Speed up file modification checks
Summary: `getModificationTime` from the `directory` package is really
slow. The `unix` package is faster but still slow. This PR brings the
time spent checking file modifications (which is required on every
change) from ~0.5s to ~0.15s.
* hie-core/test: Dedicated test suite for hie-core
* Apply hlint suggestion
Use System.Environment.Blank's setEnv which has a non-overwrite mode
that implements precisely what we were doing with more code before.
* buildifier fixes
* hie-core-exe works on Windows now - ghc-paths was fixed
* Rename reportSeriousError to reportInternalError
* Stop using logError for logging things that are warnings to the user, not errors by us
* Rename logError
* Sort the log fields properly
* Delete tagAction from Logger
* Strip down the pure logger
* Delete unused pieces of the logger
* A quick check suggests the call stack will be useful in approximately none of the callers of logging, so just remove it
* When reporting an internal error, give as much detail as we can
* Change our logger to be based on Priority values
* HLint fixes
* Rename makeNopLogger
* In hie-core say what level of message you are setting
* Delete the unused makeOneLogger
* Make sure we can show messages floating around
* If a notification/response handler throws an exception, report it upwards
* Remove reportInternalError in favour of a general logging mechanism
* Add missing dependencies
* Just call fail for a dodgy error report
* Add a FIXME
* Make missing modules just an error
* Move the hie-core demo files around (they aren't really a demo anymore)
* Split the command line parsing into a separate module
* Give messages about how long starting something takes
* Make the interactive mode say what it is doing a bit more
* Add a --cwd flag to hie-core
* Take a list of files and directories for hie-core
* Update the readme to say how to test using hie-core
* Fix up the bazel file
* Add HLint exception
* Move mergeHandlers into Server
* Make partial handlers a proper newtype
* Pass the options in to runLanguageServer
* Take in user handlers
* Remove the code lens handler since we don't advertise it
* Add setSomethingModified, rather than faking it for the LSP VFS
* Rewrite the LSP notifications
* Improve the display of info messages around openning and modifying text documents
* Make sure stdout and stderr don't have buffering, so we see their output immediately
* Handle exit properly
* Make notifications forward on to their previous values
* Remove the exit handler, HIE already has a good default for it
* Add comments on FileStore
* Delete unused package dependencies
* Eliminate the use of pretty
* Don't export some things that weren't necessary
* Add a missing dependency
* Avoid reexporting files of interest methods
* Put toIdeResult back
* Delete the entirely unused reflow function
* Don't have Data.Text.Prettyprint.Doc.Syntax reexport stuff
* Move most of the pretty printing inside hie-core
* Get rid of the prettyprinter-syntax dependency from hie-core
* Get rid of prettyprinter-syntax, by collapsing the one definition into DA.Pretty
* Move getFilesOfInterest into the IDE, and OfInterest out into DAML
* Have the IDE mode kick all active files each time
* Add a missing .cabal dependency
* Add a missing dependencies
* Use the right thing, if any files have parse errors, then don't worry
* Print messages to stderr rather than stdout
* Add a big warning that writing to stdout is a bad plan
Previously we had two layers of indirection:
In the compiler we emitted Event, we then translated this to
ClientNotification which was then translated to
FromServerMessage.
Apart from being confusing and convoluted this also resulted in us
doing the conversion to generic LSP types too late so we had scenario
specific code in places where it shouldn’t be.
This PR removes the indirection and just uses FromServerMessage
directly.
The details are described in a comment but the short story is
that a roundtrip Uri -> FilePath -> Uri necessarily loses information
on which characters were escaped. The long-term solution here is to
avoid this roundtrip altogether but this at least fixes the issue for
now.
* Use haskell-lsp’s builtin VFS in "damlc ide"
haskell-lsp has a builtin VFS that it updates automatically on the
corresponding requests. This PR removes our own VFS implementation and
uses that builtin VFS in "damlc ide". To allow the use of functions
like setBufferModified (we use that heavily in daml-ghc-shake-test-ci)
without having to spawn an LSP server, we also add a fallback where we
spin up our own LSP implementation.
* Demo program for haskell-ide-core as a library
* Fix all warnings in the Demo file
* Build the IDE demo
* Give a better error message than undefined
* HLint
* Fix copyright header
* Sort the dependencies
* Improve the comment
* Bazel formatting
* Disable building on Windows until ghc-paths is fixed
* Bazel formatting
* Specify the main function
* Rename the #ifdef for GHC_STABLE
* More CPP required for GHC stable
* Add a compatibility wrapper for HIE functionality which is new in GHC HEAD
* HLint ignores
* Finish the dummy implementations
* Add a bazel build for the ide-core library against GHC
* Add buildifier targets.
The tool allows to check and format BUILD files in the repo.
To check if files are well formatted, run:
bazel run //:buildifier
To fix badly-formatted files run:
bazel run //:buildifier-fix
* Cleanup dade-copyright-headers formatting.
* Fix dade-copyright-headers on files with just the copyright.
* Run buildifier automatically on CI via 'fmt.sh'.
* Reformat all BUILD files with buildifier.
Excludes autogenerated Bazel files.
