8 Commits
Author SHA1 Message Date
saphidandClaude Opus 5.5 989962aecc Contact email: a report's rev is its own change; another address starts fresh
- from_report applies its change and reads the id and rev together, so a
  removal made while that change is sending is newer than the report; the
  report's redaction window now starts before the address is saved.
- A report with a different address replaces the saved one with follow-up
  questions only: update notices aren't carried over to an address nobody
  agreed them for, and the form says so before sending.
- Settings refreshes after every report send, whatever the box shows by then.
- privacy.md: a report with follow-up ticked also saves and sends the address.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-01 21:25:22 +10:00
saphidandClaude Opus 5.5 08d75e3ffb Contact email: follow-up given with a report is kept and removable; match by rev
- Ticking follow-up questions on a report makes that address the contact
  email (follow-up ticked, update choice unchanged), so Settings shows it
  and Remove my email withdraws it like any other.
- Reports carry contact_rev; the inbox takes a report's follow-up
  permission back when a later change from that copy (higher rev) no
  longer agrees, whatever the clocks say.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-01 21:08:16 +10:00
saphidandClaude Opus 5.5 01d5c612c0 Contact email: withdrawal covers earlier reports; consent is a real true
- A report with follow-up ticked carries this copy's contact id, and the
  inbox marks its permission withdrawn when a later choice from that copy
  no longer agrees to follow-up questions at that address.
- The one-time prompt never appears in a visit that showed the privacy
  notice, even if the Frame connects just after it's dismissed.
- Saving contact details isn't headset work: it can't hold up switching
  headsets or be refused after a switch.
- Consent flags must be JSON true/false; "false" is no longer consent.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-01 20:59:23 +10:00
saphidandClaude Opus 5.5 2ca0e6924a Contact email tests: pin the in-gap save and same-second report timing
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 10:38:54 +10:00
saphidandClaude Opus 5.5 cf2db32721 Contact email: don't strand a change saved as a send finishes; time reports exactly
Third review follow-ups:
- A sender that found nothing waiting checks again after letting go of the
  send lock, so a change saved in that moment is sent, not left for a retrier.
- A report is compared with a removal using its full-precision start time, so
  a report sent after the address was removed is logged as sent.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 10:35:25 +10:00
saphidandClaude Opus 5.5 3f0f09b138 Contact email: don't block Save on a slow send; redact reports still in flight
Second review follow-ups:
- Saving returns once the choice is stored; a send already under way picks up
  the newest change, or the background retry is woken.
- A problem report still being sent when its address is removed is logged as
  <removed>, checked under the same lock the removal holds.
- The prompt re-checks the privacy notice after fetching its state.
- docs/privacy.md: offline contact changes are sent later by themselves; the
  prompt never follows straight after the privacy notice.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 10:29:17 +10:00
saphidandClaude Opus 5.5 b8ed53f2ff Contact email: newest choice wins by rev, removal wipes the local log
Review follow-ups:
- Each contact_consent event carries a rev that goes up with every change,
  sends are serialized, and `contacts` picks every field from the highest
  rev per copy, so a withdrawal can't lose to an earlier event sent in the
  same second or with a skewed clock.
- Removing the address also replaces it with <removed> in the local
  sent log (earlier contact events and problem reports).
- The prompt is rechecked when the Frame connects, not only at page load.
- No thanks hides the bar only once the dismissal is saved.
- docs/privacy.md: say that the analytics switches don't block a report or
  contact change the person sends deliberately.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 10:22:45 +10:00
saphidandClaude Opus 5.5 19a0d0af18 Ask for an optional contact email, with separate update and follow-up consent
Problem reports arrive with no way to reply. People can now leave an email
address with two separate opt-ins: occasional update notices, and follow-up
questions from the maintainer.

- ui/frame_contact.py keeps the address and choices locally and sends each
  change privately to PostHog as a contact_consent event under its own random
  contact id; removing the address sends a withdrawal without it. Changes made
  offline wait and are retried.
- A one-time, dismissible prompt appears after the Frame first connects; No
  thanks and showing it once are both remembered.
- Privacy & updates gains a Contact email section to add, change or remove it.
- The report form's contact field now goes with a report only when "may
  contact me with follow-up questions" is ticked (contact_followup).
- frame_report.py contacts [updates|followup] lists who agreed to what,
  using the newest event per copy.
- docs/privacy.md says what is collected, why, where and how to remove it.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 10:15:53 +10:00