[bobbai00]

What changes were proposed in this PR?

Redefines the execution result and error types shared by agent-service and the frontend. SyncExecutionResource is unchanged: adaptLegacySyncExecutionResult converts its legacy response in one place.

• OperatorExecutionSummary replaces OperatorInfo, keeping state, errors, result summary and console messages separate; WorkflowExecutionSummary replaces SyncExecutionResult.
• WorkflowExecutionState is the outcome; a legacy success=false, state=Completed becomes Failed.
• Result modes use the full WebOutputMode union, including SetDeltaMode.
• Sampled rows are [rowIndex, Tuple] pairs, not records with a __row_index__ marker.
• Compilation and execution failures share WorkflowFatalError.
• GET /agents/:id/operator-results returns only resultSummary, the field the UI reads.
• executeOperator output drops the Input operator(table shape) line.
• The popover renders as before; its never-populated Column Statistics section is removed.

Before (main)	After

Any related issues, documentation, discussions?

Closes #5750
Part of #5747
Supersedes #5927

How was this PR tested?

• agent-service: bun run typecheck, bun run format:check, bun test.
• frontend: yarn format:ci, yarn build:ci, and the agent-interaction, agent.service and workflow-editor specs.
• End to end on main's backend images, with Claude Haiku 5.5 behind access-control-service and LiteLLM. Given a penguins data-science prompt, the agent built and ran a 10-operator workflow and recovered from a failed type cast. Every popover showed its table or plot, as on main.

Was this PR authored or co-authored using generative AI tooling?

Generated-by: Claude Opus 4.8 (1M context), Claude Fable 5, Codex, Claude Code (Claude Opus 5.5)

[github-actions]

Automated Reviewer Suggestions

Based on the git blame history of the changed files, we recommend the following reviewers:

• Contributors with relevant context: @aglinxinyuan, @Xiao-zhen-Liu, @mengw15
  You can notify them by mentioning @aglinxinyuan, @Xiao-zhen-Liu, @mengw15 in a comment.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.74%. Comparing base (ed53531) to head (ff15542).

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #6009      +/-   ##
============================================
+ Coverage     91.75%   96.74%   +4.99%     
============================================
  Files          1281      436     -845     
  Lines         56777    28821   -27956     
  Branches       7454     3937    -3517     
============================================
- Hits          52096    27884   -24212     
+ Misses         2801      503    -2298     
+ Partials       1880      434    -1446     

Flag	Coverage Δ
access-control-service	?
agent-service	99.55% <100.00%> (+0.02%)	⬆️
amber	?
computing-unit-managing-service	?
config-service	?
file-service	?
frontend	96.42% <100.00%> (+0.05%)	⬆️
notebook-migration-service	?
pyamber	?
workflow-compiling-service	?

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

⚠️ Benchmark changes need a look

🟢 0 better · 🔴 6 worse · ⚪ 9 noise (<±5%) · 0 without baseline

Compared against main 6fdfe19 benchmarked on this same runner, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated.

Dashboard · Run

	config	throughput	MB/s	latency	max Δ latest / 7d
🔴	bs=10 sw=10 sl=64	392	0.239	24,977/40,305/40,305 us	🔴 +34.4% / 🔴 +162.2%
🔴	bs=100 sw=10 sl=64	795	0.485	124,362/157,297/157,297 us	🔴 +11.0% / 🔴 +47.1%
🔴	bs=1000 sw=10 sl=64	878	0.536	1,133,885/1,234,278/1,234,278 us	🔴 +8.3% / 🔴 +21.0%

Baseline details

Latest main 6fdfe19 from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	392 tuples/sec	402 tuples/sec	776.68 tuples/sec	-2.5%	-49.5%
bs=10 sw=10 sl=64	MB/s	0.239 MB/s	0.245 MB/s	0.474 MB/s	-2.4%	-49.6%
bs=10 sw=10 sl=64	p50	24,977 us	24,611 us	12,599 us	+1.5%	+98.3%
bs=10 sw=10 sl=64	p95	40,305 us	29,994 us	15,370 us	+34.4%	+162.2%
bs=10 sw=10 sl=64	p99	40,305 us	29,994 us	18,723 us	+34.4%	+115.3%
bs=100 sw=10 sl=64	throughput	795 tuples/sec	806 tuples/sec	994.07 tuples/sec	-1.4%	-20.0%
bs=100 sw=10 sl=64	MB/s	0.485 MB/s	0.492 MB/s	0.607 MB/s	-1.4%	-20.1%
bs=100 sw=10 sl=64	p50	124,362 us	122,406 us	100,754 us	+1.6%	+23.4%
bs=100 sw=10 sl=64	p95	157,297 us	141,719 us	106,898 us	+11.0%	+47.1%
bs=100 sw=10 sl=64	p99	157,297 us	141,719 us	113,210 us	+11.0%	+38.9%
bs=1000 sw=10 sl=64	throughput	878 tuples/sec	917 tuples/sec	1,027 tuples/sec	-4.3%	-14.5%
bs=1000 sw=10 sl=64	MB/s	0.536 MB/s	0.56 MB/s	0.627 MB/s	-4.3%	-14.5%
bs=1000 sw=10 sl=64	p50	1,133,885 us	1,090,660 us	979,814 us	+4.0%	+15.7%
bs=1000 sw=10 sl=64	p95	1,234,278 us	1,139,977 us	1,019,689 us	+8.3%	+21.0%
bs=1000 sw=10 sl=64	p99	1,234,278 us	1,139,977 us	1,049,906 us	+8.3%	+17.6%

Raw CSV

config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,510.52,200,128000,392,0.239,24976.86,40305.48,40305.48
1,100,10,64,20,2514.86,2000,1280000,795,0.485,124362.14,157297.35,157297.35
2,1000,10,64,20,22791.94,20000,12800000,878,0.536,1133884.88,1234278.44,1234278.44

[Copilot]

Pull request overview

This PR refactors the sync-execution wire contract across the Amber backend, agent-service, and frontend so operator execution results are represented as structured summaries (OperatorExecutionSummary / WorkflowExecutionSummary), sampled rows become { rowIndex, tuple }, and execution errors reuse the shared WorkflowFatalError shape already used by compilation errors.

Changes:

• Redesign sync-execution DTOs: replace flat per-operator result payloads with OperatorExecutionSummary (state, errors, optional result summary, optional console messages) and WorkflowExecutionSummary.
• Switch sampled output rows from embedded metadata fields (e.g., __row_index__) to an explicit SampleRow { rowIndex, tuple } shape and update formatting/mapping accordingly.
• Align execution error reporting with the engine/compilation WorkflowFatalError structure across backend + agent-service + frontend consumers, with updated unit tests.

Reviewed changes

Copilot reviewed 24 out of 24 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
frontend/src/app/workspace/service/agent/agent.service.ts	Updates frontend REST typing/storage to consume OperatorExecutionSummary (and new SampleRow/result mode/state enums).
frontend/src/app/workspace/service/agent/agent.service.spec.ts	Updates unit tests for the new operator-results REST response shape.
frontend/src/app/workspace/component/workflow-editor/workflow-editor.component.ts	Switches operator result accessors to resultSummary.sampleTuples + resultSummary.resultMode.
frontend/src/app/workspace/component/workflow-editor/workflow-editor.component.html	Wires new inputs (sampleTuples, resultMode) into the agent interaction UI.
frontend/src/app/workspace/component/agent/agent-interaction/agent-interaction.component.ts	Updates UI logic to render sample rows with explicit rowIndex and visualization via resultMode.
frontend/src/app/workspace/component/agent/agent-interaction/agent-interaction.component.html	Updates template to display a leading Row column and read values from SampleRow.tuple.
frontend/src/app/workspace/component/agent/agent-interaction/agent-interaction.component.scss	Removes column-statistics styling tied to removed statistics rendering.
frontend/src/app/workspace/component/agent/agent-interaction/agent-interaction.component.spec.ts	Adds coverage for the new table/ellipsis/visualization rendering behavior with SampleRow.
amber/src/main/scala/org/apache/texera/web/resource/SyncExecutionResource.scala	Refactors sync-execution DTOs and result sampling/truncation into helper methods; emits WorkflowFatalError consistently.
amber/src/test/scala/org/apache/texera/web/resource/SyncExecutionResourceSpec.scala	Adds extensive unit tests for extracted pure helpers (error classification, sampling/truncation, summary assembly).
agent-service/src/types/execution.ts	Redefines execution result/error interfaces (states, SampleRow, WorkflowFatalError, operator/workflow summaries).
agent-service/src/server.ts	Returns visible operator summaries verbatim (now OperatorExecutionSummary) from /operator-results.
agent-service/src/server.spec.ts	Updates server route tests to assert the new operator-results payload shape.
agent-service/src/api/compile-api.ts	Re-exports shared WorkflowFatalError from the centralized execution types module.
agent-service/src/agent/workflow-result-state.ts	Stores OperatorExecutionSummary in result state and updates accessor typing.
agent-service/src/agent/workflow-result-state.spec.ts	Updates tests for visibility/ancestor-walk semantics with new summary shape.
agent-service/src/agent/tools/workflow-execution-tools.ts	Updates execute-and-format tool to consume WorkflowExecutionSummary and derive warnings/errors from console/error summaries.
agent-service/src/agent/tools/workflow-execution-tools.spec.ts	Adds new unit tests for execution tool behavior under the redesigned model.
agent-service/src/agent/tools/tools-utility.ts	Centralizes “operator failed” detection + TSV formatting for SampleRow[]; derives warnings from console messages.
agent-service/src/agent/tools/tools-utility.spec.ts	Updates tests for operator error extraction and header handling.
agent-service/src/agent/tools/result-formatting.ts	Updates formatting of stored results to use SampleRow[] and new visualization/result mode semantics.
agent-service/src/agent/tools/result-formatting.spec.ts	Updates tests for new summary fields, warning derivation, visualization stripping, and row gap handling.
agent-service/src/agent/texera-agent.ts	Adjusts DAG-result formatting call signature after input-shape removal.
agent-service/src/agent/texera-agent.spec.ts	Adds basic coverage for formatted DAG results using the new summary shape.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Yicong-Huang]

This PR is tooo large to review. Please keep the size smaller!!! I left comments on key files.

• amber/src/main/scala/org/apache/texera/web/resource/SyncExecutionResource.scala
• agent-service/src/types/execution.ts

[Yicong-Huang]

🟢 0 resolved · 0 open · 0 new (0 new = 0 newly introduced · 0 late catches)

Verification trace

This round is a rebase onto current main, so the review checked that the rebase moved no PR code: the sorted added lines of the PR diff are identical at the old base (8b0fb16) and the new base (ed53531) across all 27 changed files, and the 6 files whose content differs differ only by absorbed main lines. The four agent-service specs pass at the new head: 142 pass, 0 fail.

[bobbai00]

@Yicong-Huang can you take another look ? I want to merge this one soon

[sshiv012]

@bobbai00 can we merge this? Do you need an additional reviewer?