-
Notifications
You must be signed in to change notification settings - Fork 13
/
hdfio.py
1455 lines (1198 loc) · 47.9 KB
/
hdfio.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# coding: utf-8
# Copyright (c) Max-Planck-Institut für Eisenforschung GmbH - Computational Materials Design (CM) Department
# Distributed under the terms of "New BSD License", see the LICENSE file.
"""
Classes to map the Python objects to HDF5 data structures
"""
import numbers
from h5io_browser import Pointer
from h5io_browser.base import (
_open_hdf,
_is_ragged_in_1st_dim_only,
_read_hdf,
_write_hdf5_with_json_support,
)
import os
import importlib
import pandas
import posixpath
import numpy as np
import sys
from typing import Union, Optional, Any, Tuple
from pyiron_base.utils.deprecate import deprecate
from pyiron_base.storage.helper_functions import (
get_h5_path,
list_groups_and_nodes,
read_dict_from_hdf,
)
from pyiron_base.interfaces.has_groups import HasGroups
from pyiron_base.state import state
from pyiron_base.jobs.job.util import _get_safe_job_name
from pyiron_base.utils.instance import static_isinstance
__author__ = "Joerg Neugebauer, Jan Janssen"
__copyright__ = (
"Copyright 2020, Max-Planck-Institut für Eisenforschung GmbH - "
"Computational Materials Design (CM) Department"
)
__version__ = "1.0"
__maintainer__ = "Jan Janssen"
__email__ = "janssen@mpie.de"
__status__ = "production"
__date__ = "Sep 1, 2017"
# for historic reasons we write str(class) into the HDF 'TYPE' field of objects, so we need to parse this back out
def _extract_fully_qualified_name(type_field: str) -> str:
return type_field.split("'")[1]
def _extract_module_class_name(type_field: str) -> Tuple[str, str]:
fully_qualified_path = _extract_fully_qualified_name(type_field)
return fully_qualified_path.rsplit(".", maxsplit=1)
def _import_class(module_path, class_name):
"""
Import given class from fully qualified name and return class object.
Args:
module_path (str): fully qualified name of a pyiron class
class_name (str): fully qualified name of a pyiron class
Returns:
type: class object of the given name
"""
# ugly dynamic import, but only needed to log the warning anyway
from pyiron_base.jobs.job.jobtype import JobTypeChoice
job_class_dict = JobTypeChoice().job_class_dict # access global singleton
if class_name in job_class_dict:
known_module_path = job_class_dict[class_name]
# entries in the job_class_dict are either strings of modules or fully
# loaded class object; in the latter case our work here is done we just
# return the class
if isinstance(known_module_path, type):
return known_module_path
if module_path != known_module_path:
state.logger.info(
f'Using registered module "{known_module_path}" instead of custom/old module "{module_path}" to'
f' import job type "{class_name}"!'
)
module_path = known_module_path
try:
return getattr(
importlib.import_module(module_path),
class_name,
)
except ImportError:
import pyiron_base.project.maintenance
if module_path in pyiron_base.project.maintenance._MODULE_CONVERSION_DICT:
raise RuntimeError(
f"Could not import {class_name} from {module_path}, but module path known to have changed. "
"Call project.maintenance.local.update_hdf_types() to upgrade storage!"
) from None
else:
raise
def _to_object(hdf, class_name=None, **kwargs):
"""
Load the full pyiron object from an HDF5 file
Args:
class_name(str, optional): if the 'TYPE' node is not available in
the HDF5 file a manual object type can be set,
must be as reported by `str(type(obj))`
**kwargs: optional parameters optional parameters to override init
parameters
Returns:
pyiron object of the given class_name
"""
if "TYPE" not in hdf.list_nodes() and class_name is None:
raise ValueError("Objects can be only recovered from hdf5 if TYPE is given")
elif class_name is not None and class_name != hdf.get("TYPE"):
raise ValueError(
"Object type in hdf5-file must be identical to input parameter"
)
type_field = class_name or hdf.get("TYPE")
module_path, class_name = _extract_module_class_name(type_field)
class_object = _import_class(module_path, class_name)
# Backwards compatibility since the format of TYPE changed
if type_field != str(class_object):
hdf["TYPE"] = str(class_object)
if hasattr(class_object, "from_hdf_args"):
init_args = class_object.from_hdf_args(hdf)
else:
init_args = {}
init_args.update(kwargs)
obj = class_object(**init_args)
obj.from_hdf(hdf=hdf.open(".."), group_name=hdf.h5_path.split("/")[-1])
if static_isinstance(obj=obj, obj_type="pyiron_base.jobs.job.generic.GenericJob"):
module_name = module_path.split(".")[0]
module = importlib.import_module(module_name)
if hasattr(module, "Project"):
obj.project_hdf5._project = getattr(module, "Project")(
obj.project_hdf5.project.path
)
return obj
class FileHDFio(HasGroups, Pointer):
"""
Class that provides all info to access a h5 file. This class is based on h5io.py, which allows to
get and put a large variety of jobs to/from h5
Implements :class:`.HasGroups`. Groups are HDF groups in the file, nodes are HDF datasets.
Args:
file_name (str): absolute path of the HDF5 file
h5_path (str): absolute path inside the h5 path - starting from the root group
mode (str): mode : {'a', 'w', 'r', 'r+'}, default 'a'
See HDFStore docstring or tables.open_file for info about modes
.. attribute:: file_name
absolute path to the HDF5 file
.. attribute:: h5_path
path inside the HDF5 file - also stored as absolute path
.. attribute:: history
previously opened groups / folders
.. attribute:: file_exists
boolean if the HDF5 was already written
.. attribute:: base_name
name of the HDF5 file but without any file extension
.. attribute:: file_path
directory where the HDF5 file is located
.. attribute:: is_root
boolean if the HDF5 object is located at the root level of the HDF5 file
.. attribute:: is_open
boolean if the HDF5 file is currently opened - if an active file handler exists
.. attribute:: is_empty
boolean if the HDF5 file is empty
"""
def __init__(self, file_name, h5_path="/", mode="a"):
Pointer.__init__(self=self, file_name=file_name, h5_path=h5_path)
self.history = []
self._filter = ["groups", "nodes", "objects"]
# MutableMapping Impl
def __contains__(self, item):
nodes_groups = self.list_all()
return item in nodes_groups["nodes"] or item in nodes_groups["groups"]
def __len__(self):
nodes_groups = self.list_all()
return len(nodes_groups["nodes"]) + len(nodes_groups["groups"])
def __iter__(self):
return iter(self.keys())
def __getitem__(self, item):
"""
Get/ read data from the HDF5 file
Args:
item (str, slice): path to the data or key of the data object
Returns:
dict, list, float, int: data or data object
"""
if isinstance(item, slice):
if not (item.start or item.stop or item.step):
return self.values()
raise NotImplementedError("Implement if needed, e.g. for [:]")
else:
try:
# fast path, a good amount of accesses will want to fetch a specific dataset it knows exists in the
# file, there's therefor no point in checking whether item is a group or a node or even worse recursing
# in case when item contains '/'. In most cases read_hdf5 will grab the correct data straight away and
# if not we will still check thoroughly below. Since list_nodes()/list_groups() each open the
# underlying file once, this reduces the number of file opens in the most-likely case from 2 to 1 (1 to
# check whether the data is there and 1 to read it) and increases in the worst case from 1 to 2 (1 to
# try to read it here and one more time to verify it's not a group below).
return _read_hdf(
hdf_filehandle=self.file_name, h5_path=self._get_h5_path(item)
)
except (ValueError, OSError, RuntimeError, NotImplementedError):
# h5io couldn't find a dataset with name item, but there still might be a group with that name, which we
# check in the rest of the method
pass
item_lst = item.split("/")
if len(item_lst) == 1 and item_lst[0] != "..":
# if item in self.list_nodes() we would have caught it in the fast path above
if item in self.list_groups():
with self.open(item) as hdf_item:
obj = hdf_item.copy()
if self._is_convertable_dtype_object_array(obj):
obj = self._convert_dtype_obj_array(obj)
return obj
raise ValueError(
"Unknown item: {} {} {}".format(item, self.file_name, self.h5_path)
)
else:
if (
item_lst[0] == ""
): # item starting with '/', thus we have an absoute HDF5 path
item_abs_lst = os.path.normpath(item).replace("\\", "/").split("/")
else: # relative HDF5 path
# The self.h5_path is an absolute path (/h5_path/in/h5/file), however, to
# reach any directory super to root, we start with a
# relative path = ./h5_path/in/h5/file and add whatever we get as item.
# The normpath finally returns a path to the item which is relative to the hdf-root.
item_abs_lst = (
os.path.normpath(os.path.join("." + self.h5_path, item))
.replace("\\", "/")
.split("/")
)
# print('h5_path=', self.h5_path, 'item=', item, 'item_abs_lst=', item_abs_lst)
if item_abs_lst[0] == "." and len(item_abs_lst) == 1:
# Here, we are asked to return the root of the HDF5-file. The resulting self.path would be the
# same as the self.file_path and, thus, the path of the pyiron Project this HDF5-file belongs to:
return self.create_project_from_hdf5()
elif item_abs_lst[0] == "..":
# Here, we are asked to return a path super to the root of the HDF5-file, a.k.a. the path of it's
# pyiron Project, thus we pass the relative path to the pyiron Project to handle it:
return self.create_project_from_hdf5()["/".join(item_abs_lst)]
else:
hdf_object = self.copy()
hdf_object.h5_path = "/".join(item_abs_lst[:-1])
return hdf_object[item_abs_lst[-1]]
# TODO: remove this function upon 1.0.0 release
@staticmethod
def _is_convertable_dtype_object_array(obj):
if isinstance(obj, np.ndarray) and obj.dtype == np.dtype(object):
first_element = obj[(0,) * obj.ndim]
last_element = obj[(-1,) * obj.ndim]
if (
isinstance(first_element, numbers.Number)
and isinstance(last_element, numbers.Number)
and not _is_ragged_in_1st_dim_only(obj)
):
return True
return False
# TODO: remove this function upon 1.0.0 release
@staticmethod
def _convert_dtype_obj_array(obj: np.ndarray):
try:
result = np.array(obj.tolist())
except ValueError:
result = np.array(obj.tolist(), dtype=object)
if result.dtype != np.dtype(object):
state.logger.warning(
f"Deprecated data structure! "
f"Returned array was converted from dtype='O' to dtype={result.dtype} "
f"via `np.array(result.tolist())`.\n"
f"Please run rewrite_hdf5() (from a job: job.project_hdf5.rewrite_hdf5() ) to update this data! "
f"To update all your data run Project.maintenance.update.base_v0_3_to_v0_4('all')."
)
return result
else:
return obj
def __setitem__(self, key, value):
"""
Store data inside the HDF5 file
Args:
key (str): key to store the data
value (pandas.DataFrame, pandas.Series, dict, list, float, int): basically any kind of data is supported
"""
if hasattr(value, "to_hdf") & (
not isinstance(value, (pandas.DataFrame, pandas.Series))
):
value.to_hdf(self, key)
return
_write_hdf5_with_json_support(
hdf_filehandle=self.file_name,
h5_path=self._get_h5_path(key),
data=value,
)
@property
def base_name(self):
"""
Name of the HDF5 file - but without the file extension .h5
Returns:
str: file name without the file extension
"""
return ".".join(posixpath.basename(self.file_name).split(".")[:-1])
@property
def file_path(self):
"""
Path where the HDF5 file is located - posixpath.dirname()
Returns:
str: HDF5 file location
"""
return posixpath.dirname(self.file_name)
def get_size(self, hdf):
"""
Get size of the groups inside the HDF5 file
Args:
hdf (FileHDFio): hdf file
Returns:
float: file size in Bytes
"""
return sum([sys.getsizeof(hdf[p]) for p in hdf.list_nodes()]) + sum(
[self.get_size(hdf[p]) for p in hdf.list_groups()]
)
def copy(self):
"""
Copy the Python object which links to the HDF5 file - in contrast to copy_to() which copies the content of the
HDF5 file to a new location.
Returns:
FileHDFio: New FileHDFio object pointing to the same HDF5 file
"""
new_h5 = FileHDFio(file_name=self.file_name, h5_path=self.h5_path)
new_h5._filter = self._filter
return new_h5
def create_group(self, name, track_order=False):
"""
Create an HDF5 group - similar to a folder in the filesystem - the HDF5 groups allow the users to structure
their data.
Args:
name (str): name of the HDF5 group
track_order (bool): if False this groups tracks its elements in
alphanumeric order, if True in insertion order
Returns:
FileHDFio: FileHDFio object pointing to the new group
"""
full_name = self._get_h5_path(name)
with _open_hdf(self.file_name, mode="a") as h:
try:
h.create_group(full_name, track_order=track_order)
except ValueError:
pass
h_new = self[name].copy()
return h_new
def remove_group(self):
"""
Remove an HDF5 group - if it exists. If the group does not exist no error message is raised.
"""
try:
with _open_hdf(self.file_name, mode="a") as hdf_file:
del hdf_file[self.h5_path]
except KeyError:
pass
def open(self, h5_rel_path):
"""
Create an HDF5 group and enter this specific group. If the group exists in the HDF5 path only the h5_path is
set correspondingly otherwise the group is created first.
Args:
h5_rel_path (str): relative path from the current HDF5 path - h5_path - to the new group
Returns:
FileHDFio: FileHDFio object pointing to the new group
"""
new_h5_path = self.copy()
if os.path.isabs(h5_rel_path):
raise ValueError(
"Absolute paths are not supported -> replace by relative path name!"
)
if h5_rel_path.strip() == ".":
h5_rel_path = ""
if h5_rel_path.strip() != "":
new_h5_path.h5_path = self._get_h5_path(h5_rel_path)
new_h5_path.history.append(h5_rel_path)
return new_h5_path
def close(self):
"""
Close the current HDF5 path and return to the path before the last open
"""
path_lst = self.h5_path.split("/")
last = self.history[-1].strip()
if len(last) > 0:
hist_lst = last.split("/")
self.h5_path = "/".join(path_lst[: -len(hist_lst)])
if len(self.h5_path.strip()) == 0:
self.h5_path = "/"
del self.history[-1]
def show_hdf(self):
"""
Iterating over the HDF5 datastructure and generating a human readable graph.
"""
self._walk()
def remove_file(self):
"""
Remove the HDF5 file with all the related content
"""
if self.file_exists:
os.remove(self.file_name)
def get_from_table(self, path, name):
"""
Get a specific value from a pandas.Dataframe
Args:
path (str): relative path to the data object
name (str): parameter key
Returns:
dict, list, float, int: the value associated to the specific parameter key
"""
df_table = self.get(path)
keys = df_table["Parameter"]
if name in keys:
job_id = keys.index(name)
return df_table["Value"][job_id]
raise ValueError("Unknown name: {0}".format(name))
def get_pandas(self, name):
"""
Load a dictionary from the HDF5 file and display the dictionary as pandas Dataframe
Args:
name (str): HDF5 node name
Returns:
pandas.Dataframe: The dictionary is returned as pandas.Dataframe object
"""
val = self.get(name)
if isinstance(val, dict):
df = pandas.DataFrame(val)
return df
def get(self, key, default=None):
"""
Internal wrapper function for __getitem__() - self[name]
Args:
key (str, slice): path to the data or key of the data object
default (object): default value to return if key doesn't exist
Returns:
dict, list, float, int: data or data object
"""
try:
return self.__getitem__(key)
except ValueError:
if default is not None:
return default
else:
raise
def put(self, key, value):
"""
Store data inside the HDF5 file
Args:
key (str): key to store the data
value (pandas.DataFrame, pandas.Series, dict, list, float, int): basically any kind of data is supported
"""
self.__setitem__(key=key, value=value)
def _list_all(self):
"""
List all groups and nodes of the HDF5 file - where groups are equivalent to directories and nodes to files.
Returns:
dict: {'groups': [list of groups], 'nodes': [list of nodes]}
"""
if self.file_exists:
with _open_hdf(self.file_name) as hdf:
groups, nodes = list_groups_and_nodes(hdf=hdf, h5_path=self.h5_path)
iopy_nodes = self._filter_io_objects(set(groups))
return {
"groups": sorted(list(set(groups) - iopy_nodes)),
"nodes": sorted(list((set(nodes) - set(groups)).union(iopy_nodes))),
}
else:
return {"groups": [], "nodes": []}
def _list_nodes(self):
return self.list_all()["nodes"]
def _list_groups(self):
return self.list_all()["groups"]
def listdirs(self):
"""
equivalent to os.listdirs (consider groups as equivalent to dirs)
Returns:
(list): list of groups in pytables for the path self.h5_path
"""
return self.list_groups()
def list_dirs(self):
"""
equivalent to os.listdirs (consider groups as equivalent to dirs)
Returns:
(list): list of groups in pytables for the path self.h5_path
"""
return self.list_groups()
def keys(self):
"""
List all groups and nodes of the HDF5 file - where groups are equivalent to directories and nodes to files.
Returns:
list: all groups and nodes
"""
list_all_dict = self.list_all()
return list_all_dict["nodes"] + list_all_dict["groups"]
def values(self):
"""
List all values for all groups and nodes of the HDF5 file
Returns:
list: list of all values
"""
return [self[key] for key in self.keys()]
def items(self):
"""
List all keys and values as items of all groups and nodes of the HDF5 file
Returns:
list: list of sets (key, value)
"""
return [(key, self[key]) for key in self.keys()]
def groups(self):
"""
Filter HDF5 file by groups
Returns:
FileHDFio: an HDF5 file which is filtered by groups
"""
new = self.copy()
new._filter = ["groups"]
return new
def nodes(self):
"""
Filter HDF5 file by nodes
Returns:
FileHDFio: an HDF5 file which is filtered by nodes
"""
new = self.copy()
new._filter = ["nodes"]
return new
def hd_copy(self, hdf_old, hdf_new, exclude_groups=None, exclude_nodes=None):
"""
args:
hdf_old (ProjectHDFio): old hdf
hdf_new (ProjectHDFio): new hdf
exclude_groups (list/None): list of groups to delete
exclude_nodes (list/None): list of nodes to delete
"""
if exclude_groups is None or len(exclude_groups) == 0:
exclude_groups_split = list()
group_list = hdf_old.list_groups()
else:
exclude_groups_split = [i.split("/", 1) for i in exclude_groups]
check_groups = [i[-1] for i in exclude_groups_split]
group_list = list(
(set(hdf_old.list_groups()) ^ set(check_groups))
& set(hdf_old.list_groups())
)
if exclude_nodes is None or len(exclude_nodes) == 0:
exclude_nodes_split = list()
node_list = hdf_old.list_nodes()
else:
exclude_nodes_split = [i.split("/", 1) for i in exclude_nodes]
check_nodes = [i[-1] for i in exclude_nodes_split]
node_list = list(
(set(hdf_old.list_nodes()) ^ set(check_nodes))
& set(hdf_old.list_nodes())
)
hdf_new.write_dict(data_dict={p: hdf_old[p] for p in node_list})
for p in group_list:
h_new = hdf_new.create_group(p)
ex_n = [e[-1] for e in exclude_nodes_split if p == e[0] or len(e) == 1]
ex_g = [e[-1] for e in exclude_groups_split if p == e[0] or len(e) == 1]
self.hd_copy(hdf_old[p], h_new, exclude_nodes=ex_n, exclude_groups=ex_g)
return hdf_new
@deprecate(job_name="ignored!", exclude_groups="ignored!", exclude_nodes="ignored!")
def rewrite_hdf5(
self, job_name=None, info=False, exclude_groups=None, exclude_nodes=None
):
"""
Rewrite the entire hdf file.
Args:
info (True/False): whether to give the information on how much space has been saved
"""
if job_name is not None:
state.logger.warning(
"Specifying job_name is deprecated and ignored! Future versions will change signature."
)
file_name = self.file_name
new_file = file_name + "_rewrite"
self_hdf = FileHDFio(file_name=file_name)
hdf_new = FileHDFio(file_name=new_file, h5_path="/")
old_logger_level = state.logger.level
state.logger.level = 50
hdf_new = self.hd_copy(self_hdf, hdf_new)
state.logger.level = old_logger_level
if info:
print(
"compression rate from old to new: {}".format(
self.file_size(self_hdf) / self.file_size(hdf_new)
)
)
print(
"data size vs file size: {}".format(
self.get_size(hdf_new) / self.file_size(hdf_new)
)
)
self.remove_file()
os.rename(hdf_new.file_name, file_name)
def __str__(self):
"""
Machine readable string representation
Returns:
str: list all nodes and groups as string
"""
return self.__repr__()
def __repr__(self):
"""
Human readable string representation
Returns:
str: list all nodes and groups as string
"""
return str(self.list_all())
def __del__(self):
del self._file_name
del self.history
del self._h5_path
def __exit__(self, exc_type, exc_val, exc_tb):
"""
Compatibility function for the with statement
"""
self.close()
try:
self._store.close()
except AttributeError:
pass
def _read(self, item):
"""
Internal read function to read data from the HDF5 file
Args:
item (str): path to the data or key of the data object
Returns:
dict, list, float, int: data or data object
"""
return _read_hdf(hdf_filehandle=self.file_name, h5_path=self._get_h5_path(item))
def write_dict_to_hdf(self, data_dict):
"""
Write a dictionary to HDF5
Args:
data_dict (dict): dictionary with objects which should be written to HDF5
"""
self.write_dict(data_dict=data_dict)
def read_dict_from_hdf(self, group_paths=[], recursive=False):
"""
Read data from HDF5 file into a dictionary - by default only the nodes are converted to dictionaries, additional
sub groups can be specified using the group_paths parameter.
Args:
group_paths (list): list of additional groups to be included in the dictionary, for example:
["input", "output", "output/generic"]
These groups are defined relative to the h5_path.
recursive (bool): Load all subgroups recursively
Returns:
dict: The loaded data. Can be of any type supported by ``write_hdf5``.
"""
return read_dict_from_hdf(
file_name=self.file_name,
h5_path=self.h5_path,
group_paths=group_paths,
recursive=recursive,
slash="ignore",
)
def create_project_from_hdf5(self):
"""
Internal function to create a pyiron project pointing to the directory where the HDF5 file is located.
Returns:
Project: pyiron project object
"""
from pyiron_base.project.generic import Project
return Project(path=self.file_path)
def _get_h5_path(self, name):
"""
Internal function to combine the current h5_path with the relative path
Args:
name (str): relative path
Returns:
str: combined path
"""
return get_h5_path(h5_path=self.h5_path, name=name)
def _get_h5io_type(self, name):
"""
Internal function to get h5io type
Args:
name (str): HDF5 key
Returns:
str: h5io type
"""
with _open_hdf(self.file_name) as store:
return str(store[self.h5_path][name].attrs.get("TITLE", ""))
def _filter_io_objects(self, groups):
"""
Internal function to extract h5io objects (which have the same type as normal groups)
Args:
groups (list, set): list of groups (as obtained e.g. from listdirs
Returns:
set: h5io objects
"""
h5io_types = (
"dict",
"list",
"tuple",
"pd_dataframe",
"pd_series",
"multiarray",
"json",
)
group_h5io = set(
[group for group in groups if self._get_h5io_type(group) in h5io_types]
)
return group_h5io
def _walk(self, level=0):
"""
Internal helper function for show_hdf() - iterating over the HDF5 datastructure and generating a human readable
graph.
Args:
level (int): iteration level
"""
l_dict = self.list_all()
indent = level * " "
for node in l_dict["nodes"]:
print(indent + "node", node)
for group in l_dict["groups"]:
print(indent + "group: ", group)
with self.open(group) as hdf_group:
hdf_group._walk(level=level + 1)
class ProjectHDFio(FileHDFio):
"""
The ProjectHDFio class connects the FileHDFio and the Project class, it is derived from the FileHDFio class but in
addition the a project object instance is located at self.project enabling direct access to the database and other
project related functionality, some of which are mapped to the ProjectHDFio class as well.
Args:
project (Project): pyiron Project the current HDF5 project is located in
file_name (str): name of the HDF5 file - in contrast to the FileHDFio object where file_name represents the
absolute path of the HDF5 file.
h5_path (str): absolute path inside the h5 path - starting from the root group
mode (str): mode : {'a', 'w', 'r', 'r+'}, default 'a'
See HDFStore docstring or tables.open_file for info about modes
Attributes:
.. attribute:: project
Project instance the ProjectHDFio object is located in
.. attribute:: root_path
the pyiron user directory, defined in the .pyiron configuration
.. attribute:: project_path
the relative path of the current project / folder starting from the root path
of the pyiron user directory
.. attribute:: path
the absolute path of the current project / folder plus the absolute path in the HDF5 file as one path
.. attribute:: file_name
absolute path to the HDF5 file
.. attribute:: h5_path
path inside the HDF5 file - also stored as absolute path
.. attribute:: history
previously opened groups / folders
.. attribute:: file_exists
boolean if the HDF5 was already written
.. attribute:: base_name
name of the HDF5 file but without any file extension
.. attribute:: file_path
directory where the HDF5 file is located
.. attribute:: is_root
boolean if the HDF5 object is located at the root level of the HDF5 file
.. attribute:: is_open
boolean if the HDF5 file is currently opened - if an active file handler exists
.. attribute:: is_empty
boolean if the HDF5 file is empty
.. attribute:: user
current unix/linux/windows user who is running pyiron
.. attribute:: sql_query
an SQL query to limit the jobs within the project to a subset which matches the SQL query.
.. attribute:: db
connection to the SQL database
.. attribute:: working_directory
working directory of the job is executed in - outside the HDF5 file
"""
def __init__(self, project, file_name, h5_path=None, mode=None):
self._file_name = _get_safe_filename(file_name)
if h5_path is None:
h5_path = "/"
self._project = project.copy()
super(ProjectHDFio, self).__init__(
file_name=os.path.join(self._project.path, self._file_name).replace(
"\\", "/"
),
h5_path=h5_path,
mode=mode,
)
@property
def base_name(self):
"""
The absolute path to of the current pyiron project - absolute path on the file system, not including the HDF5
path.
Returns:
str: current project path
"""
return self._project.path
@property
def db(self):
"""
Get connection to the SQL database
Returns:
DatabaseAccess: database conncetion
"""
return self._project.db
@property
def path(self):
"""
Absolute path of the HDF5 group starting from the system root - combination of the absolute system path plus the
absolute path inside the HDF5 file starting from the root group.
Returns:
str: absolute path
"""
return os.path.join(self._project.path, self.h5_path[1:]).replace("\\", "/")
@property
def project(self):
"""
Get the project instance the ProjectHDFio object is located in
Returns:
Project: pyiron project
"""
return self._project
@property
def project_path(self):
"""
the relative path of the current project / folder starting from the root path
of the pyiron user directory
Returns:
str: relative path of the current project / folder
"""
return self._project.project_path
@property
def root_path(self):
"""
the pyiron user directory, defined in the .pyiron configuration
Returns:
str: pyiron user directory of the current project
"""
return self._project.root_path
@property
def sql_query(self):