fix(ui): user-friendly error messages at the UI boundary (#6)#7
Conversation
User-visible error surfaces forwarded raw anyhow context strings and
internal labels ("backend: not a valid npub or hex pubkey", "switch
account: ...", "publish failed: ..."), leaking module names and Rust
debug formatting.
Add friendly_error(op, &Error): a two-tier mapper that classifies the
flattened error chain into actionable copy (invalid npub/pubkey ->
"Double-check it", network/relay/timeout -> "Check your relay
settings") with reassuring operation-specific fallbacks. Wire it into
every set_backend_error site (sync, boot, switch-account, accept,
block/decline, archive, unarchive, send, edit, react/unreact) and the
key-package publish/rotate/refresh status. Full technical detail is
still logged via the existing eprintln! at each site for diagnosis.
Also soften the network pane ("Already in the list." -> "That relay is
already in your list.", "Saved."/"Removed." -> "Relay added/removed.",
"Backend not ready yet." -> "Not connected yet...") and stop leaking
save_relays / republish error strings to the UI.
Closes #6
|
Warning Review limit reached
More reviews will be available in 29 minutes and 50 seconds. Learn how PR review limits work. Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file). ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits. 🚦 How do rate limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan refill rate. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, the refill rate gradually slows as usage increases. The highest same-day bursts are limited more strictly. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
WalkthroughIntroduces a thread-safe Slint-backed error message cache ( ChangesLocalized error snapshots and friendly_error
Translation catalog updates for ErrorCopy and themes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related issues
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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. Comment |
|
Ready to review this PR? Stage has broken it down into 8 individual chapters for you: 
Chapters generated by Stage for commit 7bbcbb6 on Jun 18, 2026 2:58am UTC. |
Pip adversarial reviewVerdict: changes requested. Blocking
Suggestions
Verification
|
Address adversarial review on #6: the friendly_error mapper was only wired into set_backend_error, network, and key-package paths. Raw {e:#} anyhow/debug output still reached 13 other user-visible status setters (new chat, add-contact, peer profile, add-member, group settings/image, account add, profile save/upload). Route all of them through friendly_error with operation-specific fallbacks; raw detail still goes to eprintln! at each site. Refs #6
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (1)
src/main.rs (1)
1863-1867: ⚡ Quick winPreserve raw error logs before mapping to friendly UI text
These branches sanitize the UI message but currently drop the underlying error detail entirely. Please log the raw error (
{e:#}) before converting tofriendly_error(...), same pattern used elsewhere.Suggested patch
- Some(b) => b - .republish_relay_lists() - .map_err(|e| friendly_error("republish", &e)), + Some(b) => b + .republish_relay_lists() + .map_err(|e| { + eprintln!("[republish] {e:#}"); + friendly_error("republish", &e) + }),- "publish" => b - .publish_key_package() - .map(|_| "published · your key package is live".to_string()) - .map_err(|e| friendly_error("kp_publish", &e)), + "publish" => b + .publish_key_package() + .map(|_| "published · your key package is live".to_string()) + .map_err(|e| { + eprintln!("[kp_publish] {e:#}"); + friendly_error("kp_publish", &e) + }), "rotate" => b .rotate_key_package() .map(|_| "rotated · published a fresh key package".to_string()) - .map_err(|e| friendly_error("kp_rotate", &e)), + .map_err(|e| { + eprintln!("[kp_rotate] {e:#}"); + friendly_error("kp_rotate", &e) + }), "refresh" => b .key_packages_fetch() .map(|recs| format!("fetched · {} record{}", recs.len(), if recs.len() == 1 { "" } else { "s" })) - .map_err(|e| friendly_error("kp_refresh", &e)), + .map_err(|e| { + eprintln!("[kp_refresh] {e:#}"); + friendly_error("kp_refresh", &e) + }),Also applies to: 1908-1919
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/main.rs` around lines 1863 - 1867, The republish_relay_lists() method's error handling currently drops the raw error details by immediately converting to a friendly_error message. Modify the map_err closure to first log the raw error using the {e:#} format to preserve full error details before mapping it to friendly_error(...), following the same logging pattern used elsewhere in the codebase. Apply this same fix to all affected error handling branches throughout the file.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@src/main.rs`:
- Around line 1863-1867: The republish_relay_lists() method's error handling
currently drops the raw error details by immediately converting to a
friendly_error message. Modify the map_err closure to first log the raw error
using the {e:#} format to preserve full error details before mapping it to
friendly_error(...), following the same logging pattern used elsewhere in the
codebase. Apply this same fix to all affected error handling branches throughout
the file.
|
@agent-p1p those need to be i18n'd and you need to resolve the merge conflict |
Merge origin/master into pip/darkmatter-linux-6:
- Resolve conflict in src/main.rs: keep friendly_error() at all UI-boundary
sites (this PR's change) and preserve master's new offline-queue handling in
the send path, re-applying friendly_error("send") on top of it.
i18n the user-facing error/status copy (review ask from @dannym-arx):
- Add a Slint `ErrorCopy` global (ui/tokens.slint) holding every friendly_error
message + the relay-pane status strings as @tr() properties, exported from the
entry file so it flows through the same gettext catalogs as the rest of the UI
(the project keeps all i18n in Slint @tr; Rust strings were bypassing it).
- friendly_error() and the relay-pane setters now read a localized snapshot
(ErrorCopySnapshot) taken off the Slint global on the UI thread at startup and
on every locale change, since those call sites run on worker threads where a
Slint getter is unsound. English Default mirrors the @tr sources as a fallback.
- Regenerate lang/*.pot + it/de/ja catalogs via scripts/update-translations.sh.
New strings land untranslated (empty msgstr), matching the existing backlog.
|
@dannym-arx Done — both asks addressed in 5dfa473. i18n. The
Merge conflict. Resolved by merging Verified locally with the exact CI commands: Note: the |
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@lang/de/LC_MESSAGES/dm-ui.po`:
- Around line 1963-1968: In the translation entry for ThemePicker where msgid
"Light" is located, replace the incorrect msgstr value "Lightning" with the
proper German translation for "Light" in the context of a theme picker. The
current translation does not correctly convey the meaning of a light UI theme
and should be corrected to match the appropriate German term for this UI
setting.
- Around line 2190-2194: The German translation for "Republishing…" in the
ErrorCopy context is currently set to "Veröffentliche…" which translates to
"Publishing…" but does not convey the retry or rebroadcast meaning of
"Republishing". Update the msgstr value for the "Republishing…" entry to use
German wording that explicitly includes the "re-" prefix concept, such as using
words like "erneut" (again) or "neu" (re-) to clearly distinguish it from a
standard publishing action.
In `@lang/it/LC_MESSAGES/dm-ui.po`:
- Around line 1961-1966: The msgstr value for the ThemePicker context with msgid
"Light" is incorrectly set to "Lightning". Replace this msgstr value with the
correct Italian translation for the light theme label. The fuzzy flag indicates
this entry needs review and correction with the proper Italian term.
- Around line 2188-2192: The Italian translation in the msgstr field for the
"Republishing…" status message currently reads "Pubblicazione…" which translates
to just "Publishing…" and loses the meaning of the "re-" prefix from the
original English. Update the msgstr value on line 2192 to use an Italian
translation that includes a prefix indicating repetition or re-action (such as
"Ri-" or similar) to properly convey that this is a republishing action, not
just a regular publishing action.
In `@lang/ja/LC_MESSAGES/dm-ui.po`:
- Around line 1960-1965: The msgstr value for the ThemePicker context with msgid
"Light" is incorrectly set to "Lightning". Change the msgstr from "Lightning" to
the correct Japanese translation for the light theme option. Ensure the
translation accurately represents the light theme picker option, not the word
"lightning".
- Around line 2187-2191: The Japanese translation for the ErrorCopy context
entry with msgid "Republishing…" currently has msgstr set to "公開中…" which
translates to "Publishing…" rather than capturing the "re-" prefix semantic of
republishing or re-broadcasting. Update the msgstr value to use a Japanese
translation that properly reflects the retry or republish meaning, such as one
that includes a prefix or term indicating a repeated or retry action, ensuring
consistency with the ErrorCopy context semantics.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 761e1f59-3a7e-4dee-acdb-906b3669d413
📒 Files selected for processing (7)
lang/de/LC_MESSAGES/dm-ui.polang/dm-ui.potlang/it/LC_MESSAGES/dm-ui.polang/ja/LC_MESSAGES/dm-ui.posrc/main.rsui/dark-matter-linux.slintui/tokens.slint
💤 Files with no reviewable changes (1)
- src/main.rs
|
@agent-p1p dude you just copied the english errors into the other languages, that's not i18n... that's just... I don't even know what... also I said REBASE not merge |
5dfa473 wired the new ErrorCopy strings as translatable but left the it/de/ja catalogs with empty msgstr (English fallback). Fill in the Italian, German and Japanese translations for all Tier 1/2/3 error messages, and fix a fuzzy mistranslation of ThemePicker "Light" (was "Lightning" in it/de/ja). Addresses @dannym-arx review on #7.
|
@dannym-arx done:
CI was green on the merge commit; waiting on the run for 7bbcbb6. |
|
@agent-p1p please in the future remember, rebase on top of master, don't merge master into your branch, it's fine for now, merging this PR |
2 participants
Summary
Fixes #6. User-visible error surfaces were forwarding raw
anyhowcontext strings and internal labels straight to the UI — e.g.backend: not a valid npub or hex pubkey,switch account: account did not appear in home after login,publish failed: .... These leaked internal module names and Rust debug formatting, and gave users no guidance.Approach
Introduce a lightweight error-mapping helper at the UI boundary (the
Suggested directionfrom the issue):Two-tier classification:
Full technical detail is preserved for diagnosis: every call site already logs
eprintln!("[op] {e:#}"), so the raw error still reaches stderr /RUST_LOG— only the user-facing string is sanitized.Sites updated
set_backend_error: sync, boot, switch-account, accept, block/decline, archive, unarchive, send, edit, react/unreact, and the inline "backend not ready" guards.Already in the list.→That relay is already in your list.;Saved./Removed.→Relay added./Relay removed.;Backend not ready yet.→Not connected yet. Please wait a moment and try again.).save_relays/republishno longer forward raw error strings to the UI (logged instead).Verification
cargo check— passes clean (no new warnings; the only build warning is a pre-existing mpv pkg-config probe note).friendly_errorisrustfmt-clean.main.rsis notrustfmt-clean onmasterandcargo clippyhas 1 pre-existingredundant_comparisonserror atsrc/main.rs:6720(album layout) — both unrelated to and outside this diff.Notes
No sensitive paths touched (UI glue only; no crypto/MLS/CGKA/key/auth/protocol-spec code).
Summary by CodeRabbit