* Add exception tests
changelog_begin
changelog_end
* Add another test
This one tests a transaction that can only be
produced by a privacy-aware ledger.
* Address review comments
@S11001001 is taking care of 1.15.0-snapshot.20210622.7213.0.d867d904 (#10085), so they get pushed back to the end of the line.
Please do not merge this before #10085.
CHANGELOG_BEGIN
CHANGELOG_END
Co-authored-by: Azure Pipelines DAML Build <support@digitalasset.com>
* CHANGELOG_BEGIN
Log ledger api validation failures at info level
CHANGELOG_END
* log validation failures in Api*Service family of classes
* address review comments
* WIP: reinterpret must fail
changelog_begin
changelog_end
EXPLORE: add x/y prefix to reinterpret
MiniEngineTest.scala
MiniTests.daml
setup reinterpret testing one step down, to be able to see rollback nodes
MiniEngineTest.scala - refactor yreinterpret
first minimal reinterpret test
setup testcase for reinterpret on a command which throws
catch exception during reinterpret; and test
introduce type for Validating; avoid boolean blindness
remove exploratory MiniTests
remove exploratory x/y prefix on reinterpret
question
fix build
* small cleanup
* add `unsafeCompileForReinterpretation`; preserve interface of existing `unsafeCompile` (command list)
* remove the Validating type (not needed for this work)
* undo format change
* more private
* update comment for Engine.reinterpret
* test ensure failure works correctly; no rollback expected
* refactor to share duplicated code in reinterpret tests
* remove unhelpful create/ensure tests
* test that reinterpretation of an uncatchable error remains a failure
* test that rollback-version-check-error remains a failure for reinterpretation
* remove dev comment left in by accident
* Introduce shell.nix to expose the Nix tooling, if required.
* dev-env: Rewrite the ghcide script to use nix-shell.
* language-support/hs: Use `nix-shell` for export-package.sh.
* Nix: Add a shebang to the Bazel wrapper script.
CHANGELOG_BEGIN
CHANGELOG_END
* dev-env: Standardize the `set` header in `ghcide`.
* dev-env: Replace `gpg` and `make` with symlinks.
* dev-env: Remove sphinx-autobuild. It's brought in by pipenv.
* dev-env: Remove Python 3.6 references. They seem to be unused.
CHANGELOG_BEGIN
CHANGELOG_END
Since per-request offset can be described strictly as a special case of the new
per-query offset semantics, go ahead and describe it that way, so that really
only one model needs to be understood to fully comprehend the query request
semantics.
CHANGELOG_BEGIN
CHANGELOG_END
changelog_begin
- [JSON-API] Timing metrics which measure how long the processing of a command submission request takes on the ledger are now available
changelog_end
This PR drops two things:
1. The check that the benchmark hasn’t been modified. This hasn’t ever
been useful and it keeps being annoying.
2. It stops the comparison against the old version and instead just
benchmarks the current version. We really only care about the day to
day changes. Comparing against an arbitrary year old version has lost
all meaning at this point.
changelog_begin
changelog_end
* Run scenarios in off-ledger machine
This PR builds on the previous PR that split scenario execution in two
different speedy machines and now actually makes the machine that runs
scenarios run in off-ledger mode just like we handle Daml Script.
This required a bunch of refactoring to make it nice so apologies for
the slightly large PR. Hopefully it’s still relatively easy to follow
and luckily it deletes more code than it adds.
changelog_begin
changelog_end
* Apply suggestions from code review
Co-authored-by: Stefano Baghino <43749967+stefanobaghino-da@users.noreply.github.com>
* review comments
changelog_begin
changelog_end
Co-authored-by: Stefano Baghino <43749967+stefanobaghino-da@users.noreply.github.com>
This morning we started with very restricted CI pools (2/6 for Windows
and 7/20 for Linux), apparently because the region we run in (us-east1)
has three zones, two of them were unable to allocate new nodes, and the
default policy is to distribute nodes evenly between zones.
I've manually changed the distribution policy. Unfortunately this option
is not yet available in our version of the GCP Terraform plugin.
CHANGELOG_BEGIN
CHANGELOG_END
changelog_begin
- [JSON-API] The database operations (regardless of in-memory or postgres) contain now metrics for fetching contracts by id or key (seperate metrics foreach)
- [JSON-API] The time for a repsonse payload construction of a request is now also tracked in a metric
changelog_end
It makes no sense to pass NotVisible to Speedy especially since that
is not how visibility is handled for the Engine. Also it lets us
delete code and I like deleted code.
changelog_begin
changelog_end
* [JSON-API] Add more timing metrics
changelog_begin
- [JSON-API] Timing metrics are now available for party management, package management, command submission and query endpoints.
- [JSON-API] Also added a timing metric for parsing and decoding of incoming json payloads
changelog_end
* Add comments to new metrics
* Split metrics up more & remove obsolete metric
* Split up timers for query endpoints
* Creating OR predicate to reduce length of string literals
CHANGELOG_BEGIN
CHANGELOG_END
* removing unnecessary thrown exception
CHANGELOG_BEGIN
CHANGELOG_END
* switching to camel case for constants
CHANGELOG_BEGIN
CHANGELOG_END
* run format
CHANGELOG_BEGIN
CHANGELOG_END
* remove tolist conversion
CHANGELOG_BEGIN
CHANGELOG_END
* scalafmt
Co-authored-by: Brian Healey <brian.healey@digitalasset.com>
* Use ScenarioRunner.submit in Daml Script
changelog_begin
changelog_end
* privatize ledger variable
changelog_begin
changelog_end
* drop space
changelog_begin
changelog_end
* nvarchar2 keys are text-incompatible, but varchar2 keys are fine
* commit the ACS update before query
* add changelog
CHANGELOG_BEGIN
- [JSON API] The Oracle database schema has changed; if using
``--query-store-jdbc-config``, you must rebuild the database by adding
``,createSchema=true``. See #9895.
CHANGELOG_END
* expand the InitDdl set to include materialized views
* replace search index with a materialized view that expands the stakeholders
* allow materialized views to be created in Oracle testing
* join and query the contract_stakeholders table for party-set membership
- restoring a few elements removed by 3e6661128d (#9484)
This solves two warts in the code:
- the validate/createUnsafe double-parse, because scopt doesn't let you flatMap;
- the non-JdbcConfig sub-configs appeared to need to know the JDBC drivers when
they really don't, because of a quirk in the inherited implementation
That coherence of scopt.Read instances calls for all its dependencies to be
coherent leads us to treat supportedJdbcDriverNames as a nullary typeclass
instance. This is a nullary typeclass by the same justification as
SupportedJdbcDriver; see scaladoc there for more.
And we solve the latter problem by...adding a type parameter, how else.
CHANGELOG_BEGIN
CHANGELOG_END
* Preparation: pull up initialisation
* Preparation: introduce partyArrayContext for more flexibility
* Preparation: add append-only-from-scratch feature to FlywayMigrations
* Add squashed Flyway migration for append-only H2
* Add support for H2 StorageBackend
* Extend conformance test suites for H2
* Extend JDBCLedgerDao unit test suites for H2
* Extend sandbox-classic unit test suites for H2
* Fixing typo, some renames based on review
changelog_begin
changelog_end
@stefanobaghino-da is taking care of 1.14.0-snapshot.20210615.7169.0.adeba206 (#10018), so they get pushed back to the end of the line.
Please do not merge this before #10018.
CHANGELOG_BEGIN
CHANGELOG_END
Co-authored-by: Azure Pipelines DAML Build <support@digitalasset.com>
* Do not terminate transactions/transaction trees streams when no end-offset provided in the ledger-api-bench-tool
CHANGELOG_BEGIN
CHANGELOG_END
* Minor change
* Split scenario & ledger execution
This PR by no means aims to solve everything we can do here. It is
rather the minimal change that I could get to work that provides us
with one Speedy machine for scenario execution (which is still an
on-ledger machine to avoid having to change the callsites to much in
this PR) and one speedy machine per submission.
There is tons of cleanup we can do afterwards but this should
hopefully set the right foundations.
changelog_begin
changelog_end
