Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

agda pr 7349 #2441

Closed
wants to merge 316 commits into from
Closed

agda pr 7349 #2441

wants to merge 316 commits into from

Conversation

andreasabel
Copy link
Member

Taneb and others added 30 commits July 10, 2024 10:43
* Eliminate many propositional equality imports

* Fix merge conflict in Data.Bool.Properties

Co-authored-by: G. Allais <[email protected]>

---------

Co-authored-by: G. Allais <[email protected]>
To fix agda/agda#6654, we've decided that large indices will no longer
be allowed by default. There is an infective flag `--large-indices` to
bring them back, but none of the uses of large indices in the standard
library were essential: to avoid complicated mutually-recursive PRs
across repos, I adjusted the levels to check with `--no-large-indices`
instead of adding the flag to the modules that used them.
* Simplify some `Relation.Binary` imports

* Format

* Better naming scheme

* Fix tests

* Merge issues
* Simplify leftover `Function` imports

* `Function` -> `Function.Base`

Co-authored-by: G. Allais <[email protected]>

---------

Co-authored-by: G. Allais <[email protected]>
* Simplify more `Data.Product` import to `Data.Product.Base`

* Missed an import
…#2019)

* Simplify import of `Data.List.Relation.Binary.Pointwise` in agda-stdl

* rectifications on Data.List.Relation.Binary.Pointwise import

* rectifications on Data.List.Relation.Binary.Pointwise import

* Delete weights-README.dot

* Delete weight.png

* Delete README.dot
* Simplify more `Data.Product` import to `Data.Product.Base`

* Simplify more `Data.Product` import to `Data.Product.Base`

* Indent
andreasabel and others added 28 commits July 10, 2024 10:43
* Haskell CI (for GenerateEverything) and dependencies bumped to GHC 9.10.1

* CI: bump some versions, satisfy some shellcheck complaints
* refactor `scanr` etc.

* restore `inits` etc. but lazier; plus knock-on consequences

* more refactoring; plus knock-on consequences

* tidy up

* refactored into `Base`

* ... and `Properties`

* fix-up `inits` and `tails`

* fix up `import`s

* knock-ons

* Andreas's suggestions

* further, better, refactoring is possible

* yet further, better, refactoring is possible: remove explicit equational reasoning!

* Update CHANGELOG.md

Fix deprecated names

* Update Base.agda

Fix deprecations

* Update Properties.agda

Fix deprecations

* Update CHANGELOG.md

Fix deprecated names
* refactored from #2350

* enriched the `module` contents in response to review comments
* More list properties about `catMaybes/mapMaybe`

- Follow-up to PR #2361
- Ported from my [formal-prelude](https://github.com/omelkonian/formal-prelude/tree/c10fe94876a7378088f2e4cf68d9b18858dcc256)

* Revert irrefutable `with`s for inductive hypotheses.
* add some 'very dependent' maps to Data.Product. More documentation to come later.

* and document

* make imports more precise

* minimal properties of very-dependen map and zipWith. Structured to make it easy to add more.

* whitespace

* implement new names and suggestions about using pattern-matching in the type

* whitespace in CHANGELOG

* small cleanup based on latest round of comments

* and fix the names in the comments too.

---------

Co-authored-by: jamesmckinna <[email protected]>
)

* refactor towards `if_then_else_`

* layout

* `let` into `where`
* prototype for fixing #2199

* delegate to `Relation.Nullary.Negation.Core.weak-contradiction`

* refactoring: lift out `Recomputable` in its own right

* forgot to add this module

* refactoring: tweak

* tidying up; added `CHANGELOG`

* more tidying up

* streamlined `import`s

* removed `Recomputable` from `Relation.Nullary`

* fixed multiple definitions in `Relation.Unary`

* reorder `CHANGELOG` entries after merge

* `contradiciton` via `weak-contradiction`

* irrefutable `with`

* use `of`

* only use *irrelevant* `⊥-elim-irr`

* tightened `import`s

* removed `irr-contradiction`

* inspired by #2329

* conjecture: all uses of `contradiction` can be made weak

* second thoughts: reverted last round of chnages

* lazier pattern analysis cf. #2055

* dependencies: uncoupled from `Recomputable`

* moved `⊥` and `¬ A` properties to this one place

* removed `contradictionᵒ` and rephrased everything in terms of `weak-contradiction`

* knock-on consequences; simplified `import`s

* narrow `import`s

* narrow `import`s; irrefutable `with`

* narrow `import`s; `CHANGELOG`

* narrow `import`s

* response to review comments

* irrelevance of `recompute`

* knock-on, plus proof of `UIP` from #2149

* knock-ons from renaming

* knock-on from renaming

* pushed proof `recompute-constant` to `Recomputable`
* `contradiction` against `⊥-elim`

* tightened `import`s

* added two operations `∃∈` and `∀∈`

* added to `CHANGELOG`

* knock-on for the `Propositional` case

* refactor `lose∘find` and `find∘lose` in terms of new lemmas `∃∈-Any∘find` and `find∘∃∈-Any`

* `CHANGELOG`

* reorder

* knock-on viscosity

* knock-on viscosity; plus refactoring of `×↔` for intelligibility

* knock-on viscosity

* tightened `import`s

* misc. import and other tweaks

* misc. import and other tweaks

* misc. import and other tweaks

* removed spurious module name

* better definition of `find`

* make intermediate terms explicit in proof of `to∘from`

* tweaks

* Update Setoid.agda

Remove redundant import of `index` from `Any`

* Update Setoid.agda

Removed more redundant `import`ed operations

* Update Properties.agda

Another redundant `import`

* Update Properties.agda

Another redundant `import`ed operation

* `with` into `let`

* `with` into `let`

* `with` into `let`

* `with` into `let`

* indentation

* fix `universal-U`

* added `map-cong`

* deprecate `map-compose` in favour of `map-∘`

* explicit `let` in statement of `find∘map`

* knock-on viscosity: hide `map-cong`

* use of `Product.map₁`

* revert use of `Product.map₁`

* inlined lemmas!

* alpha conversion and further inlining!

* correctted use of `Product.map₂`

* added `syntax` declarations for the new combinators

* remove`⊥-elim`

* reordered `CHANGELOG`

* revise combinator names

* tighten `import`s

* tighten `import`s

* fixed merge conflict bug

* removed new syntax

* knock-on
* port over two modules

* and add to CHANGELOG

* fix whitespace

* fix warning: it was pointing to a record that did not exist.

* fix things as per Matthew's review - though this remains a breaking change.

* take care of comments from James.

* adjust CHANGELOG for what will be implemented shortly

* Revert "take care of comments from James."

This reverts commit 93e9e0f.

* Revert "fix things as per Matthew's review - though this remains a breaking change."

This reverts commit d1cae72.

* Revert "fix whitespace"

This reverts commit 81230ec.

* Revert "port over two modules"

This reverts commit 6619f11.

* rename these

* fix tiny merge issue

* get deprecations right (remove where not needed, make more global where needed)

* style guide - missing blank lines

* fix a bad merge

* fixed deprecations

* fix #2394

* minor tweaks

---------

Co-authored-by: James McKinna <[email protected]>
…es` (#2402)

* fix #2138

* fixed `Structures`; added `Bundles`

* added fields; plus redefined `IsSemilattice` and `IsBoundedSemilattice` as aliases

* `fix-whitespace`

* reexported `comm`

* fixed `Lattice.Bundles`

* knock-on

* added text about redefinition of `Is(Bounded)Semilattice`

* add manifest fields to `IsIdempotentSemiring`

* final missing `Bundles`

* `CHANGELOG`
* [ admin ] deprecate Word -> Word64

* [ new ] a type of bytes

* [ new ] Bytestrings and builders

* [ test ] for bytestrings, builders, word conversion, show, etc.

* [ ci ] bump ghc & cabal numbers

* [ fix ] actually set the bits ya weapon

* [ ci ] bump options to match makefile

* [ ci ] more heap

* [ more ] add hexadecimal show functions too
* Update LICENSE year

* Remove extraneous 'i'

---------

Co-authored-by: Lex van der Stoep <[email protected]>
* additional proofs and patterns in `Data.Nat.Properties`

* added two projections; refactored `pad*` operations

* `CHANGELOG`

* removed one more use

* removed final uses of `<″-offset` outside `Data.Nat.Base|Properties`

* rename `≤-proof` to `m≤n⇒∃[o]m+o≡n`

* removed new pattern synonyms

* interim commit: deprecate everything!

* add guarded monus; make arguments more irrelevant

* fixed up `CHANGELOG`

* propagate use of irrelevance

* tidy up deprecations; reinstate `s<″s⁻¹` for `Data.Fin.Properties`

* tightened up the deprecation story

* paragraph on use of `pattern` synonyms

* removed `import`

* Update CHANGELOG.md

Fix refs to Algebra.Definitions.RawMagma

* Update Base.agda

Fix refs. to Algebra.Definitions.RawMagma

* inlined guarded monus definition in property

* remove comment about guarded monus
…ary.Decidable.Core` (#2405)

* fixes #2059 on the model of #2336

* fixed `import` dependencies

* simplified `import` dependencies

* final tweak

* `CHANGELOG`
* Make the main home of `decToMaybe` be `Relation.Nullary.Decidable.Core`, but
deprecate the one in `Data.Maybe.Base` instead of removing it entirely.
Fix the library accordingly.

Note that this forces `Relation.Nullary.Decidable.Core` to use `Agda.Builtin.Maybe`
to avoid a cyclic import. This can be fixed once the deprecation is done.

* Update src/Relation/Nullary/Decidable/Core.agda

Good idea.

Co-authored-by: G. Allais <[email protected]>

* simplified the deprecation

* `CHANGELOG`

* narrowed import too far

* tweak a couple of the solvers for consistency, as per suggestions.

* chnage to `public` re-export of `Relation.Nullary.Decidable.Core.decToMaybe`

* Revert "chnage to `public` re-export of `Relation.Nullary.Decidable.Core.decToMaybe`"

This reverts commit 256a505.

* `fix-whitespace`

* simplify `import`s

* make consistent with `Idempotent` case

* tidy up

* instead of an alias, open public so that Agda knows these really are the same. This is a better deprecation strategy.

* rename(provisional)+deprecate

* knock-on

* knock-on: refactor via `dec⇒maybe`

* deprecation via delegation

* fix `CHANGELOG`

---------

Co-authored-by: G. Allais <[email protected]>
Co-authored-by: James McKinna <[email protected]>
Co-authored-by: jamesmckinna <[email protected]>
* fixes #2411

* now via local -defined auxiliaries
* Tidy up CHANGELOG in preparation for v2.1 release candidate

* Fixed WHITESPACE

* Fixed James' feedback and improved alphabetical order
* fixes #2400: use explicit quantification

* fix knock-ons
…2123) (#2365)" (#2423)

This reverts commit 438f9ed.

Specifically, it restores `with`-based definitions of the
`Decidable`-definable functions on `List`s, incl. `filter`
Fixes #2415
couple fixes for changelog rendering
@andreasabel
Copy link
Member Author

Oops wrong repo.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment