Public landing page is now served in English and Italian via path-
prefixed URLs (/en/ and /it/), with the bare / detecting the
visitor's language and 302-ing to the right one.
Routing
-------
* GET / : if authed → dashboard (unchanged). Otherwise the visitor's
language is resolved from (in priority order) the rtm.lang cookie,
the Accept-Language header, the cf-ipcountry geolocation header,
and finally DEFAULT_LANG, then a 302 redirects to /<lang>/.
* GET /en/ + /it/ : render the localised landing template, set the
rtm.lang cookie (1-year, SameSite=Lax) so the next /-visit goes
straight to the same translation. Logged-in users with user.lang
set bypass detection for / (same priority chain — user.lang wins
on every public surface they touch).
Storage
-------
* app/locales/<lang>.yaml — flat-ish nested copy files. YAML chosen
so a future translator can edit without touching Python. Strings
containing inline HTML (<strong>, <em>, <a>) are rendered with the
Jinja `safe` filter in the template.
* app/services/locales.py — loads at startup, exposes get_locale()
and detect_public_lang(). Wraps each YAML tree in a small _Dotted
view so templates can write {{ t.hero.subhead }} (deliberately NOT
a dict subclass — dict's built-in method names would shadow YAML
keys like `items`).
Template + chrome
-----------------
* landing.html ported in full to {{ t.<key> }} references.
* public_base.html gets <html lang="…"> + hreflang link tags (en/it/
x-default) when a route opts in via lang_switch=true. A tiny
EN | IT link group lands in the public header, only visible on
surfaces that opt in.
* public.css picks up the small lang-switch widget styles.
Scope (intentionally narrow)
----------------------------
* Only the landing page is localised. Pricing, terms, privacy,
disclaimer, login, verify all stay English-only for now; their
header chrome stays English too because localising labels there
while the linked content is still EN would be a worse mismatch.
* When other public pages get translated, lang_switch=true on those
routes will surface the same widget there with no template changes.
Tests
-----
* tests/test_locales.py covers YAML load parity (every active
language has a file), dotted access through the tree, the
detection precedence chain, and the unknown-locale fallback.
Deps
----
* pyyaml was already in requirements.lock as a transitive but not
declared. Added to pyproject so it stays pinned as an explicit
direct dependency.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Adds the @media (max-width: 480px) blocks specified in the design:
- dashboard.css: indicator table hides the 'mobile-hide'-tagged
columns (Label, Ccy, 1y, anchor, as-of), keeping Symbol / Price /
1d / 1m. Cell padding + font shrink. Group-tab buttons get a
bigger touch target.
- panels.css: header padding tightens, scroll-body max-height drops
to 60vh so log/news stay above the fold in the stacked layout.
- portfolio.css: overall grid keeps 2 cols (already at 640px) with
tighter gap; action buttons wrap; composer input goes full-width.
- log-chat.css: chat bubbles edge-to-edge, input row stacks, font-
size:14px on form fields to avoid iOS Safari zoom-on-focus.
- news.css: row collapses to age | (title / source) — source moves
under the title. Tag filter strip wraps.
- settings.css: form rows stack (label above input). Import picker
becomes single-column. Buttons full-width.
- auth.css: card padding tightens to free up vertical space when the
iOS keyboard is up. font-size:14px on inputs.
- public.css: hero headline clamp() lower bound drops to 22px; CTAs
stack full-width; pricing tier-grid stacks.
indicators.html: tagged the secondary cells with .mobile-hide rather
than relying on positional nth-child — the anchor column is
conditional and would have shifted positions.
336 tests still pass.
Splits the 2571-line cassandra.css into ten focused stylesheets:
tokens (palette + fonts), layout (chrome), panels, dashboard,
portfolio, log-chat, auth, settings, news, public. base.html and
public_base.html load only what they need; auth pages (login,
verify, unsubscribe confirm) load tokens + layout + auth.
Brand drift-detection test repointed at tokens.css (where the
palette now lives). 291 tests still pass.
