Buzz QoL bundle for v2026.7.30 - #1
Draft
OthmaneN wants to merge 17 commits into
Draft
Conversation
Signed-off-by: bbamnesia <68875027+bbamnesia@users.noreply.github.com>
…e relay
send() forwards metadata["thread_id"] as --reply-to, but this adapter
never sets thread metadata on inbound — every inbound event logs
reply_to_id=None. That value therefore arrives from generic gateway
machinery and can name an event this relay has never seen, so buzz-cli
rejects the entire send:
error: parent event <id> not found (exit 4)
The message is then lost, because the plain-text fallback in
gateway/platforms/base.py re-sends with the same reply_to and metadata
and varies only the text — it is a fallback for formatting problems, not
for argument-level rejections. Two replies were lost this way within ten
minutes of ordinary use against a hosted relay, in a channel where the
triggering messages were plain top-level posts.
Threading is a presentation detail; losing the message over it is not.
On that specific error, strip --reply-to and retry once. Unrelated
failures are untouched and still surface as before.
The deeper cause is that the adapter does not model Buzz threads at all —
it only ever reports chat_type "group" or "dm", never "thread", so
gateway thread state and Buzz thread state are never reconciled. That is
a design change and belongs in its own issue; this only stops the
mismatch from destroying messages.
Review caught that the first commit only covered text send(). Native local-file uploads and _standalone_send build the same --reply-to argument and return the same parent-not-found failure, so cron delivery (deliver=buzz) and image replies could still be lost outright. Extract the bounded retry into _send_retrying_unthreaded() and use it from all three call sites. The helper takes a runner callable returning (code, stdout, stderr), which lets the adapter's _run_cli and the module-level _exec_buzz share one implementation without _standalone_send needing an adapter instance. Behaviour is unchanged: still a single retry, still only on parent-not-found, every other failure surfaces as before. Tests cover the two newly-guarded paths, including that the --file upload survives the retry.
The Buzz adapter always passes --reply-to with the triggering message's event id, causing every outbound reply to chain into the source message's Nostr thread. In multi-turn conversations this produces deeply nested threads (thread-inside-thread-inside-thread) that hurt readability. Add a config option (default: true, backwards-compatible) that controls whether --reply-to is emitted on send/send_image/standalone delivery. Set to false to post flat, top-level replies instead. Config (config.yaml): gateway.platforms.buzz.extra.reply_in_thread: false Or env override: BUZZ_REPLY_IN_THREAD=false Applies to three send paths: send(), send_image(), and _standalone_send() (cron delivery).
OthmaneN
pushed a commit
that referenced
this pull request
Aug 17, 2026
Addresses both review findings on the remote-gateway download PR: 1. Unbounded buffering (finding #1). fetchBuffer / fetchBufferViaOauthSession accumulated the entire response (then copied it again via Buffer.concat) before saveGatewayFile even opened the save dialog, so a large gateway file could exhaust the native process. Both auth paths now stream: once response headers arrive the connect timeout is cleared, the filename is derived, the save dialog is shown, and the body is piped to the chosen destination with backpressure. A read/write error tears down the stream and unlinks the partial file. The byte-moving, data-URL decoding, and filename/path helpers are extracted into gateway-file-download.ts so they're unit-testable without Electron. 2. No fallback for older gateways (finding NousResearch#2). saveGatewayFile required the new /api/fs/download route. Desktop and the remote gateway update independently, so a gateway predating this PR 404s. Added a 404-only compatibility fallback to the existing capped /api/fs/read-data-url route (bounded, so it only serves smaller files — enough to keep older backends working). Tests: gateway-file-download.test.ts covers streaming, backpressure, error-cleanup (unlink on write/response error), data-URL decoding, filename derivation (incl. traversal reduction), and 404 detection; gateway-file-download-transport.test.ts asserts both transports stream (no whole-body Buffer.concat) and that the 404 fallback is wired. Both registered in the desktop platform test list. Server-side /api/fs/download tests (streaming + sensitive-file reject) already pass. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
OthmaneN
pushed a commit
that referenced
this pull request
Aug 17, 2026
…-renders (NousResearch#81726) The scoped find walker wraps transcript text nodes in <mark> elements that React does not own. Assistant responses stream through markdown-text.tsx, which rebuilds the markdown DOM on every delta, and a new message is appended whenever the assistant answers — so a re-render of a changed region detaches the marks we inserted, dropping the user's highlights while the bar stays open. Watch the captured scope with a MutationObserver and re-wrap only when an unmarked occurrence of the active query actually reappears. The observer is gated behind a re-entrancy flag while the walker is mutating, coalesced to one re-apply per microtask, torn down when the bar closes or the query clears, and restores the active ordinal so a mid-stream re-render doesn't reset the user's place to match #1. An append that adds no matching text is a no-op; re-wrapping only fires when highlights genuinely went stale. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
8 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Base
Built from Hermes release
v2026.7.30(commitcc4cab469cd2da24ddedeb68d2f06f9ec3226875).Included
Overlapping thread-root logic was deduplicated onto NousResearch#74516. The canonical root now scopes Hermes session memory and anchors outbound text/image replies, avoiding nested Buzz subthreads.
reply_in_threadis YAML/config only; no extra environment-variable path.Verification