Skip to content

LFS: beta rules toggle + severity filter on the Flows page#723

Closed
RubenHalman wants to merge 4 commits into
SalesforceLabs:mainfrom
RubenHalman:lfs-beta-rules-severity-filter
Closed

LFS: beta rules toggle + severity filter on the Flows page#723
RubenHalman wants to merge 4 commits into
SalesforceLabs:mainfrom
RubenHalman:lfs-beta-rules-severity-filter

Conversation

@RubenHalman
Copy link
Copy Markdown
Contributor

@RubenHalman RubenHalman commented May 17, 2026

Hey @VinceFINET 👋

Two small quality-of-life improvements for the Lightning Flow Scanner integration on the Flows page.

What's in here

Beta rules toggle
Added a checkbox in the Flows section to include or exclude LFS beta rules. When you flip it, the flows dataset cache is busted and the scan reruns with the updated ruleset — so you always get results for the right set of rules without having to hit the global refresh button.

Severity filter
Added a dropdown next to the checkbox: All severities / Errors only / Warnings & above / Notes & above. Since the violations are already stored after the scan, changing the filter just re-mixes the cached data — no extra calls to Salesforce.

Two new options in the Flows section powered by the Lightning Flow Scanner integration.

- Added a checkbox to include/exclude beta rules in the scan. Toggling it busts the dataset cache so flows are re-scanned with the right ruleset.
- Added a severity dropdown (All / Errors only / Warnings & above / Notes & above). Filtering on severity re-mixes already-cached data, so no extra API calls to Salesforce.

On the data side, lfsViolations now stores { name, severity } objects instead of plain strings, which is what makes the severity filter possible. Updated secretsauce formulas and table column definitions accordingly.
@VinceFINET
Copy link
Copy Markdown
Collaborator

I really appreciate the time you put in this PR. Don't you think that this option would complexify the app? I would prefer you to tell me which is the best options we can put by default in the app. By experience the less option i set in the app the better. Users in the end do not get the impact on changing the options, and finally set all options to true and launch the scanning... So it's maybe better to come up with a default setting that would be the same one for everyone. what do you think? and maybe if they want to go deeper in the analysis they can use directly LFS directly?! Thank you.

@RubenHalman
Copy link
Copy Markdown
Contributor Author

RubenHalman commented May 18, 2026

Thank you for the feedback @VinceFINET.

What I’d recommend is excluding note severity findings by default. In practice, warning and error levels are usually the ones that surface the important actionable issues. So I think a better approach will be:

  • Remove the severity dropdown entirely
  • Keep beta rules disabled by default
  • Use a fixed minimum severity threshold of warning as single flow scanner setting.

That keeps the experience simple👍

Based on feedback — fewer options is better for users who don't know
the impact of changing them.

Beta rules stay off (false) and severity is fixed at 'warning' and above.
Notes are filtered out by default as they tend to be low-signal for most orgs.
The infrastructure to drive these from parameters is still in place if needed.
@RubenHalman
Copy link
Copy Markdown
Contributor Author

RubenHalman commented May 23, 2026

Good call @VinceFINET — updated! Removed both UI controls. Beta rules stay off and severity is now fixed at warning and above (notes filtered out silently). I will try to open a new PR with that approach. Severity will be hardcoded to warning and above for now. Do you want to keep it that way, or is there a place in the app where you'd want it to be configurable (e.g. a settings section, localStorage, something else)?

@RubenHalman RubenHalman deleted the lfs-beta-rules-severity-filter branch May 23, 2026 09:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants