stdpp merge requestshttps://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests2021-12-09T20:51:27Zhttps://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/347add bool_decide_negb2021-12-09T20:51:27ZRalf Jungjung@mpi-sws.orgadd bool_decide_negbI had to move up all the `Decision` stuff since the new lemma needs the `not_dec` instance.
Proof by Yun-Sheng.I had to move up all the `Decision` stuff since the new lemma needs the `not_dec` instance.
Proof by Yun-Sheng.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/350remove a Global Arguments Pos.of_nat from the middle of a proof2021-12-13T19:13:40ZRalf Jungjung@mpi-sws.orgremove a Global Arguments Pos.of_nat from the middle of a proofThis doesn't seem to actually be required, at least not in this file...This doesn't seem to actually be required, at least not in this file...https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/352Add tactics `destruct select <pat>` and `destruct select <pat> as <intro_pat>`.2021-12-16T17:29:42ZRobbert KrebbersAdd tactics `destruct select <pat>` and `destruct select <pat> as <intro_pat>`.These tactics are consistent with existing tactics like `revert select` and `rename select`.These tactics are consistent with existing tactics like `revert select` and `rename select`.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/351add some lemmas about `Finite` and `pred_finite`2021-12-22T21:06:21ZGlen Méveladd some lemmas about `Finite` and `pred_finite`Merge it if you like it. Feel free to suggest better naming and placement.Merge it if you like it. Feel free to suggest better naming and placement.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/355Added some useful lemmas for the [last] function2022-01-12T15:13:03ZJonas KastbergAdded some useful lemmas for the [last] functionAdded some utility lemmas for the [last] function.
(precursor to the !353 MR)Added some utility lemmas for the [last] function.
(precursor to the !353 MR)https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/354Add lemmas `elem_of_prefix` and `elem_of_suffix`2022-01-12T15:13:39ZJonas KastbergAdd lemmas `elem_of_prefix` and `elem_of_suffix`Added a lemma that captures that any element [x] that is in a prefix [l1] is also in the full list [l2].
The property is only trivial from the definition of `l1 prefix_of l2` (i.e. `∃ l, l2 = l1 ++ l2`).
Adding the lemma thus avoids brea...Added a lemma that captures that any element [x] that is in a prefix [l1] is also in the full list [l2].
The property is only trivial from the definition of `l1 prefix_of l2` (i.e. `∃ l, l2 = l1 ++ l2`).
Adding the lemma thus avoids breaking abstraction.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/273Declare `equiv` as a rightful rewrite relation, when an `Equiv` instance is available.2022-01-12T17:33:55ZMatthieu SozeauDeclare `equiv` as a rightful rewrite relation, when an `Equiv` instance is available.This makes Iris & Perennial compatible with [Coq PR #13969](https://github.com/coq/coq/pull/13969). Should be entirely backwards compatible.This makes Iris & Perennial compatible with [Coq PR #13969](https://github.com/coq/coq/pull/13969). Should be entirely backwards compatible.Robbert KrebbersRobbert Krebbershttps://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/360Set the priority of the rewrite relation for equiv to not take precedence over...2022-01-13T16:37:57ZMatthieu SozeauSet the priority of the rewrite relation for equiv to not take precedence over...Set the priority of the rewrite relation for equiv to not take precedence over the eq instance from Coq.
This is more backwards compatible performance-wise with the old resolution where `eq` was favored (otherwise we get a 70% slowdown i...Set the priority of the rewrite relation for equiv to not take precedence over the eq instance from Coq.
This is more backwards compatible performance-wise with the old resolution where `eq` was favored (otherwise we get a 70% slowdown in perennial).https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/362Set the priority of the rewrite relation for sqsubseteq to not take precedence over...2022-01-14T10:59:45ZMatthieu SozeauSet the priority of the rewrite relation for sqsubseteq to not take precedence over...Set the priority of the rewrite relation for sqsubseteq to not take precedence over the eq instance from Coq or equiv.Set the priority of the rewrite relation for sqsubseteq to not take precedence over the eq instance from Coq or equiv.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/359Added some useful lemmas about [list_subseteq]2022-01-14T11:03:41ZJonas KastbergAdded some useful lemmas about [list_subseteq]Added some generally useful lemmas about the [list_subseteq] functionAdded some generally useful lemmas about the [list_subseteq] functionhttps://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/361Add `last_cons_Some_ne` lemma for the `last` function2022-01-14T11:58:54ZJonas KastbergAdd `last_cons_Some_ne` lemma for the `last` functionA useful lemma for the [last] function, to reduce `last (x::l) = Some y` to `last l = Some y` whenever `x ≠ y`.
This is useful since using `last_cons` leaves us with
```{coq}
match last l with
| Some y0 => Some y0
| None => Some x
end = ...A useful lemma for the [last] function, to reduce `last (x::l) = Some y` to `last l = Some y` whenever `x ≠ y`.
This is useful since using `last_cons` leaves us with
```{coq}
match last l with
| Some y0 => Some y0
| None => Some x
end = Some y
```
which only reduces through a subgoal.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/363Add lemmas `app_cons_eq_inv_{l,r}`.2022-01-14T15:31:17ZRobbert KrebbersAdd lemmas `app_cons_eq_inv_{l,r}`.This MR contains two commits: One with the actual lemmas, and one to perform some rearrangements. To prove the `_r` variant, I needed properties of `reverse`. I thus moved the `reverse` lemmas above the `elem_of` lemmas.
The lemmas in t...This MR contains two commits: One with the actual lemmas, and one to perform some rearrangements. To prove the `_r` variant, I needed properties of `reverse`. I thus moved the `reverse` lemmas above the `elem_of` lemmas.
The lemmas in this MR generalize the ones in !356. They also allow proving a variant for `suffix_of`, which is missing in that MR.
/cc @jihgfeehttps://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/353Add `filter_reverse`, `head_filter_Some`, `last_filter_Some` lemmas2022-01-19T09:39:46ZJonas KastbergAdd `filter_reverse`, `head_filter_Some`, `last_filter_Some` lemmasAdded the lemma:
```coq
Lemma last_filter_postfix P `{!∀ x : A, Decision (P x)} l x
last (filter P l) = Some x →
∃ l1 l2, l = l1 ++ [x] ++ l2 ∧ Forall (λ z, ¬ (P z)) l2.
```
(and its counterpart for `head`)
The lemmas seem useful an...Added the lemma:
```coq
Lemma last_filter_postfix P `{!∀ x : A, Decision (P x)} l x
last (filter P l) = Some x →
∃ l1 l2, l = l1 ++ [x] ++ l2 ∧ Forall (λ z, ¬ (P z)) l2.
```
(and its counterpart for `head`)
The lemmas seem useful and general enough to fit in stdpp.
The `last` lemma derives the property `Forall (λ z, ¬ (P z)) l2` of the postfix of a split `l1 ++ [x] ++ l2` for a list `l` whenever `last (filter P l) = Some x`.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/356Renamed `app_cons_eq_inv_{l,r}`, and added derived versions for `prefix/suffix`2022-01-19T16:49:29ZJonas KastbergRenamed `app_cons_eq_inv_{l,r}`, and added derived versions for `prefix/suffix`Renamed the lemmas `app_cons_eq_inv_{l,r}`, to `elem_of_app_cons_inv_{l,r}`, to be more descriptive and consistent with existing lemmas.
Additionally, derived versions for `prefix/suffix`, as they recall some non-immediate properties of...Renamed the lemmas `app_cons_eq_inv_{l,r}`, to `elem_of_app_cons_inv_{l,r}`, to be more descriptive and consistent with existing lemmas.
Additionally, derived versions for `prefix/suffix`, as they recall some non-immediate properties of the relations on similar list structures.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/364std++ 1.7.0 release notes2022-01-22T17:44:23ZTej Chajedtchajed@gmail.comstd++ 1.7.0 release notesFirst step towards a release, which will help with #129.First step towards a release, which will help with #129.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/366Add set_unfold_list_bind (+ test)2022-02-20T18:22:59ZPaolo G. GiarrussoAdd set_unfold_list_bind (+ test)`set_unfold_bind` for sets already exists; this brings the list variant on par.`set_unfold_bind` for sets already exists; this brings the list variant on par.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/367Workaround to avoid `injection` from unfolding equalities on `dom`2022-02-24T15:41:35ZRobbert KrebbersWorkaround to avoid `injection` from unfolding equalities on `dom`As reported by @vsiles and @msammler on [Mattermost](https://mattermost.mpi-sws.org/iris/pl/6gkx9t39rpnxmghhdq7jkn45fe) the tactics `simplify_eq` and `injection` accidentally unfold equalities `dom (gset A) m1 = dom (gset A) m2` into `me...As reported by @vsiles and @msammler on [Mattermost](https://mattermost.mpi-sws.org/iris/pl/6gkx9t39rpnxmghhdq7jkn45fe) the tactics `simplify_eq` and `injection` accidentally unfold equalities `dom (gset A) m1 = dom (gset A) m2` into `merge ...`.
The culprit is that the definition of `dom` on `gmap` starts with a constructor `MapSet`. This MR tweaks the definition a bit, by inserting an eta expansion, to avoid that.https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/370Avoid universe bumps of `gset`/`mapset` (fixes #134)2022-03-16T10:04:56ZRobbert KrebbersAvoid universe bumps of `gset`/`mapset` (fixes #134)I don't understand universes well enough to see why the previous version does not work. So if anyone can explain that, that would be great :).I don't understand universes well enough to see why the previous version does not work. So if anyone can explain that, that would be great :).https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/369Add lemmas list_to_set_disj_elem_of_list and gmultiset_difference_disj_union_{l,r}2022-03-16T10:14:29ZLéo StefanescoAdd lemmas list_to_set_disj_elem_of_list and gmultiset_difference_disj_union_{l,r}https://gitlab.mpi-sws.org/iris/stdpp/-/merge_requests/371Clarify relationship between `gset_to_gmap` and `set_to_map`.2022-04-07T20:10:33ZRobbert KrebbersClarify relationship between `gset_to_gmap` and `set_to_map`.As a response to a question by @jung at Mattermost: https://mattermost.mpi-sws.org/iris/pl/xdgqgkyw9ifcmj3zdsy8dyb18e I added some comments.
Aside from the differences pointed out in the added comments, the `gset` function was there bef...As a response to a question by @jung at Mattermost: https://mattermost.mpi-sws.org/iris/pl/xdgqgkyw9ifcmj3zdsy8dyb18e I added some comments.
Aside from the differences pointed out in the added comments, the `gset` function was there before the generic `set` function.