| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
| |
from ghc-9.2 on, the “default” langauge of GHC is expected to change
more wildly. To prepare for that (and unblock
https://gitlab.haskell.org/ghc/ghc/-/merge_requests/4853), this sets the
language for all the test files to `Haskell2010`. This should insolate
this test suite against changes to the default.
Cherry-picked from https://github.com/haskell/haddock/pull/1341
|
| |\ |
|
| | |\ |
|
| | | | |
|
| | | | |
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This means that `html-test`, `latex-test`, `hoogle-test`, and
`hypsrc-test` now only depend on GHC boot libs. So we should
now be able to build and run these as part of GHC's testsuite. \o/
The reference output has changed very slightly, in three ways:
* we don't convert quotes back into `"` as the `xml` lib did
* we don't add extra ` ` as the `xml` lib did
* we now remove the entire footer `div` (instead of just emptying it)
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Now that Haddock is moving towards working entirely over `.hi` and `.hie` files,
all declarations and types are going to be synthesized via the `Convert` module.
In preparation for this change, here are a bunch of fixes to this module:
* Add kind annotations to type variables in `forall`'s whose kind is not `Type`,
unless the kind can be inferred from some later use of the variable. See
`implicitForAll` and `noKindTyVars` in particular if you wish to dive into this.
* Properly detect `HsQualTy` in `synifyType`. This is done by following suit with
what GHC's `toIfaceTypeX` does and checking the first argument of
`FunTy{} :: Type` to see if it classified as a given/wanted in the typechecker
(see `isPredTy`).
* Beef up the logic around figuring out when an explicit `forall` is needed. This
includes: observing if any of the type variables will need kind signatures, if the
inferred type variable order _without_ a forall will still match the one GHC
claims, and some other small things.
* Add some (not yet used) functionality for default levity polymorphic type
signatures. This functionality similar to `fprint-explicit-runtime-reps`.
Couple other smaller fixes only worth mentioning:
* Show the family result signature only when it isn't `Type`
* Fix rendering of implicit parameters in the LaTeX and Hoogle backends
* Better handling of the return kind of polykinded H98 data declarations
* Class decls produced by `tyThingToLHsDecl` now contain associated type
defaults and default method signatures when appropriate
* Filter out more `forall`'s in pattern synonyms
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| | |
Correspondingly, we wrap all inline/diplay math in
<span class="mathjax"> ... the math .... </span>
This fixes #959.
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| | |
This solves an issue reported about the content looking
incredibly small on mobile devices.
|
| | |
| |
| |
| |
| | |
The package name element in the package-header is now a div instead of a
paragraph, and it is now above the menu ul.links instead of below.
|
| |/ |
|
| | |
|
| |
|
|
|
|
|
|
| |
* Fix minimal pragma handling
Class declarations contain 'ClassOpSig' not 'Typesig'. This should fix #834.
* Accept html-test output
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
| |
* Remove unnecessary call to 'collapseSection'
The call is unnecessary since there is no corresponding toggle for hiding the
section of orphan instances.
* Use <details> for collapsibles
This makes them work even when JS is disabled. Closes #560.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add compile step that bundles and compresses JS files
Also, manage dependencies on third-party JS libraries using NPM.
* Compile JS from TypeScript
* Enable 'noImplicitAny' in TypeScript
* QuickJump: use JSX syntax
* Generate source maps from TypeScript for easier debugging
* TypeScript: more accurate type
* Separate quick jump css file from ocean theme
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
18ed871afb82560d5433b2f53e31b4db9353a74e switched to a new MathJax URL
but didn't update the tests.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Test: a correct behaviour for fields comma-separating values. I'm
surprised we had no bug open for this. Maybe it affects how #301 renders
now but I doubt.
Operators: Seems GHC is giving us a new order for operators, something
must have changed on their side again. cc @haasn , this makes the fixity
to the side not match the order on the LHS which is a bit unpleasant.
Maybe the fixity can be made to match the GHC order?
Bug335: We expand examples by default now.
Bug310: Now inferred safe.
|
| | |
|
| | |
|
| |
|
|
|
|
| |
Conflicts:
haddock-api/src/Haddock/Backends/Xhtml/Decl.hs
haddock-api/src/Haddock/Convert.hs
|
| |
|
|
|
| |
This was long overdue, now running ./accept.lhs on a clean test from
master will not generate a bunch of changes.
|
| |
|
|
|
|
| |
Because the change is in GHC 7.9 and we now work against 7.8.3, this
test no longer makes sense. We revert it until 7.10 becomes the standard
version. If anything, there should be a test for this in GHC itself.
|
| |
|
|
|
|
|
| |
The order of signature groups has been corrected upstream. Here we add a
test case and update some existing test-cases to reflect this change. We
remove grouped signature in test cases that we can (Minimal,
BugDeprecated &c) so that the test is as self-contained as possible.
|
| |
|
|
| |
These were accidentally left there when the tests were originally added
|
| | |
|
| |
|
|
|
|
|
| |
This moves them in-line with their corresponding lines, similar to a
presentation envision by @hvr and described in #ghc.
Redundant operator names are also omitted when no ambiguity is present.
|
| |
|
|
|
| |
Identical fixities declared for the same line should now render using
syntax like: infix 4 <, >=, >, <=
|
|
|
Affects functions, type synonyms, type families, class names, data type
names, constructors, data families, associated TFs/DFs, type synonyms,
pattern synonyms and everything else I could think of.
|
