Feature #134 closing studies based on selection

[junaidferoz]

Selective Study Closing with Workflow and User Filters

Previously, users could only close all studies at once. This feature adds the ability to selectively close studies by filtering them based on workflow type and/or user, reducing the risk of unintentionally closing studies and providing finer-grained control over bulk operations.

New User Features

• Workflow filter: Select a specific workflow type when closing studies, or choose "All workflows" to close studies across all workflows.
• User filter: Optionally filter studies by the user who owns them, with a default "Any user" option for unfiltered behavior.
• Scoped confirmation: The confirmation dialog now displays which workflow and user filters are applied, along with an approximate count of studies to be closed.

New Dev Features

• Extended studyCloseBulk handler: Backend now accepts optional workflowId and userId parameters to filter study lists before closing.
• Server-side validation: All filtering is enforced on the backend to ensure malicious client requests cannot close out-of-scope studies.

Improvements

• Filtered progress reporting: Progress indicators now reflect only the studies being closed within the selected scope, providing accurate feedback.
• Enhanced safety: Users can preview exactly which studies will be affected before confirming the bulk close operation

[Copilot]

Pull request overview

This PR refines “Bulk Close Studies” so operators can scope bulk-closing to a subset of open studies (by workflow / owner / guest-role), preview the matching count in the UI, and have the backend close exactly the study IDs computed by the client (with a strict mode to avoid falling back to “close everything”).

Changes:

• Updated the bulk-close modal UI to support workflow/user scoping, show a live match count, and send an explicit study id list (studyIds + studyIdsJson) with bulkCloseUseIdList: true.
• Updated the backend studyCloseBulk handler to resolve id lists robustly (array/object/JSON) and return closedCount.
• Adjusted supporting config/ops files (sequelize CLI env loading; additional gitignore entries).

Reviewed changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
frontend/src/components/dashboard/study/BulkCloseModal.vue	Adds workflow/user/guest scoping UI, match count, and emits bulk close with explicit study id list.
frontend/src/components/dashboard/Study.vue	Updates bulk-close button label/tooltip.
backend/webserver/sockets/study.js	Adds id-list resolution + strict mode; filters/loads studies and returns closedCount.
backend/db/.sequelizerc	Loads repo-root .env (and .env.$ENV) for sequelize-cli runs.
.gitignore	Ignores a specific root artifact filename and root logs/.

---

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

[Copilot]

studyCloseBulk currently performs a destructive operation (closing studies) without any server-side authorization check. Since any socket client can emit this event directly, this should enforce a permission (e.g., await this.isAdmin() or await this.hasAccess("frontend.dashboard.studies.closeAllStudies")) before proceeding, otherwise users can bypass the UI restriction and close studies they shouldn't.

Suggested change

	// Enforce server-side authorization for bulk-closing studies.
	if (!(await this.isAdmin()) && !(await this.hasAccess("frontend.dashboard.studies.closeAllStudies"))) {
	throw new Error("No permission to bulk close studies");
	}

[Copilot]

loadStudiesForBulkCloseByIds issues one getById query per id and awaits them sequentially, which becomes an N+1 bottleneck for large bulk-closes. Consider fetching all studies in one query (e.g. this.models.study.getAllByKeyValues('id', ids) / findAll({ where: { id: { [Op.in]: ids } } })) and then filtering by projectId/template in-memory.

[Copilot]

There are several new branching behaviors here (strict bulkCloseUseIdList mode, studyIdsJson parsing, and id-list vs legacy filter selection), but there are no socket tests covering studyCloseBulk in backend/tests/socket.study.test.js. Adding integration tests for the success path (closes exactly the provided ids) and error path (strict mode with missing/invalid id list) would help prevent regressions.

[junaidferoz]

Made the changes based on the review comments. Also metioned which commit is the change made