You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: HISTORY.md
+25-12Lines changed: 25 additions & 12 deletions
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,18 @@
1
1
# Revision History for `enumlib`
2
2
3
+
## 2.0.4 (GLWH)
4
+
- At the request of `mkhorton` (Issue \#89), added the functionality for the enumeration code to print out a version number each time `gen_multilatice_derivatives` is called. The filename is `VERSON.enum`. The version number written to the file is generated by `git describe --tags --abbrev=4 --always`. The `Makefile` embeds the output of `git describe` into the source code at compile time.
5
+
- This change requires the developers to 1) tag their releases, and 2) push the tags (`git push` does not do this by default). Suggest that each time one of us commits that we do the following: (see `https://git-scm.com/book/en/v2/Git-Basics-Tagging`)
6
+
7
+
`git tag vX.Y.Z` where the version number matches the rev. number in `HISTORY.md`
8
+
`git push origin vX.Y.Z`
9
+
10
+
11
+
## 2.0.3 (GLWH)
12
+
- Made changes in `aux_src/compare_two_enum_files.f90` so that two output files could be compared, including considering the automorphism induced by changes to the Smith Normal Form routine (that made in more robust against overflow when used in the "on the fly" generalized regular kpoint grid generation code). This allowed us to update all the unit tests after the changes to the SNF routine in `symlib` which impacted the output in many cases. When the left transform matrix associated with the transformation of an HNF matrix to SNF matrix changes (it is not unique), that change affects the enumeration output. The enumeration is *equivalent*, but not always identical to, the previous (pre-SNF change) enumeration. Figuring out how to compute and apply the automorphism, how to take into account permutations of labelings in the old and new files, and how to take into account complementary cases (when labels are exchanged) took months.
13
+
- Many other small changes (improved comments throughout, removal of debugging print statements, etc.).
14
+
- Now that these changes have been made, the rest of the updates needed to complete the "inactive site" work of version 2.0.0 can be made.
15
+
3
16
## 2.0.2 (JJ)
4
17
- Fixed functions in `enumlib/wrap` to reflect changes to `get_dvector_permutations`.
5
18
@@ -99,11 +112,11 @@ changes, so committing/pushing again.
99
112
- Fixed the maxLabLength in makePerovStr.f90 to be 100. Also fixed the
100
113
call to reduce_to_shortest_basis as described in issue #38.
101
114
102
-
## Revision 1.0.6
115
+
## Revision 1.0.6 (WSM)
103
116
104
117
- The maximum allowed binomial, max_binomial in
105
118
src/derivative_structure_generator.f90, had been set to
106
-
2.63E14. This number was to large and allowed the code to enter the
119
+
2.63E14. This number was too large and allowed the code to enter the
107
120
generate_permutation_labelings algorithm when it should have been
108
121
entering the recursively_stabilized_enum algorithm. The max_binomial
109
122
is now 1E10 to force the correct behavior.
@@ -120,25 +133,25 @@ changes, so committing/pushing again.
120
133
- Fixed bug in makeStr.py, the lattice vectors in struct_enum.out were
121
134
being treated as column and not row vectors.
122
135
123
-
## Revision 1.0.4
136
+
## Revision 1.0.4 (WSM)
124
137
125
-
- Removed depricated code from line 443 of tree_class.f90.
138
+
- Removed deprecated code from line 443 of tree_class.f90.
126
139
127
140
- Added more extensive comments to explain the loop structure of line
128
-
171 of labelin_related.f90.
141
+
171 of `labeling_related.f90`.
129
142
130
143
- Removed the `if (d==0) then` statement from addArrowsToEnumeration
131
144
of tree_class.f90 since the code now handles all depths
132
145
appropriately.
133
146
134
-
- Fixed merge conflits with master branch.
147
+
- Fixed merge conflicts with master branch.
135
148
136
149
## Revision 1.0.3
137
150
138
151
- Fixed a bug number of bugs in the addArrows subroutine of
139
-
tree_class.f90.
152
+
`tree_class.f90`.
140
153
141
-
- Added nArrows as an input to tree_class and as an output to
154
+
- Added `nArrows` as an input to tree_class and as an output to
142
155
arrow_concs of arrow_related.
143
156
144
157
- Fixed the generateArrowingFromIndex subroutine so that it returns
@@ -148,7 +161,7 @@ changes, so committing/pushing again.
148
161
- Fixed the color_mapping of the arrowed labels for the site
149
162
restricted cases in labeling_related.
150
163
151
-
- Added makeStr.py to the repo to replace makeStr.x.
164
+
- Added `makeStr.py` to the repo to replace makeStr.x.
152
165
153
166
## Revision 1.0.2
154
167
@@ -160,7 +173,7 @@ changes, so committing/pushing again.
160
173
.and. .not. use_arrows)) stop "(d,g)-->(d',g') mapping failed in
161
174
get_rotation_perm_lists"`.
162
175
163
-
## Revision 1.0.1
176
+
## Revision 1.0.1 (WSM)
164
177
- Updated the polya submodule.
165
178
- Fixed a bug in tree_class.f90 that was causing a segmentation fault
166
179
in the arrow enumeration. The error was caused because using
@@ -170,8 +183,8 @@ changes, so committing/pushing again.
170
183
this permutation should.
171
184
172
185
- Fixed the final color mapping in the arrow enumeration. The location
173
-
of the old coloring in the old coloring was incorect because the
174
-
computation used the sizee of the color map as a metric rather than
186
+
of the old coloring in the old coloring was incorrect because the
187
+
computation used the size of the color map as a metric rather than
175
188
the difference in the number of colors vs the number of arrows. As a
176
189
result self%color_map(coloring(site_i)-size(self%color_map,1),2) -1
![TODO] Get rid of all the junk that crept in (writing files, making inactives table, etc. These should all be in routines so that this main routine is still readable)
0 commit comments