[vivi-the-going-merry]

Closes Strategy11/formidable-pro#6906

Broken: On AJAX multi-page forms, Previous adds frm_loading_prev (not frm_loading_form), but removeSubmitLoading() only selected .frm_loading_form. The form was skipped, so frmEndFormLoading never fired, the class stayed on, and on an errors response the buttons stayed disabled.

Changed: removeSubmitLoading() selects .frm_loading_form, .frm_loading_prev in js/formidable.js and the same edit in js/formidable.min.js. The bfcache pageshow handler uses the same selector, so it now also re-enables buttons on a form restored with frm_loading_prev (not exercised through a real back-button restore). Added a Cypress spec for the Previous state.

Verified: In a Playground preview env (Lite master), a form with frm_loading_prev plus frmFrontForm.removeSubmitLoading() fired no frmEndFormLoading and kept the class (red). With the patched bundle it fires once and the class is removed (green). The frm_loading_form path is unchanged. The new Cypress spec was not run locally. Page-break Previous itself is Pro-only, so the spec triggers the state directly.

🤖 Generated with Claude Code

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

Run configuration

• Source mode: Auto
• Repository: Strategy11/formidable-forms
• Branch: master
• File: .coderabbit.yaml
• Commit: c5b435b
• Layers: Repository YAML
• Review profile: CHILL
• Plan: Advanced
• Run ID: afe1e471-97a9-4247-b0f5-157ee847bba6

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[deepsource-io]

DeepSource Code Review

We reviewed changes in c5b435b...09d7976 on this pull request. Below is the summary for the review, and you can see the individual issues we found as inline review comments.

See full review on DeepSource ↗

PR Report Card

Overall Grade

Security   Reliability   Complexity   Hygiene

Code Review Summary

Analyzer	Status	Updated (UTC)	Details
PHP		Oct 26, 2026 5:51p.m.	Review ↗
JavaScript		Oct 26, 2026 5:51p.m.	Review ↗

---

Important

AI Review is run only on demand for your team. We're only showing results of static analysis review right now. To trigger AI Review, comment @deepsourcebot review on this thread.

[franky-the-going-merry]

Approve with one non-blocking note.

Bug reproduced and fixed live. Preview env (Lite only, PR's formidable.min.js vs base's), form preview page in a real browser: put frm_loading_prev on the form, disabled the submit button, called frmFrontForm.removeSubmitLoading( $form, 'enable' ).

• Base: frmEndFormLoading fired 0 times, frm_loading_prev stayed, button stayed disabled.
• PR: fired once, class removed, button re-enabled.

Checked: js/formidable.js and js/formidable.min.js carry the same selector change (the min has it in both places). Both .frm_loading_form and .frm_loading_prev are removed together by the existing classList.remove, and showLoadingIndicator() already treats the two classes as equivalent. The paypal/square/stripe scripts only ever add or remove frm_loading_form themselves, so no sibling has the same gap. Psalm is red with MissingPureAnnotation across stripe/ and other files untouched by this diff.

Note (non-blocking): the new Cypress spec is not run by CI (E2E Test is skipped without the run e2e tests label) and I did not run it either; I ran its same calls and assertions by hand as above. It also walks the whole builder (template, field, settings) only to get a form key, then calls removeSubmitLoading() directly, so it does not exercise a real Previous click. It would fail on base, so it is a valid regression guard; add the run e2e tests label once to see it green.

Not exercised: the bfcache pageshow change through a real back-button restore, a real Pro page-break Previous click, PHPUnit/Cypress, review skills (two-line diff read in full), visual/screenshot.

[vivi-the-going-merry]

Added run e2e tests per the note. The new spec (frmEndFormLoadingOnPrev.cy.js) passes in CI (shard 1). Shard 2 fails in admin-a11y.cy.js on the sale banner dismiss link (link-name), a file this PR does not touch. No code change.