UN-3770 [FEAT] Opt-in pagination for list endpoints; wire Workflows page#2187
Merged
Conversation
Four list endpoints (workflows, prompt studio, adapters, connectors)
returned every row as a bare array and filtered client-side, making the
pages slow to load as orgs grow.
Backend: add `OptionalPagination` — it paginates only when the caller sends
`?page`/`?page_size`, otherwise returns None so DRF serialises the bare
array. These endpoints are shared with dropdown/selector consumers that
expect an array, so the opt-in keeps their responses byte-identical while
the listing page opts in and receives the {count,next,previous,results}
envelope. Attach it to the four viewsets and add server-side name search;
Workflow also gets a deterministic order (id tiebreaker) for stable pages.
Frontend: the Workflows page now paginates + searches server-side, reusing
the existing `usePaginatedList` hook and the Pipelines wiring pattern. This
is the first vertical slice — the prompt-studio, adapters and connectors
pages will follow (their backends already ship the dormant pagination).
Test: utils/tests/test_pagination.py pins the opt-in contract (bare array
without a page param; envelope with one).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
🚧 Files skipped from review as they are similar to previous changes (1)
Summary by CodeRabbit
WalkthroughThe change adds opt-in pagination and server-side name search to backend list endpoints, then updates the workflow frontend to request, render, and refresh paginated results while preserving bare-array responses when pagination is not requested. ChangesPagination and search flow
Estimated code review effort: 3 (Moderate) | ~25 minutes Sequence Diagram(s)sequenceDiagram
participant Workflows
participant usePaginatedList
participant workflowService
participant WorkflowViewSet
Workflows->>usePaginatedList: request page, page size, and search
usePaginatedList->>workflowService: getProjectList(params)
workflowService->>WorkflowViewSet: GET workflow list
WorkflowViewSet-->>workflowService: filtered results and pagination metadata
workflowService-->>Workflows: update project list and pagination state
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
SonarCloud python:S5332 flags the clear-text http:// scheme in the request-stub build_absolute_uri; the scheme is inert in this DB-free contract test but tripping the B security rating on the PR. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z
chandrasekharan-zipstack
marked this pull request as ready for review
July 20, 2026 15:15
Contributor
|
| Filename | Overview |
|---|---|
| backend/utils/pagination.py | Adds optional pagination that preserves bare-array responses when no pagination parameter is provided. |
| backend/utils/tests/test_pagination.py | Adds contract tests for the opt-in pagination behavior and blank pagination parameters. |
| backend/workflow_manager/workflow_v2/views.py | Adds optional pagination, server-side workflow search, and deterministic workflow ordering. |
| backend/adapter_processor_v2/views.py | Adds optional pagination, adapter name search, and deterministic adapter ordering. |
| backend/connector_v2/views.py | Adds optional pagination, connector name search, and deterministic connector ordering. |
| backend/prompt_studio/prompt_studio_core_v2/views.py | Adds optional pagination, prompt studio search, and deterministic prompt studio ordering. |
| frontend/src/components/workflows/workflow/Workflows.jsx | Updates the Workflows page to fetch paginated results, run server-side search, and refresh the active page after mutations. |
| frontend/src/components/workflows/workflow/workflow-service.js | Updates the workflow list API helper to send pagination and search parameters. |
| frontend/src/components/workflows/workflow/Workflows.css | Adds layout styling for the Workflows pagination control. |
Reviews (4): Last reviewed commit: "UN-3770 [FIX] Lock blank-pagination-para..." | Re-trigger Greptile
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 4
🧹 Nitpick comments (2)
frontend/src/components/workflows/workflow/Workflows.jsx (1)
2-2: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueRemove unused
Typographyimport.The
Typographycomponent fromantdis imported but never used in this file.♻️ Proposed refactor
-import { Pagination, Typography } from "antd"; +import { Pagination } from "antd";🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@frontend/src/components/workflows/workflow/Workflows.jsx` at line 2, Remove the unused Typography import from the antd import declaration in Workflows.jsx, while retaining the Pagination import.frontend/src/components/workflows/workflow/workflow-service.js (1)
27-34: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueDeclare
optionsto prevent global variable leakage.The
optionsobject is assigned without a variable declaration, which creates an implicit global variable. While this appears to be a pre-existing pattern in this service file, it is best practice to declare variables locally.♻️ Proposed refactor
getProjectList: (params = {}) => { - options = { + const options = { url: `${path}/workflow/`, method: "GET", params, }; return axiosPrivate(options); },🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@frontend/src/components/workflows/workflow/workflow-service.js` around lines 27 - 34, Declare options locally within getProjectList before assigning the request configuration, preventing implicit global leakage while preserving the existing axiosPrivate call and request parameters.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@backend/adapter_processor_v2/views.py`:
- Around line 193-196: Ensure stable pagination ordering after the for_user()
queryset filtering: in backend/adapter_processor_v2/views.py lines 193-196,
backend/connector_v2/views.py lines 108-112, and
backend/prompt_studio/prompt_studio_core_v2/views.py lines 177-180, apply an
explicit descending modified_at ordering with ascending id as the unique
tiebreaker before returning each queryset (including qs in the prompt studio
view).
In `@frontend/src/components/workflows/workflow/Workflows.jsx`:
- Around line 155-163: Remove the unconditional handleListRefresh() call after
setAlertDetails in the workflow update success handler, preserving the existing
refresh inside the editingProject?.name branch while retaining the refresh
behavior for non-edit updates.
- Around line 139-141: Update the getProjectList failure handler in the workflow
component to set projectList to an empty array before logging the error,
ensuring the existing projectList-based rendering shows the empty state rather
than an indefinite SpinnerLoader after the initial request fails.
- Around line 128-137: Update the paginated workflow fetch handling around the
results assignment and setProjectList call so an empty results array on a page
greater than the first automatically refetches or navigates to the previous page
before updating the displayed list. Keep page 1 empty behavior unchanged, and
ensure pagination state reflects the fallback page so controls remain usable.
---
Nitpick comments:
In `@frontend/src/components/workflows/workflow/workflow-service.js`:
- Around line 27-34: Declare options locally within getProjectList before
assigning the request configuration, preventing implicit global leakage while
preserving the existing axiosPrivate call and request parameters.
In `@frontend/src/components/workflows/workflow/Workflows.jsx`:
- Line 2: Remove the unused Typography import from the antd import declaration
in Workflows.jsx, while retaining the Pagination import.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 8c025dc8-c7d4-4a4e-b448-90c184af9b36
📒 Files selected for processing (9)
backend/adapter_processor_v2/views.pybackend/connector_v2/views.pybackend/prompt_studio/prompt_studio_core_v2/views.pybackend/utils/pagination.pybackend/utils/tests/test_pagination.pybackend/workflow_manager/workflow_v2/views.pyfrontend/src/components/workflows/workflow/Workflows.cssfrontend/src/components/workflows/workflow/Workflows.jsxfrontend/src/components/workflows/workflow/workflow-service.js
…pty-page fallback - OptionalPagination: treat blank ?page=/?page_size= as not opting in so shared callers keep their bare-array response (Greptile). - adapter/connector/prompt-studio get_queryset: order by the DISTINCT ON field for deterministic pagination, incl. the admin/service branch that has no distinct; not modified_at (conflicts with DISTINCT ON) (CodeRabbit). - Workflows.jsx: step back a page when a delete empties the current page (CodeRabbit). - Drop redundant "server-side name search" WHAT comments. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z
…tch failure - Drop the unconditional handleListRefresh() after edit; the branch already refreshes existing edits and navigates for new ones (CodeRabbit). - Default projectList to [] on fetch failure so the empty state shows instead of an indefinite spinner (CodeRabbit). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z
Contributor
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
frontend/src/components/workflows/workflow/Workflows.jsx (1)
148-150: 🩺 Stability & Availability | 🟡 Minor | ⚡ Quick winPrevent early clearance of loading state during chained requests.
Using an unconditional
.finally()block to clear the sharedloadingstate inadvertently hides the spinner when a chained operation has already initiated a new request. This causes the UI to temporarily freeze while the second network call is in flight.
frontend/src/components/workflows/workflow/Workflows.jsx#L148-L150: IngetProjectList,finallyclears the loading state even when falling back to the previous page (getProjectList(page - 1)). Remove.finally()and explicitly callsetLoading(false)inside.then()(after state updates, skipping the fallback return) and.catch().frontend/src/components/workflows/workflow/Workflows.jsx#L179-L181: IneditProject,finallyclears the loading state immediately afterhandleListRefresh()triggers a new list fetch. Remove.finally()and let the list refresh manage the loading state, while explicitly clearing it in theopenProjectand.catch()paths.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@frontend/src/components/workflows/workflow/Workflows.jsx` around lines 148 - 150, The loading state is cleared prematurely by unconditional finally blocks during chained requests. In frontend/src/components/workflows/workflow/Workflows.jsx lines 148-150, update getProjectList to remove finally and call setLoading(false) in then after state updates, except before returning for the previous-page fallback, and in catch. In lines 179-181, update editProject to remove finally, let handleListRefresh manage loading after refresh, and explicitly clear loading in the openProject and catch paths.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@frontend/src/components/workflows/workflow/Workflows.jsx`:
- Around line 148-150: The loading state is cleared prematurely by unconditional
finally blocks during chained requests. In
frontend/src/components/workflows/workflow/Workflows.jsx lines 148-150, update
getProjectList to remove finally and call setLoading(false) in then after state
updates, except before returning for the previous-page fallback, and in catch.
In lines 179-181, update editProject to remove finally, let handleListRefresh
manage loading after refresh, and explicitly clear loading in the openProject
and catch paths.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 7153dbac-8d3d-4cc9-ab52-3e1ac91d512a
📒 Files selected for processing (7)
backend/adapter_processor_v2/views.pybackend/connector_v2/views.pybackend/prompt_studio/prompt_studio_core_v2/views.pybackend/utils/pagination.pybackend/utils/tests/test_pagination.pybackend/workflow_manager/workflow_v2/views.pyfrontend/src/components/workflows/workflow/Workflows.jsx
💤 Files with no reviewable changes (1)
- backend/workflow_manager/workflow_v2/views.py
🚧 Files skipped from review as they are similar to previous changes (5)
- backend/utils/pagination.py
- backend/utils/tests/test_pagination.py
- backend/prompt_studio/prompt_studio_core_v2/views.py
- backend/adapter_processor_v2/views.py
- backend/connector_v2/views.py
pk-zipstack
approved these changes
Jul 21, 2026
…e size
Cover the mixed blank/non-blank pagination params. A blank ?page= alongside a
real ?page_size= serves the first page rather than raising NotFound, since DRF
resolves the page number as `query_params.get("page") or 1`.
Also drop the Workflows page size from 12 to 10 to match the other paginated
listing pages, which use the usePaginatedList default.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z
Contributor
Frontend Lint Report (Biome)✅ All checks passed! No linting or formatting issues found. |
|
Contributor
Unstract test resultsPer-group results
Critical paths
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What & why
Jira: UN-3770
Four listing endpoints — workflows, prompt studio, adapters, connectors — return every row as a bare array and filter client-side, so the pages get slow to load as an org grows (noticeably Prompt Studio projects and Workflows). This adds opt-in server-side pagination.
The regression constraint
These endpoints are shared. Besides the listing page they feed dropdowns/selectors that expect a bare array:
/adapter?adapter_type=…→CombinedOutput,OutputForDocModal,AdapterSelectionModal,AddLlmProfile,DefaultTriadconnector/→ConfigureDsworkflow/list →EtlTaskDeployAttaching a DRF
pagination_classunconditionally would turn all their responses into the{count,next,previous,results}envelope and break every one of those consumers.Approach
OptionalPagination(utils/pagination.py): paginates only when the request carries?page/?page_size, otherwise returnsNone→ DRF serialises the bare array. Only the listing page opts in.?search=icontains).WorkflowViewSetgets a deterministic order (-modified_at, id) — its manager uses plain.distinct()with no default order, which pagination needs for stable pages. The other three already order byDISTINCT ON (id/tool_id).usePaginatedListhook and the Pipelines wiring pattern. Bare-array fallback (data.results ?? data) kept.Scope / follow-ups
This is the first vertical slice. The prompt-studio, adapters and connectors frontend pages are intentionally not converted yet — their backends already ship the (dormant) pagination, and each page has intricate search/refresh flows that warrant live validation one at a time under the same ticket.
Testing
backend/utils/tests/test_pagination.pypins the opt-in contract (bare array without a page param; envelope with one). Verified standalone (backend test tier needs the full env).ruff+ frontendbiomeclean.🤖 Generated with Claude Code
https://claude.ai/code/session_01XzUpb2Dyunz2DCdwegR62Z