[PM-39212] Add entry for account.rogers.com

[jprusik]

🎟️ Tracking

PM-39212

[github-actions]

🤖 Bitwarden Claude Code Review

Overall Assessment: APPROVE

This PR adds a single Forms Map entry for account.rogers.com (a 36-line addition to maps/forms/forms.jsonc). The entry models Rogers' two-step Angular login flow (username step wrapped in <ng-component>, then password step wrapped in <password-form>) as one logical account-login form with next and submit actions. I reviewed it against forms.v1.schema.json, the selector lint rules in scripts/lib/lint-selectors.mjs, and the authoring conventions in maps/forms/README.md.

Code Review Details

No blocking findings. Verification notes:

• Schema-valid: correct category, container/fields/actions shapes, and all field/action keys are within the allowed enums.
• Selector lint rules satisfied: every segment carries a tag anchor, no bare/class-only/universal/positional/state pseudos at the top level, container selectors anchor on form, and :has() is supported (already used by existing entries and the recently merged coinbase form).
• Convention-aligned: the multi-step single-URL flow with multiple container alternatives mirrors the existing coinbase and trip.com entries, the host is placed alphabetically, selectors are self-contained, and the non-obvious DOM structure is documented inline per authoring guideline 12.
• Considered but not flagged: the username-step input uses the email field key while carrying autocomplete='username' and a "registered username" title. Because the input is type='email' and the README directs using email when an email is used to authenticate, the author's choice is defensible and treated as an intentional judgment call rather than a finding.

[blackwood]

using the title text here feels brittle, but I'm assuming there was no other readily available distinguishing selector?

[jprusik]

It should be somewhat brittle; it's more that it was a meaningfully differentiating quality.