Inline Execution

[4e6]

Description

The engine should allow the execution of arbitrary expressions at the given point in the program with access to local variables defined in the scope.

Currently, the language server provides two endpoints with similar functionality:

• executionContext/executeExpression
• executionContext/attachVisualization

The difference is that the executeExpression evaluates the value once and then "detaches" from the expression, and the attachVisualization leaves the expression attached and re-evaluates when the program changes.

ToDo

The limitation of the current functionality is that it does not allow to access expressions in the local scope, for example given the program,

main =
    operator1 = ...
    operator2 = ...

the IDE should be able to request the execution of an expression in the scope of the main method, and the expression can contain operator1 and operator2 symbols.

Notes

• When designing the new JSON-RPC protocol, how should the scope be specified? One way would be to provide the scope as an expressionId, and then assume that the execution happens in the same scope as expressionId, after the expression is defined.

[JaroslavTulach]

Where to find inline parsing related functionality:

• Compiler.scala#runInline handles the conversion into IR
• it is used by EnsoLanguage.parse(InlineParsingRequest request) to integrate that with other Truffle languages
• the EvalNode - it probably does exactly what the IDE needs, so possibly just use it "somehow"
• and last, but not least there is the ReplDebuggerInstrument and it uses the node: private @Child EvalNode evalNode = EvalNode.buildWithResultScopeCapture();

The right functionality must be somewhere along these lines.

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-10-23):

Progress: Started working on the task. Had to refresh in my head the visualization infrastructure. Drafted the API. Started experimenting with the inline context. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-10-24):

Progress: Continue working on the task. Drafted the runtime infrastructure for getting the inline context and evaluating the expression. Started working on the execution of the evaluated expression. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-10-25):

Progress: Continue working on the task. Debugging failures related to the evaluated expression. Fixed an issue with the execution. Debugging an issue with the inability to access the local scope during the execution. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-10-26):

Progress: Continue working on the task. Fixed the issue with the local scope. Implemented dedicated logic that evaluates the expression in the local scope. Started working on the runtime tests. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-10-27):

Progress: Continue working on the task. Started working on wiring the logic to the language server. Updated the engine API. Added the tests. Started to work on the language server part. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-10-30):

Progress: Continue working on the task. Tried to unify the two methods of evaluating visualization expressions - in global and local scope. Turned out that the evaluation in local scope is quite different and it is better to implement a separate handling for it. It should be finished by 2023-10-30.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new 🔴 DELAY for today (2023-10-31):

Summary: There is 2 days delay in implementation of the Inline Execution (#8132) task.
It will cause 2 days delay for the delivery of this weekly plan.

Delay Cause: Need extra time to implement commodity logic around executing expressions in the local scope. Initially thought that I can reuse the infrastructure for visualizations, but turned out that it is a bit different and requires special handling

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-10-31):

Progress: Continue working on the task. Finished implementing the new engine API for executing expressions in the local scope. Fixed the engine tests. It should be finished by 2023-11-01.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-11-01):

Progress: Continue working on the task. Implemented the tests for the new ExecuteExpression message. Fixed related errors. Updated the PR. Spent some time debugging the language server initialization issue in the cloud. It should be finished by 2023-11-01.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new 🔴 DELAY for yesterday (2023-11-02):

Summary: There is 5 days delay in implementation of the Inline Execution (#8132) task.
It will cause 5 days delay for the delivery of this weekly plan.

Delay Cause: After the discussion of initial implementation, decided that the expression evaluation logic should be on the fast path. This will require some modification of the instrumentation logic

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-11-02):

Progress: Continue working on the task. Started working on moving the execution logic on the fast path. Separated the Callbacks interface into the node execution callbacks and execution support callbacks. Started implementation of the new methods. It should be finished by 2023-11-06.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-11-03):

Progress: Continue working on the task. Implemented the expression execution as a part of execution callbacks. Implemented a new onExpressionExecuted callback that sends the result of the computed expression to the user. Started working on fixing the tests. It should be finished by 2023-11-06.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-11-06):

Progress: Continue working on the task. Was working on the review comments. Reverted the changes to the callbacks interface. Cleaned up the module and instrumentor changes. It should be finished by 2023-11-06.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new 🔴 DELAY for today (2023-11-10):

Summary: There is 7 days delay in implementation of the Inline Execution (#8132) task.
It will cause 1 day delay for the delivery of this weekly plan.

Delay Cause: Returning to the task after a break. Need one day to implement the new instrumentation interface.

[enso-bot]

Dmitry Bushev reports a new STANDUP for today (2023-11-10):

Progress: Continue working on the task. Updated the branch to a new develop. Implemented the new instrumentation interface. Fixed the tests. It should be finished by 2023-11-13.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-11-09):

Progress: Started working on the task. Updated the runner logic. Started working on the language server part. Implemented requests. It should be finished by 2023-11-13.

Next Day: Next day I will be working on the #8132 task. Continue working on the task

[enso-bot]

Dmitry Bushev reports a new STANDUP for yesterday (2023-11-13):

Progress: Continue working on the task. Updated the PR. Fixed the review comments. Final testing. Merged. It should be finished by 2023-11-13.

Next Day: Next day I will be working on the #8132 task. Continue working on the task