forked from spglib/spglib
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ChangeLog
1440 lines (943 loc) · 45 KB
/
ChangeLog
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
2019-02-06 Atsushi Togo <[email protected]>
* Version 1.12.2
Minor release for memory leak fix (PR#71).
2019-01-30 Atsushi Togo <[email protected]>
* Version 1.12.1
* src/spglib.h:
Bug fix of PR #70. #include <stddef.h> is inserted. In principle,
this line is necessary to compile.
2019-01-29 Atsushi Togo <[email protected]>
* Version 1.12.0
After this minor version, addition of new features will be stopped
until version 2.0 will be released.
* src/sitesym_database.c (ssmdb_get_site_symmetry_symbol):
Experimental support of site symmetry symbols.
* src/pointgroup.c:
Comparisons of basis vector lengths are more carefully done for
letting them work for different computer architectures to treat
very small numerical different that is close to the machine
precision.
* src/mathfunc.c (mat_Dmod1):
Similarly, MOD-Z operation (bring values x inside the rage of 0<=
x < 1) is now done considering the difference of machine
precision.
* src/kpoint.c (kpt_get_irreducible_reciprocal_mesh):
To enable dense k-point mesh sampling, integer types of variables
that represent grid point indices were changed from int to size_t.
But to support current functions with the same API, adapters were
written for these functions (having 'dense' in their function
names until v2.0 will be released) and currently they are working
in the same way as before. Finally the API change will be realized
at the v2.0
2018-12-07 Atsushi Togo <[email protected]>
* Version 1.11.2
* src/refinement.c (find_similar_bravais_lattice):
Bug fix. Again, origin shift was still not correctly computed when
the standardized basis vectors are rotated in this method to find
most similar choice of basis vectors to the idealized ones. This
bug affected to 'origin_shift' and 'std_positions' in spglib
dataset and the functions 'spg_standardize_cell',
'spg_find_primitive', and 'spg_refine_cell'. Although there
existed this bug, tests unfortunately passed. So a new test to
check Wyckoff positions consistency was added to find possible
bugs related to origin shift because this bug was found by getting
wrong Wyckoff positions reported by a user.
2018-11-12 Atsushi Togo <[email protected]>
* Version 1.11.1
* src/refinement.c (find_similar_bravais_lattice):
Bug fix. Origin shift was not correctly computed when the
standardized basis vectors are rotated in this method to find most
similar choice of basis vectors to the idealized ones. This bug
affected to 'origin_shift' and 'std_positions' in spglib dataset
and the functions 'spg_standardize_cell', 'spg_find_primitive',
and 'spg_refine_cell'.
2018-11-09 Atsushi Togo <[email protected]>
* Version 1.11.0
* src/refinement.c (find_similar_bravais_lattice):
This looks for a
similar choice of starndardized basis vectors to that of idealized
starndardized basis vectors among those generated by proper
symmetry operations.
* src/refinement.c (measure_rigid_rotation):
This computes a rigid rotation matrix introduced
in the idealization step. The rotation is made to agree a-axes
and a-b planes between systems before and after idealization.
2018-08-01 Atsushi Togo <[email protected]>
* Version 1.10.4
Collection of minor fixes.
A text on spglib algorithm is prepared for this version. This
text will be available soon at arXiv. To be consistent between
text and code, the code refactoring was made.
2018-01-13 Atsushi Togo <[email protected]>
* Version 1.10.3
* src/overlap.c:
New algorithm to search translation part of space group operation
is implmented by Michael Lamparski to improving the search speed
especially for large unit cell systems. For this, new files
'overlap.*' are added and so the version number is incremented.
2017-12-13 Atsushi Togo <[email protected]>
* Version 1.10.2
* src/Makefile.am:
Automake didn't work due to the lack of determination.{c,h}
lines. This was fixed.
2017-10-27 Atsushi Togo <[email protected]>
* Version 1.10.1
* src/spglib.c (get_dataset):
A large refactoring was made to replace the outer-most iteration
loop of symmetry search algorithm to a new file 'determination.c'.
* src/spglib.c (spg_get_schoenflies,spg_get_international):
These routines had accessed directly to a function close to
spacegroup.c. Now these are replaced by passing through
'get_dataset'.
2017-10-22 Atsushi Togo <[email protected]>
* Version 1.10.0
'mapping_to_primitive' and 'std_mapping_to_primitive' in the
spglib dataset are put available on the spglib
document. 'spg_get_hall_number_from_symmetry' function
('get_hall_number_from_symmetry' method for python) is also
available on the spglib document.
2017-10-02 Atsushi Togo <[email protected]>
* Version 1.9.10
Many minor fixes and updates after many months from v1.9.9.
* src/symmetry.c (search_pure_translations):
Pure translation search is made faster using a property of
group, which is significantly effective for supercells with large
multiplicity.
2016-12-14 Atsushi Togo <[email protected]>
* Version 1.9.9
* src/kpoint.c:
'get_ir_reciprocal_mesh' was supporsed to work with mesh numbers
that are consistent with symmetry. Now this function somehow
supports when mesh numbers are symmetrically broken.
2016-12-09 Atsushi Togo <[email protected]>
* src/symmetry.c:
static variable of 'angle_tolerance', 'sym_set_angle_tolerance',
'sym_get_angle_tolerance' were removed toward letting it
thread-safe. Now 'angle_tolerance' is passed as an argument to
many functions in spglib.
2016-11-02 Atsushi Togo <[email protected]>
* Version 1.9.8
* src/cell.c (cel_is_overlap_with_same_type):
'cel_is_overlap_with_same_type' is implemented to check
overlapping of atoms along with checking atom type
agreement. Using this, atoms that have about the same position but
with different types can be separately handled.
2016-10-19 Atsushi Togo <[email protected]>
* Version 1.9.6
Collection of minor fixes.
* src/cell.c (cel_set_cell):
Reduced atomic positions of the input cell are inisitalized by
their moving in the range between -0.5 and 0.5. The cell shape is
kept unchanged but this may need to be considered to handle some
extreme case in the future.
2016-09-05 Atsushi Togo <[email protected]>
* Version 1.9.5
* src/spglib.h (SpglibError):
'spg_get_error_code' and 'spg_get_error_message' were made to
handle a few types of internal errors.
* src/spglib.c, src/cell.c (cel_any_overlap_with_same_type):
To check overlap of atoms, 'cel_any_overlap' and
'cel_any_overlap_with_same_type' were implemented. The former
doesn't check atomic type but the later does. Currently this check
is done before symmetry search and the later is used for
it. Therefore overlap of different types of atoms is allowed.
2016-05-06 Atsushi Togo <[email protected]>
* Version 1.9.4
* src/spglib.h (SpglibDataset):
The member of SpglibDataset, 'setting', is renamed to 'choice',
and 'pointgroup_number' is removed.
* src/cell.c (get_overlap_table):
Fix unexpected behavior. This might happen in a rare case with
large tolerance.
2016-04-28 Atsushi Togo <[email protected]>
* src/spglib.c (spg_get_spacegroup_type):
Arithmetic crystal class number and symbol are returned. Python
interface 'get_spacegroup_type' was created.
* src/spglib.c (get_dataset):
* src/site_symmetry.c (ssm_get_exact_positions, set_exact_location):
Two iterative loops are implemented. The outer one is in
get_dataset, and the inner one is in ssm_get_exact_positions.
Each site-symmetry is checked in set_exact_location. In the inner loop,
symmetry tolerance is controled until all site-symmetries become
reasonable. If the inner loop finally fails, the symmetry
tolerance in the outer loop is reduced and the symmetry search
starts from the begginig.
2016-04-17 Atsushi Togo <[email protected]>
* Version 1.9.2
* src/symmetry.c (get_lattice_symmetry):
The treatment after the conditon, order of point symmetry
becomes larger than 48, is found was bad and it induced
segfault. It is fixed and a better treatment for angle_tolerance
is implemented as an iterative method.
* src/spglib.c (spg_delaunay_reduce):
spg_delaunay_reduce is made as an interface to the Delaunay
reduction.
* src/spacegroup.c (search_spacegroup):
A check if excess symmetry operations are found for a primitive
cell is inserted by watching its point group. This made the
algorithm of database matching safer.
2016-04-11 Atsushi Togo <[email protected]>
* src/delaunay.c:
lattice.{c,h} are renamed to delaunay.{c.h},
respectively. Function names in them, *get_smallest_lattice, are
changed to *delaunay_reduce.
2016-04-06 Atsushi Togo <[email protected]>
* Version 1.9.1
This is a minor update. Only a small notable change is in the
python wrapper, where for crystal structure, now a tuple format
can be sued.
2015-12-20 Atsushi Togo <[email protected]>
* Version 1.9.0
Pyspglib was considered as a special extention of spglib, but now
it starts to be a part of spglib as a usual extention. Therefore
the package name is changed to spglib, and this python wrapper
will be maintained to work at every update of c-spglib.
Scripts for pypi and conda package environments are prepared in
python wrapper by a lot of help from Paweł T. Jochym.
In previous versions, tests were in the test directory and space
group numbers were checked for many POSCARs by using combination
of ruby-spglib and shell script. Now a similar test is written in
a form of python unit test in the 'python' directory. Previous
'test' directory is renamed to 'ruby' and the POSCARs were moved
under the 'python/test' directory.
2015-12-13 Atsushi Togo <[email protected]>
* Version 1.8.3
* src/version.h:
Made functions to return spglib version number,
'spg_get_major_version()', 'spg_get_minor_version()',
'spg_get_micro_version()' in 'spglib.c'. This version number is
stored in 'version.h'.
2015-08-19 Atsushi Togo <[email protected]>
* Version 1.8.2.2
* src/spglib.c (is_rhombohedral):
Regardless of rhombohedral or hexagonal setting, a primitive
rhombohedral is found when using finding primitive cell functions.
2015-08-03 Atsushi Togo <[email protected]>
* Version 1.8.2.1
* src/spacegroup.c (spa_transform_to_primitive):
Bug fix of transformation matrix with centering. The side to be
multiplied was wrong.
* Version 1.8.2
* k-points and tetrahedron method
k-point grid definitions are separated to 'kgrid.c'.
'kgrid.{c,h}' and 'tetrahedron_method.{c,h}' are moved to a new
git repository of kspclib (https://github.com/atztogo/kspclib).
Tetrahedron method related spglib-APIs are removed.
* Choice of triclinic basis vector directions is implemented and
the definition is written in the document.
2015-07-09 Atsushi Togo <[email protected]>
* Version 1.8.1
In SpglibDataset, definitions of transformation matrix and origin
shift are changed, member names of '*brv_*' are changed to
'*std_*', and the members, pointgroup_number and
pointgroup_symbol, are added.
New function 'spg_standardize_cell' is implemented. This can also
give 'standardized primitive cell' with 'to_primitive =
1'.
Due to change of internal implementation, the order of symmetry
operations obtained can be different from the recently released
versions, but the set is the same.
2015-06-25 Atsushi Togo <[email protected]>
* Version 1.8
Relatively large code refactoring was made. Especially
'primitive.c' was changed a lot.
Write license statement (New BSD) in the source codes.
* src/spglib.c:
A new function 'spg_find_standardized_primitive' is
implemented. This function allows to create a primitive cell from
the obtained standardized conventional unit cell with
transformation using pre-determined choice of centring basis
vectors.
2015-06-15 Atsushi Togo <[email protected]>
* Version 1.7.4
Wyckoff position database for rhombohedral cell was
wrong. Probably this bug was hidden by the previous default choice
of basis vectors but from version 1.7.3 it became revealed.
2015-06-14 Atsushi Togo <[email protected]>
* src/sitesym_database.c
Bug in database of hP of rhombohedral was fixed. This bug was
affected after version 1.7.3. 'make_Wyckoff_db.py' is fixed and
then the database in 'sitesym_database.c' is recreated by
make_Wyckoff_db.py.
2015-04-06 Atsushi Togo <[email protected]>
* Version 1.7.3
2015-04-05 Atsushi Togo <[email protected]>
* src/triplet_kpoint.{c,h}:
Code to handle triplets of kpoints was separated from
kpoint.c. Detail of functions was revisited and modified, e.g.,
taking modulo of grid address is now included in
get_grid_point_double_mesh in most cases. This is probably safe
change, but may induce some tricky bug, which has to be
remembered. kpt_get_grid_point should be moved to spglib.c with
calling kpt_get_grid_point_double_mesh someday soon. It is noted
here for not forgetting that get_third_q_of_triplets_at_q in
triplet_kpoint.c might be written better around the point using
mat_modulo_i3. In triplet_kpoint.c, mat_modulo_i3 is used only
here, so eliminating this can move mat_modulo_i3 to kpoint.c as a
local function as before.
2015-04-03 Atsushi Togo <[email protected]>
* src/spacegroup.c (match_hall_symbol_db_ortho):
Change the convention of lengths of basis vector preference to:
a < b < c for hall number 290 and a < c < b for 291 or
a < b < c for hall number 291 and a < c < b for 290,
to preserve a < b convention found in Parthe et al, Journal of
Alloys and Compounds, 197 (1993) 291-301.
2015-03-31 Atsushi Togo <[email protected]>
* src/spacegroup.c (match_hall_symbol_db_ortho):
Fix choice of lengths of basis vectors in orthorhombic Pbca. There
are two choices for Pbca. The basis vectors can be chosen so as to
a < b < c or b < a < c, which correspond to these two choices,
respectively. The correspondance can be flipped, i.e., in a case,
a < b < c for hall number 290 and b < a < c for 291, but in some
other case, a < b < c for hall number 291 and b < a < c for 290.
2015-03-20 Atsushi Togo <[email protected]>
* src/hall_symbol.c (find_hall_symbol):
A critical bug was found in version 1.7.2, where origin shift was
given for the primitive cell but not for the Bravais lattice. This
was fixed. The version 1.7.2 was removed from the download site
and temporary 1.7.3-pre is put on the download site.
2015-03-08 Atsushi Togo <[email protected]>
* Version 1.7.2
* src/niggli.*:
Routine of Niggli cell reduction is imported from
https://github.com/atztogo/niggli.
2015-02-24 Atsushi Togo <[email protected]>
* src/spacegroup.c (get_symmetry_settings):
Monoclinic lattice vectors are chosen as |a| < |c| with b as the
unique axis. This is achieved by get_Delaunay_reduction_2D.
* src/lattice.c (get_Delaunay_reduction_2D):
2D Delaunay reduction is implemented based on that of 3D.
* src/lattice.c (lat_smallest_lattice_vector_2D):
Create an interface to access 2D Delaunay reduction.
2015-02-11 Atsushi Togo <[email protected]>
* Version 1.7.1
There are minor changes to make functions robust and code clean.
* src/spglib.c
spg(at)_get_symmetry_with_collinear_spin is changed to return
equivalent atoms.
spg_get_international and spg_get_schoenflies call iterative
routine to determine space group type.
* src/primitive.h:
Primitive structure is defined. This is expected to make routines
related to primitive cell comprehensive. Memory for a pointer of
the Primitive variable may be allocated using prm_alloc_primitive
and freed by prm_free_primitive.
* src/spacegroup.c:
spa_get_spacegroup was replaced by iterative routine and now
returns a Primitive structure.
2014-11-19 Atsushi Togo <[email protected]>
* Version 1.7
Space group operations for non-standard setting can be determined
using spg_get_dataset_with_hall_number. To achieve this, many
lines are added and changed. Due to the big change of the code,
there may be bugs, and maybe it's safer to wait for a while to use
this version for mission critical systems.
Two important changes are:
* src/spglib.c (spg_get_dataset_with_hall_number)
In space group type determination, Hall symbol serial number can
be specified.
* src/spglib.h (SpglibDataset)
Crystallographic setting information is stored in
'setting'. Symmetrized and standarized crsytal structure is stored
in 'n_brv_atoms', 'brv_lattice', 'brv_types', and 'brv_positions',
which are the same information obtained using 'spg_refine_cell',
except that non-standard setting can be specified in
'spg_get_dataset_with_hall_number'.
2014-10-16 Atsushi Togo <[email protected]>
* Version 1.6.4
* src/refinement.c (refine_cell):
An iterative approach is implemented for refine_cell.
2014-09-12 Atsushi Togo <[email protected]>
* Version 1.6.3
* src/primitive.c:
Atom type is checked in finding a primitive cell. This enables to
handle different types of atoms that locate at the same position.
for the case of an automatically generated crystal structure data
from a fractional occupancy, etc.
* Version 1.6.2
Minor bug fixes.
* src/spglib.c (spg_get_symmetry_from_database):
An API to access space group operations database is implemented.
2014-08-04 Atsushi Togo <[email protected]>
* Version 1.6.1:
* src/{refinement.*,primitive.*,spglib.c}
Fix symmetry finding bug for supercell that has lower point group
than the primitive cell.
* src/spglib.*
New functions of spg_get_spacegroup_type and
spg_get_grid_point. The former offers difrect access to space
group type database using hall symbol number. The later perfoms
converting grid address to grid point index on the reciprocal
mesh.
* pyspglib update.
2014-02-17 Atsushi Togo <[email protected]>
* Version 1.6.0:
* src/spglib.c (spg_get_spacegroup_type):
A function is made for accessing space-group-type database.
2014-01-07 Atsushi Togo <[email protected]>
* src/tetrahedron_method.*:
Tetrahedron method is implemented. Example is found in the example
directory.
2013-09-10 Atsushi Togo <[email protected]>
* Version 1.5.2:
* kpoint.c (kpt_relocate_BZ_grid_address):
New function kpt_relocate_BZ_grid_address is implemented in
kpoint.c. k-points on the mesh of the reciprocal primitive lattice
are relocated to fit in the Brillouin zone. To treat
translationally equivalent k-points, the memory space of
bz_grid_address[prod(mesh + 1)][3] is required. bz_map[prod(mesh
* 2 - 1)] is used to map k-points on the expanded k-space where
the shape is same as the reciprocal primitive lattice but the size
along each lattice vector is multiplied by mesh * 2 - 1. BZ zone
is included in this expanded k-space, so the bz_map can hold the
location of relocated k-points. Irreducible k-points are stored
from 0 to index of prod(mesh) - 1 in bz_grid_address in the same
order of grid_address, and the translationally equivalent k-points
are stored after that. Fast indexing of the location of relocated
k-points can be executed using bz_map with similar way to usual
grid_address scheme.
* kpoint.c (kpt_get_BZ_triplets_at_q):
New function kpt_get_BZ_triplets_at_q is implemented using
relocate_BZ_grid_address. This generates a set of triplets that
cleanly fit in BZ.
2013-09-10 Atsushi Togo <[email protected]>
* kpoint.c:
Remove get_triplets_reciprocal_mesh and
extract_triplets_reciprocal_mesh_at_q. Add
get_grid_triplets_at_q. This function treats q-points on boundary
correctly.
* spglib.c (spg_get_symmetry):
Previously symmetry operations were obtained by sym_get_operation.
Now it is changed so that they are extracted from spglib_dataset.
2013-04-16 Atsushi Togo <[email protected]>
* symmetry.c (get_operation):
In some case, get_space_group_operation returns more than 48
symmetry operations for a primitive cell. To avoid this, an
iterative approach, which is just reducing tolerance gradually to
achieve number of symmetry operations <= 48, is implemented.
* symmetry.c:
In the symmetry operation search, rot[] and trans[] were used. Now
they are replaced by Symmetry *symmetry in symmetry.c.
2013-04-14 Atsushi Togo <[email protected]>
* Fortran interface given by Dimitar Pashov:
The new fortran interface is given by Dimitar Pashov, which is
much safer than the previous one. The new fortran interface, its
example, and makefile are found in the example directory as
spglib_f08.f90, example_f08.f90 and Makefile_f08.
* spglib.h, spglib.c, kpoint.h, kpoint,c:
Symprec parameter was removed from spg_get_stabilized_reciprocal_mesh.
2013-01-28 Atsushi Togo <[email protected]>
* spglib.c, spglib_f.c:
Pointerd out by Michael Rutter that the length of charasters of
space-group type symbol (International table) was wrong in
spglib_f.c. That of Schoenflies symbol was also found wrong. The
former and latter were fixed as from 21 to 11 and from 10 to 7,
respectively.
2012-10-10 Atsushi Togo <[email protected]>
* Version 1.3
Rethink algorithms to search symmetry operations for speed-up and
stability. Many small modifications are applied. Some parts of the
codes that look unnecessary were removed but this change might
make the code unstable.
* symmetry.c (get_translation):
Improve OpenMP efficiency. Number of atoms to enable OpenMP
concurrency is set as 1000 (NUM_ATOMS_CRITERION_FOR_OPENMP).
2012-10-09 Atsushi Togo <[email protected]>
* symmetry.c (sym_get_pure_translation, get_translation):
In 'sym_get_pure_translation', iterative approach to search
pure_translation was employed. But 'sym_get_pure_translation' was
only used from two functions in primitive.c. In these two
functions, tolerance is iteratively reduced when the search
fails. Because there is no need to do iteration twice, the
iteration part in 'sym_get_pure_translation' was removed.
* primitive.{c,h}:
'get_primitive_with_pure_translation' was replaced by
'get_primitive_and_pure_translation'. The former was only used
from 'get_operation' in symmetry.c and the interative approach was
implemented in 'get_operation'. The iterative part was moved into
'get_primitive_and_pure_translation'.
The Primitive structure was defined in primitive.h. This just
packs Cell * primitive_cell and VecDBL * pure_trans. The Primitive
structure is only used in 'get_operation' in symmetry.c and
'(prm_)get_primitive_and_pure_translation' in primitive.c.
2012-10-08 Atsushi Togo <[email protected]>
* spacegroup.c (get_hall_number_local_iteration):
REDUCE_RATE to reduce number of symmetry operations is changed
from 0.2 to 0.95. Symmetry operations depend on tolerance
delicately.
* Remove many lines of debug_print:
Now to follow the flow is more important than to see the
detailed behavior of implementation.
2012-10-03 Atsushi Togo <[email protected]>
* Version 1.2.5
* primitive.c:
Separate out get_primitive_with_mapping_table from
prm_get_primitive_with_mapping_table. Rename
prm_get_primitive_with_all to
get_primitive_with_pure_translation.
In get_primitive_with_pure_translation, it was
if (pure_trans->size > 0)
but not it was fixed to
if (pure_trans->size > 1)
not to search primitive cell when pure_trans->size == 1.
2012-10-01 Atsushi Togo <[email protected]>
* Version 1.2.4
* kpoint.c (get_ir_reciprocal_mesh):
In spglib, k-point grid is indexed as, for example,
0 1 2 3 4 5 6 -5 -4 -3 -2 -1
for 12 points mesh. Now the different index style such as
0 1 2 3 4 5 -6 -5 -4 -3 -2 -1
is activated by CFLAGS='-DGRID_BOUNDARY_AS_NEGATIVE' before
running configure script at the compilation time.
* symmetry.c (get_operation):
In the iterative reduction of tolerance of primitive cell search,
pure_trans was searched by a fixed tolerance rather than the
reduced one. This is fixed to use the reduced tolerance at each
iteration.
2012-08-13 Atsushi Togo <[email protected]>
* Version 1.2.3
* symmetry.c (is_identity_metric):
To check identity of metric tensors, too small angle induces
numerical size error, so if the dtheta is calculated smaller than
1e-6, automatically set the angles be identical.
* spacegroup.c (get_hall_number_local_iteration):
Implementation of iterative reduction of symmetry tolerance to
search symmetry operation to get Hall symbol is separated as the
function get_hall_number_local_iteration and the tolerance
reduction rate of REDUCE_RATE is changed from 0.2 from 0.95.
2012-07-17 Atsushi Togo <[email protected]>
* Version 1.2.2
* lattice.c (get_centering):
Tranformation A-centring monoclinic lattice to C-centring
monoclinic lattice preserving b-axis is not well handled and now
the matrix 'monocli_a2c' is used for it.
2012-07-14 Atsushi Togo <[email protected]>
* Version 1.2.1
* pointgroup.c (laue_one_axis):
Transformation matrix is expected to be always positive, but it
was not correctly implemented.
2012-05-28 Atsushi Togo <[email protected]>
* refinement.c (reduce_symmetry_in_frame):
Fix memory leak by freeing lattice_trans.
2012-04-26 Atsushi Togo <[email protected]>
The numbering of the hall symbol is included in the 'struct
SpglibDataset'. This starts from 1 to 530.
2012-04-25 Atsushi Togo <[email protected]>
* spglib.c, spglib.h:
To find symmetry with angle tolerance, for spg_* functions,
respective spgat_* functions are prepared. In the spgat_*
functions, the last variable is the angle tolerance (in degrees).
* symmetry.c:
The way to find lattice symmetry is changed. Now a new static
variable, 'angle_tolerance', can be used for the tolerance of
angle. Therefore the non-diagonal elements of metric tensor are
directry compared with this angle. 'angle_tolerance' can be
accessed via 'sym_set_angle_tolerance' inside spglib. However the
default behavior does not use the angle tolerance. In the default
behavior, averaged_lattice_vector_length * sin(delta_theta) is
compared with usual 'symprec'. The default behavior is invoked
when 'angle_tolerance' has negative value. These are implemented
in 'is_identity_metric' in symmetry.c.
2012-04-12 Atsushi Togo <[email protected]>
* lattice.c (get_centering):
Correction matrix for I-centring monoclinic was implemented.
2012-03-03 Atsushi Togo <[email protected]>
* primitive.c:
The functions of 'prm_get_current_tolerance',
'set_current_tolerance' and the value of static double
current_tolerance were introduced in 'primitive.c'. These are used
to return reduced tolerance value that is iteratively determined
in 'prm_get_primitive_with_mapping_table'.
The interative technique is used in several parts to obtain
possible primitive cell parameters by reducing symmetry
tolerance. Only in 'get_overlap_table' in 'primitive.c', the
interative techinique is used with reducing and increasing the
tolerance, i.e., in the other parts, only reducing.
2012-03-02 Atsushi Togo <[email protected]>
* Version 1.1.2
* refinement.c, primitive.c, symmetry.c:
Modifications at many small parts of the code for improve
robustness with large symmetry tolerance values.
2012-02-28 Atsushi Togo <[email protected]>
* primitive.c (get_overlap_table):
Simplify the algorithm for search overlapping table. The new
algorithm is as follows: Symmetry tolerance monotonically
decreases by REDUCE_RATE = 0.95 when the search is failed. When
number of overlapping atoms is found smaller than the ratio of
number of atoms between original and primitive cells, the symmetry
tolerance is doubled. For this algorithm, 'check_overlap_table'
function was implemented.
2012-02-09 Atsushi Togo <[email protected]>
* spin.{c,h}:
'spin.{c,h}' are added. Colliear spins on atoms are employed as
additional constraints to search symmetry. The API is
'spg_get_symmetry_with_collinear_spin'. (reduce_symmetry_in_frame):
Bug fix to get refined symmetry operations. When the original cell
is not primitive and the lattice symmetry is lower than that
primitive lattice, the symmetry of primitive has to be
reduced to those belongs to the lower symmetry. But this was not
correctly done.
2011-10-05 Atsushi Togo <[email protected]>
* spglib.c (spg_get_pointgroup):
'spg_get_pointgroup' is implemented. This is used to obtain point
group symbols.
2011-09-30 Atsushi Togo <[email protected]>
* site_symmetry.c (get_Wyckoff_notation):
Bug fix of assignment of Wyckoff letters
'sitesym_database.c' is compressed. Now only the first element of
'coordinates' is used to aasign a Wyckoff letter.
2011-08-09 Atsushi Togo <[email protected]>
* symmetry.c (sym_get_pure_translation):
Iterative reduction of symmetry tolerance to find pure
translations is implemented as well as different parts.
2011-07-29 Atsushi Togo <[email protected]>
* kpoint.c, kpoint.h:
symmetry_kpoint.c and symmetry_kpoint.h are renamed as kpoint.c
and kpoint.h, respectively.
2011-07-06 Atsushi Togo <[email protected]>
* site_symmetry.c:
A set of equivalent atoms is returned. It is accessible
through spg_get_dataset. (spglib-1.0.7)
2011-07-05 Atsushi Togo <[email protected]>
* site_symmetry.c, sitesym_database.c, spglib-1.0.6:
Newly added two source files (and also *.h). The part related to
site symmetry in refinement.c is moved to site_symmetry.c. Wyckoff
position determination is implemented in site_symmetry.c and the
database is in sitesym_database.c. The pointers in the structure
SpglibDataset are made free by calling spg_free_dataset. It was
difficult to implement Wyckoff position determination without
Yusuke Seko's formated crystallographic database. I appreciate his
kind offer. (spglib-1.0.6)
* primitive.c (prm_get_primitive):
The function is modified to return mapping table from input cell
to primitive cell as the first argument.
2011-06-18 Atsushi Togo <[email protected]>
* symmetry.c:
OpenMP is included to search translations with each rotation.
Compilation with gcc is achived by '-lgomp' and '-fopenmp'.
* refinement.c:
Performance tuning was made for 'get_refined_symmetry_operations'.
2011-06-18 Atsushi Togo <[email protected]>
* refinement.c:
In 'ref_get_refined_symmetry_operations', it was forgotten to
consider origin shift to obtain symmetry operations from
database. This is fixed by modifying translations in
'set_translation_with_origin_shift'.
2011-06-15 Atsushi Togo <[email protected]>
* spglib-1.0.4:
The new version of spglib is released. To the next version,
implementation of Wyckoff positions is planed.
2011-06-15 Atsushi Togo <[email protected]>
* pyspglib:
pyspglib didn't work correctly. It was fixed.
2011-06-15 Atsushi Togo <[email protected]>
* spglib.c:
A new function 'spg_get_dataset is implmented. This returns a
'SpglibDataset' type structure that is defined in 'spglib.h'. This
includes refined informations of crystal symmetry. For example,
the symmetry operations are derived based on the operations in
database and so they are different from those obtained from
'spg_get_symmetry'. Wyckoff positions are to be impmemented.
2011-06-13 Atsushi Togo <[email protected]>
* spg_database.c:
A static array of 'symmetry_operations' is compressed because it
was very large. It contains symmetry operations of 530 hall
symbols. Each space group operation requires a rotation 3x3 matrix
and a translation vector, therefore 12 elements were implemented
for that. Now it is compressed using ternary numerical system for
rotation and duodecimal system for translation. This is achieved because
each element of rotation matrix can have only one of {-1,0,1}, and
the translation can have one of {0,2,3,4,6,8,9,10} divided by
12. Therefore 3^9 * 12^3 = 34012224 different values can map space
group operations. In principle, octal numerical system can be used
for translation, but duodecimal system is used for the convenient.
The fact that determinant of rotation matrix is -1 or 1 could be
used as a constraint to reduce the number of mapping space. These
may be implemented if more compression is required.