* simple test to hedge against existing sql migration files
* copyright
* adding better failure message
* adding intructions on generating the digest file
* Migration guide for new assistant
* Add some more words about the changes and start explaining config file.
* Flesh out migration of configs.
* Fix summary of changes.
* Fix formatting of list
* Tidying up migration guide
For long builds, it’s nice to know what damlc is doing. We might want
to consider making this configurable via some --quiet flag but given
that we already output the “Created out.ar.” message unconditionally
it seems reasonable to add this without an option for now.
* Add configuration to LedgerInitialConditions
* Wait until index state is established in reference index service
This fixes a race condition where the daml-on-x server has been
brought up before the index service can serve it.
* Remove ledgerId checks from index service
Now that we have decided on non-changing ledger identifiers, remove
the ledgerId checks from index service and simplify the interface.
* Reformat and fix leftover ??? from DamlOnXTransactionService
Seems like an array that is only declared but never assigned to cannot
be dereferenced with `arr[@]` under `set -u`. We can easily work around
this by using an empty string as default value like `:-`.
* Remove the requirement for Binary on Shake rules (was not used)
* Add a deriving Show on Event, easier for external integrations
* Rename GeneratePackageState to LoadPackageState and move its fields to the rule, rather than the key
* Inline getPackageState away
* Change to passing a ModRenaming to the package loader. Two reasons:
1) When loading non-DAML things we might want to omit the renaming
2) The type ModRenaming has documentation of semantics, unlike [(String, String)]
- change jobs names to match the names Azure uses, so GitHub doesn't get
confused. These were the names when the pipeline was originally set
up; since I changed them it looks like Azure and GitHub don't agree on
test names, resulting in some occurrences of GitHub forever expecting
more info about `linux` while `Linux` is completed.
- add `-i` to cURL invocations, so we can see the response headers. This
is an attempt to debug a situation I have seen three times now where
Azure says it executed the cURL step successfully, yet no message
appears in Slack.
- add test name to failure message sent to Slack.
* Delete Test.Tasty.HUnit.Extended - no one was using the extended pieces
* Delete stuff from TH.Extended that was unused
* Delete stuff from Control.Monad.Except.Extended and simplify
* Delete unused stuff from Data.Text.Extended
* Inline the one use of runDefaultExceptT
* Delete Control.Monad.Except.Extended
* Clean up getAssociatedVirtualResources by inlining and thinking
* Fix dumb typos in getVirtualResources
* Remove usages of Control.Monad.Except.Extended
* Add a missing package import
This reverts commit 3d8acde916.
For some reason that commit seems to have resulted in a lot of
"unexpected end of file" errors during cache downloads. I do not know
what is going on here or how to fix it so let’s revert it for now.
* Use DA.Prelude a lot less
* Make LANGUAGE pragmas have the right case
* Don't reexport DA.Prelude
* Remove more DA.Prelude imports
* Remove the nastiest renaming DA.Prelude
* Final removal of DA.Prelude
* Kill DA.Prelude
* Add missing dependencies
* More missing package imports
* More missing package imports
* Update compiler/daml-lf-ast/src/DA/Daml/LF/Ast/Base.hs
* Update compiler/daml-lf-ast/src/DA/Daml/LF/Ast/Base.hs
This PR copies the hack that I used for the other
data-default-instances-* packages to data-default-instances-dlist. The
exact conditions under which this seems to be required probably depend
on things like your username and the directory of your project which
is why this failed for Neil but not for me.
* Make record constructors in contract keys work
Currently, they don't always work because we don't inline record constructors
aggressively enough. The PR address this flaw by inlining more constructors.
Fixes#890.
* Remove left over comment
* Support the inline return type syntax for recursive functions
Currently, the inline return type syntax does not work for recursive
functions as it produces a locat letrec which the compiler can't handle
yet. The PR works around this problem by doing a very limited form of
lambda lifting.
This fixes#747.
* Apply Neil's suggestions
I often end up in situations where I need to to look at the GHC Core or
the DAML-LF before type checking. In the current parallel setting they
ususally don't get written to disk before the first error pops up. By
running the compiler stages one after another, we make sure that the
outputs of successful stages can be written to disk before starting the
next stage.
* webide: add vscode settings to disable telementry
* webide: remove keybindings
* Adding settings.json to webide server image, in order to disable
telemetry popup.
* clean webide 400 and 500 error messages (which reveal too much about
software)
* disable daml guidelines link until it can be fixed
* more hiding of content from front end
Back in the day, every test case needed a `main` scenario and that was the
only scenario tested. Now, all scenarios are tested and hence a few of these
`main` scenarios are no longer necessary.
* Add Functor/Applicative/Action instances for reader monad
This is possible because we dropped DAML-LF 1.0 support from the
compiler lately.
* Add release notes
* Fix language-server-tests
* Deprecate DA.{Map, Set} in favour of DA.Next.{Map, Set}
`DA.Next.{Map, Set}` were known as `DA.Experimental.{Map, Set}` until now.
We also remove the tests for `DA.{Map, Set}` because our test infrastructure
can't nicely deal with tons of deprecation warning. Since we're not going to
change these files anymore, this seems to be a fair trade-off.
* Fix code spans in release notes
Submitting a command via the CommandService now returns either the
transaction id (SubmitAndWaitForTransactionId), the flat transaction
(SubmitAndWaitForTransactionResponse), or the transaction tree
(SubmitAndWaitForTransactionTreeResponse).
This means that users don't have to wade through the transaction stream
to retrieve the resulting transaction. This is particularly useful in
combination with #479.
Fixes#406
* Allow to mangle names used in scenarios before their executions.
This allows to run a scenario against a long-running server repeatedly and avoid
clashes between runs, since each party is unique (up to the randomness used).
* ScenarioRunner: add test for partyNameMangler
Before it was garbled and wrong:
Range: Range Start: Position Line: 2Character: 15End: Position Line: 2Character: 15
After it's correct and like it was before:
Range: 3:15-3:17
