|
|
80
|
+
|
|
81
|
+
def test_generic_register_helper_errors():
|
|
82
|
+
with pytest.raises(ValueError, match="'groups_or_paritions' cannot be 'None'"):
|
|
83
|
+
_generic_register_helper([1, 2, 3], None)
|
|
84
|
+
|
|
85
|
+
with pytest.raises(ValueError, match="'groups_or_paritions' must be a group vector or a Partition object"):
|
|
86
|
+
_generic_register_helper([1, 2, 3], {"a": 1})
|
|
@@ -78,3 +78,41 @@ def test_getitem():
|
|
|
78
78
|
|
|
79
79
|
with pytest.raises(TypeError):
|
|
80
80
|
part["invalid"]
|
|
81
|
+
|
|
82
|
+
|
|
83
|
+
def test_partitioning_empty():
|
|
84
|
+
part = Partitioning([])
|
|
85
|
+
assert len(part) == 0
|
|
86
|
+
assert part.nobj() == 0
|
|
87
|
+
assert list(part.get_element_lengths()) == []
|
|
88
|
+
assert part.get_names() is None
|
|
89
|
+
|
|
90
|
+
|
|
91
|
+
def test_partitioning_from_list_non_sequence():
|
|
92
|
+
# The implementation checks for __len__ or assumes 1
|
|
93
|
+
lst = [1, [2, 3], "foo", (4, 5, 6)]
|
|
94
|
+
part = Partitioning.from_list(lst)
|
|
95
|
+
assert list(part.get_element_lengths()) == [1, 2, 3, 3]
|
|
96
|
+
assert list(part.ends) == [1, 3, 6, 9]
|
|
97
|
+
|
|
98
|
+
|
|
99
|
+
def test_partitioning_getitem_slices():
|
|
100
|
+
part = Partitioning.from_lengths([2, 3, 5, 1])
|
|
101
|
+
|
|
102
|
+
assert part[:] == [(0, 2), (2, 5), (5, 10), (10, 11)]
|
|
103
|
+
assert part[-2:] == [(5, 10), (10, 11)]
|
|
104
|
+
assert part[::2] == [(0, 2), (5, 10)]
|
|
105
|
+
|
|
106
|
+
|
|
107
|
+
def test_partitioning_set_names_inplace():
|
|
108
|
+
part = Partitioning.from_lengths([2, 3], names=["A", "B"])
|
|
109
|
+
|
|
110
|
+
part_new = part.set_names(["X", "Y"], in_place=False)
|
|
111
|
+
assert list(part.names) == ["A", "B"]
|
|
112
|
+
assert list(part_new.names) == ["X", "Y"]
|
|
113
|
+
|
|
114
|
+
part.set_names(["X", "Y"], in_place=True)
|
|
115
|
+
assert list(part.names) == ["X", "Y"]
|
|
116
|
+
|
|
117
|
+
with pytest.raises(ValueError, match="Length of names must match"):
|
|
118
|
+
part.set_names(["X"], in_place=True)
|
|
@@ -0,0 +1,32 @@
|
|
|
1
|
+
import biocutils as ut
|
|
2
|
+
import pytest
|
|
3
|
+
|
|
4
|
+
from compressed_lists import (
|
|
5
|
+
CompressedBooleanList,
|
|
6
|
+
CompressedFloatList,
|
|
7
|
+
CompressedIntegerList,
|
|
8
|
+
CompressedStringList,
|
|
9
|
+
Partitioning,
|
|
10
|
+
)
|
|
11
|
+
|
|
12
|
+
__author__ = "Jayaram Kancherla"
|
|
13
|
+
__copyright__ = "Jayaram Kancherla"
|
|
14
|
+
__license__ = "MIT"
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
@pytest.mark.parametrize(
|
|
18
|
+
"Cls, data, unlist_type",
|
|
19
|
+
[
|
|
20
|
+
(CompressedIntegerList, [1, 2, 3], ut.IntegerList),
|
|
21
|
+
(CompressedFloatList, [1.1, 2.2], ut.FloatList),
|
|
22
|
+
(CompressedBooleanList, [True, False, True], ut.BooleanList),
|
|
23
|
+
(CompressedStringList, ["a", "b", "c"], ut.StringList),
|
|
24
|
+
],
|
|
25
|
+
)
|
|
26
|
+
def test_coercion_warning(Cls, data, unlist_type):
|
|
27
|
+
part = Partitioning.from_lengths([len(data)])
|
|
28
|
+
with pytest.warns(UserWarning, match="trying to coerce"):
|
|
29
|
+
clist = Cls(data, part)
|
|
30
|
+
|
|
31
|
+
assert isinstance(clist.unlist_data, unlist_type)
|
|
32
|
+
assert list(clist.unlist_data) == data
|
|
@@ -1,10 +0,0 @@
|
|
|
1
|
-
# Changelog
|
|
2
|
-
|
|
3
|
-
## Version 0.2.0
|
|
4
|
-
|
|
5
|
-
- Major changes to the package; Switch to typed lists from the biocutils package.
|
|
6
|
-
|
|
7
|
-
## Version 0.1.0 - 0.1.1
|
|
8
|
-
|
|
9
|
-
- Initial implementation of various classes - Partitioning and CompressedLists.
|
|
10
|
-
- Udpate documentation, tutorial and tests.
|
|
@@ -1,24 +0,0 @@
|
|
|
1
|
-
import biocutils as ut
|
|
2
|
-
|
|
3
|
-
from compressed_lists import CompressedFloatList, CompressedIntegerList, Partitioning, splitAsCompressedList
|
|
4
|
-
|
|
5
|
-
__author__ = "Jayaram Kancherla"
|
|
6
|
-
__copyright__ = "Jayaram Kancherla"
|
|
7
|
-
__license__ = "MIT"
|
|
8
|
-
|
|
9
|
-
|
|
10
|
-
def test_groups():
|
|
11
|
-
float_vec = ut.FloatList([1.1, 1.2, 2.1, 2.2, 2.3, 3.0])
|
|
12
|
-
groups = [1, 2, 3, 1, 2, 3]
|
|
13
|
-
|
|
14
|
-
clist = splitAsCompressedList(float_vec, groups_or_partitions=groups)
|
|
15
|
-
|
|
16
|
-
assert isinstance(clist, CompressedFloatList)
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
def test_partitions():
|
|
20
|
-
int_list = splitAsCompressedList(
|
|
21
|
-
ut.IntegerList([1, 2, 3, 4, 5, 6, 7, 8, 9]), groups_or_partitions=Partitioning(ends=[3, 5, 9])
|
|
22
|
-
)
|
|
23
|
-
|
|
24
|
-
assert isinstance(int_list, CompressedIntegerList)
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/dependency_links.txt
RENAMED
|
File without changes
|
{compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/not-zip-safe
RENAMED
|
File without changes
|
{compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/requires.txt
RENAMED
|
File without changes
|
{compressed_lists-0.2.0 → compressed_lists-0.3.0}/src/compressed_lists.egg-info/top_level.txt
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|