aboutsummaryrefslogtreecommitdiff
path: root/haddock-api/src
Commit message (Collapse)AuthorAgeFilesLines
* Implement simple string chunking based on HsColour library.Łukasz Hanuszczak2015-06-301-1/+27
|
* Implement function for tagging parsed chunks with source spans.Łukasz Hanuszczak2015-06-301-1/+8
|
* Create scaffolding for Haskell source parser module.Łukasz Hanuszczak2015-06-301-0/+36
|
* Fix alignment of Source links in instance table in FirefoxPhil Ruffwind2015-06-121-2/+3
| | | | | | | | | | | | | | | | | | | Due to a Firefox bug [1], a combination of 'whitespace: nowrap' on the parent element with 'float: right' on the inner element can cause the floated element to be displaced downwards for no apparent reason. To work around this, the left side is wrapped in its own <span> and set to 'float: left'. As a precautionary measure to prevent the parent element from collapsing entirely, we also add the classic "clearfix" hack. The latter is not strictly needed but it helps prevent bugs if the layout is altered again in the future. Fixes #384. Remark: line 159 of src/Haddock/Backends/Xhtml/Layout.hs was indented to prevent confusion over the operator precedence of (<+>) vs (<<). [1]: https://bugzilla.mozilla.org/show_bug.cgi?id=488725
* Attach to instance location the name that has the same location filejpmoresmau2015-06-126-25/+39
| | | | Fixes #383
* Fix haddock: internal error: spliceURL UnhelpfulSpan (#207)Bartosz Nitka2015-06-121-5/+11
| | | | | | | | Inferred type signatures don't have SrcSpans, so let's use the one from the declaration. I've tested this manually on the test-case from #207, but I got stuck at trying to run the test-suite.
* Change ModuleTree Node to carry PackageKey and SourcePackageId to resolve #385Murray Campbell2015-05-112-21/+24
| | | | | Signed-off-by: Austin Seipp <aseipp@pobox.com> (cherry picked from commit 2380f07c430c525b205ce2eae6dab23c8388d899)
* Do not insert anchor for section headings in contents boxwatashi2015-04-261-1/+4
|
* Don't print instance safety information in HoogleMateusz Kowalczyk2015-03-271-2/+11
| | | | Fixes #168
* Output method documentation in Hoogle backendMateusz Kowalczyk2015-03-271-11/+21
| | | | | | | | | One thing of note is that we no longer preserve grouping of methods and print each method on its own line. We could preserve it if no documentation is present for any methods in the group if someone asks for it though. Fixes #259
* Fully qualify names in Hoogle instances outputMateusz Kowalczyk2015-03-271-4/+6
| | | | Closes #263
* Fix Hoogle display of constructorsMateusz Kowalczyk2015-03-271-1/+4
| | | | Fixes #361
* Clearly default to variables in out of scope caseMateusz Kowalczyk2015-03-271-23/+40
|
* Remove now redundant importsMateusz Kowalczyk2015-03-268-10/+0
|
* Make the error encountered when a package can't be found moreBen Gamari2015-03-251-12/+21
| | | | | | user-friendly Closes #369
* Follow API changes in D538Alan Zimmerman2015-01-238-69/+61
| | | | | Signed-off-by: Austin Seipp <aseipp@pobox.com> (cherry picked from commit d61bbc75890e4eb0ad508b9c2a27b91f691213e6)
* Track naming change in DataConSimon Peyton Jones2015-01-231-3/+3
| | | | (cherry picked from commit 04cf63d0195837ed52075ed7d2676e71831e8a0b)
* Sort out some module import warningsMateusz Kowalczyk2015-01-223-4/+1
|
* --package-name and --package-version flagsMateusz Kowalczyk2015-01-223-25/+53
| | | | | | Used for --hoogle amongst other things. Now we need to teach cabal to use it. The situation is still a bit sub-par because if the flags aren't passed in, the crash will occur. Closes #353.
* Links to source location of class instance definitionsjpmoresmau2015-01-226-32/+53
|
* Add missing import for standalone haddock-api packageLuite Stegeman2014-12-221-0/+1
|
* hide projectVersion from DynFlags since it clashes with ↵Luite Stegeman2014-12-191-1/+1
| | | | Haddock.Version.projectVersion
* Print missing docs by defaultMateusz Kowalczyk2014-12-182-4/+4
| | | | Adds --no-print-missing-docs
* Only keep one Version instead of blindly appendingMateusz Kowalczyk2014-12-171-3/+3
|
* Treat GHC 7.10 the same as GHC 7.9Herbert Valerio Riedel2014-12-141-1/+1
| | | | ...since the current GHC 7.9 is going to become GHC 7.10 real-soon-now anyway
* Remove redundant wild-card pattern matchHerbert Valerio Riedel2014-12-141-1/+0
| | | | (this would otherwise cause a build-failure with `-Werror`)
* Various fixups and bumps for next releaseMateusz Kowalczyk2014-12-125-18/+6
|
* For pattern synonyms, render "pattern" as a keywordDr. ERDI Gergo2014-12-121-1/+1
|
* Support for PartialTypeSignaturesThomas Winant2014-12-128-55/+72
| | | | | | | Conflicts: haddock-api/src/Haddock/Backends/Xhtml/Decl.hs haddock-api/src/Haddock/Convert.hs haddock-api/src/Haddock/Interface/Create.hs
* Follow API changes in D426Alan Zimmerman2014-12-128-88/+120
| | | | | | | | | Signed-off-by: Austin Seipp <aseipp@pobox.com> Conflicts: haddock-api/src/Haddock/Backends/LaTeX.hs haddock-api/src/Haddock/Backends/Xhtml/Decl.hs haddock-api/src/Haddock/Convert.hs
* Changes to reflect refactoring in GHC as part of #7484Richard Eisenberg2014-12-121-0/+1
|
* Follow changes from #9812Jan Stolarek2014-12-121-11/+8
| | | | | Conflicts: haddock-api/src/Haddock/Convert.hs
* Update Haddock to new pattern synonym type signature syntaxDr. ERDI Gergo2014-12-125-92/+85
| | | | | | Conflicts: haddock-api/src/Haddock/Backends/Xhtml/Decl.hs haddock-api/src/Haddock/Convert.hs
* Make compatible with `deepseq-1.4.0.0`Herbert Valerio Riedel2014-12-121-3/+3
| | | | ...by not relying on the default method implementation of `rnf`
* Remove overlapping pattern matchAustin Seipp2014-12-121-1/+0
| | | | Signed-off-by: Austin Seipp <aseipp@pobox.com>
* reflect ForeignType constructore removalYuras Shumovich2014-12-121-4/+0
| | | | | | | | Reviewers: austin Reviewed By: austin Differential Revision: https://phabricator.haskell.org/D358
* Collapse user-defined section by default (re #335)Herbert Valerio Riedel2014-12-121-1/+1
| | | | | Conflicts: haddock-api/src/Haddock/Backends/Xhtml/DocMarkup.hs
* Properly render package ID (not package key) in index, fixes #329.Edward Z. Yang2014-12-123-13/+18
| | | | | | | Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu> Conflicts: haddock-api/src/Haddock/ModuleTree.hs
* Followup changes to addition of -fwarn-context-quantification (GHC Trac #4426)Krzysztof Gogolewski2014-12-123-2/+7
|
* Follow changes to TypeAnnot in GHC HEADAlan Zimmerman2014-12-122-11/+26
| | | | | | | Signed-off-by: Austin Seipp <aseipp@pobox.com> Conflicts: haddock-api/src/Haddock/Convert.hs
* Changes due to ghc api changes in package representationDuncan Coutts2014-12-123-27/+17
| | | | | | | | | | Also fix a bug with finding the package name and version given a module. This had become wrong due to the package key changes (it was very hacky in the first place). We now look up the package key in the package db to get the package info properly. Conflicts: haddock-api/src/Haddock.hs
* Revert "Merge branch 'reverts'"Mateusz Kowalczyk2014-12-1213-57/+65
| | | | | | | | This reverts commit 5c93cc347773c7634321edd5f808d5b55b46301f, reversing changes made to 5b81a9e53894d2ae591ca0c6c96199632d39eb06. Conflicts: haddock-api/src/Haddock/Convert.hs
* header could contain several linesjpmoresmau2014-12-101-1/+1
| | | | Closes #348
* Allow the parser to spit out meta-infoMateusz Kowalczyk2014-12-1016-102/+144
| | | | | | | | | | | | | | | Currently we only use it only for ‘since’ annotations but with these patches it should be fairly simple to add new attributes if we wish to. Closes #26. It seems to work fine but due to 7.10 rush I don't have the chance to do more exhaustive testing right now. The way the meta is output (emphasis at the end of the whole comment) is fairly arbitrary and subject to bikeshedding. Note that this makes test for Bug310 fail due to interface version bump: it can't find the docs for base with this interface version so it fails. There is not much we can do to help this because it tests for ’built-in’ identifier, not something we can provide ourselves.
* Fix warningsMateusz Kowalczyk2014-11-043-6/+4
|
* Turn some uses of error into recoverable warningsMateusz Kowalczyk2014-11-043-106/+141
| | | | | | | This should at the very least not abort when something weird happens. It does feel like we should have a type that carries these errors until the end however as the user might not see them unless they are printed at the end.
* Fix re-exports of built-in type familiesMateusz Kowalczyk2014-11-041-4/+7
| | | | Fixes #310
* whitespace onlyMateusz Kowalczyk2014-11-041-9/+9
|
* Experimental support for collapsable headersMateusz Kowalczyk2014-10-294-29/+117
| | | | Closes #335
* Fix improper lazy IO usetreeowl2014-10-091-1/+1
| | | Make `getPrologue` force `parseParas dflags str` before returning. Without this, it will attempt to read from the file after it is closed, with unspecified and generally bad results.