feat(telegram): add reaction indicator for message processing#433
Merged
feat(telegram): add reaction indicator for message processing#433
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (3)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughPropagates Changes
Sequence DiagramsequenceDiagram
actor User as User/Telegram
participant TG as Telegram API
participant Ch as TelegramChannel
participant Agent as Agent Loop
User->>TG: Send message
TG->>Ch: Deliver inbound message (includes telegram_message_id)
Ch->>Ch: Optionally set 👀 reaction (if enabled)
Ch->>Agent: Publish InboundMessage (with telegram_message_id metadata)
Agent->>Agent: Process and propagate telegram_message_id into OutboundMessage
Agent->>Ch: Send outbound response
Ch->>TG: Post reply (use ReplyParameters: reply_to or telegram_message_id)
Ch->>TG: Optionally set ✅ reaction on original message
TG->>User: Deliver threaded reply and reactions
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@src/agent/loop.rs`:
- Around line 240-244: The outbound metadata propagation currently copies only
"telegram_thread_id" in the direct tool publish paths, so ensure you also copy
"telegram_message_id" from inbound to outbound there; specifically, locate the
places that insert "telegram_thread_id" into outbound.metadata (the same scope
where inbound.metadata.get("telegram_thread_id") is read) and add the same
pattern used at the final/error/timeout path—check
inbound.metadata.get("telegram_message_id") and, if Some(mid), call
outbound.metadata.insert("telegram_message_id".to_string(), mid.clone()) so that
for_user tool messages preserve the Telegram message id for reply-thread
fallback and reaction updates.
In `@src/channels/telegram.rs`:
- Around line 1222-1232: The code sets the completion reaction (✅) immediately
after reading telegram_message_id, which can mark a message as processed even if
the outbound reply fails; move the bot.set_message_reaction call to run only
after the outgoing send succeeds—i.e., after the successful await of the send
function (the call around send_message at lines near where send_message is
awaited), so wrap or relocate the logic that checks
msg.metadata.get("telegram_message_id") and calls
bot.set_message_reaction(ChatId(chat_id), MessageId(mid)).reaction(...) to
execute only when the send_message future returns Ok; keep the same parsing of
mid_str.parse::<i32>() and error handling but perform it post-send to ensure the
✅ reaction is only applied on successful send.
- Around line 781-790: The reaction is being set unconditionally with
bot.set_message_reaction(...) (the 👀 emoji) before the early-return branches
(/model and /persona short-circuits and the if let Some(text) gate), which
leaves many inbound messages stuck at 👀; move the reaction setup so it only
runs when the message will reach the completion path (i.e., after the /model and
/persona short-circuits and after the if let Some(text) branch), or refactor by
calling the same completion helper used by the inline bot.send_message(...)
paths (the paths that currently add ✅ in send()) so both inline and async
completion flows reuse the same reaction logic instead of setting 👀
unconditionally.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 2173bad7-79c2-4a9c-bb56-431a48069768
📒 Files selected for processing (2)
src/agent/loop.rssrc/channels/telegram.rs
Comment on lines
+781
to
+790
| // React with 👀 to acknowledge receipt before processing. | ||
| { | ||
| use teloxide::types::ReactionType; | ||
| let _ = bot | ||
| .set_message_reaction(msg.chat.id, msg.id) | ||
| .reaction(vec![ReactionType::Emoji { | ||
| emoji: "\u{1F440}".to_string(), | ||
| }]) | ||
| .await; | ||
| } |
There was a problem hiding this comment.
Only add 👀 on paths that will hit the completion hook.
✅ is only set in send(), but this block runs before the /model and /persona short-circuits and before the if let Some(text) gate. Messages handled inline at Lines 853-930, 934-1014, and 1084-1089 — plus unsupported updates that never enter Lines 800-803 — will leave the inbound message stuck at 👀.
Either move the reaction setup below the early-return branches or reuse the same completion helper from those inline bot.send_message(...) paths.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/channels/telegram.rs` around lines 781 - 790, The reaction is being set
unconditionally with bot.set_message_reaction(...) (the 👀 emoji) before the
early-return branches (/model and /persona short-circuits and the if let
Some(text) gate), which leaves many inbound messages stuck at 👀; move the
reaction setup so it only runs when the message will reach the completion path
(i.e., after the /model and /persona short-circuits and after the if let
Some(text) branch), or refactor by calling the same completion helper used by
the inline bot.send_message(...) paths (the paths that currently add ✅ in
send()) so both inline and async completion flows reuse the same reaction logic
instead of setting 👀 unconditionally.
Comment on lines
+1222
to
+1232
| // Clear the processing reaction (👀) set when the message was received. | ||
| if let Some(mid_str) = msg.metadata.get("telegram_message_id") { | ||
| if let Ok(mid) = mid_str.parse::<i32>() { | ||
| let _ = bot | ||
| .set_message_reaction(ChatId(chat_id), MessageId(mid)) | ||
| .reaction(vec![ReactionType::Emoji { | ||
| emoji: "\u{2705}".to_string(), | ||
| }]) | ||
| .await; | ||
| } | ||
| } |
There was a problem hiding this comment.
Set ✅ only after Telegram accepts the outbound message.
This marks the inbound message as complete before Line 1262 actually sends anything. If send_message fails (for example, bad HTML or a transport error), users will still see ✅ even though no reply was delivered.
🔧 Move the completion reaction after a successful send
- // Clear the processing reaction (👀) set when the message was received.
- if let Some(mid_str) = msg.metadata.get("telegram_message_id") {
- if let Ok(mid) = mid_str.parse::<i32>() {
- let _ = bot
- .set_message_reaction(ChatId(chat_id), MessageId(mid))
- .reaction(vec![ReactionType::Emoji {
- emoji: "\u{2705}".to_string(),
- }])
- .await;
- }
- }
-
let rendered = render_telegram_html(&msg.content);
let mut req = bot
.send_message(ChatId(chat_id), rendered)
.parse_mode(ParseMode::Html);
@@
req.await
.map_err(|e| ZeptoError::Channel(format!("Failed to send Telegram message: {}", e)))?;
+
+ if let Some(mid_str) = msg.metadata.get("telegram_message_id") {
+ if let Ok(mid) = mid_str.parse::<i32>() {
+ let _ = bot
+ .set_message_reaction(ChatId(chat_id), MessageId(mid))
+ .reaction(vec![ReactionType::Emoji {
+ emoji: "\u{2705}".to_string(),
+ }])
+ .await;
+ }
+ }Also applies to: 1262-1263
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/channels/telegram.rs` around lines 1222 - 1232, The code sets the
completion reaction (✅) immediately after reading telegram_message_id, which can
mark a message as processed even if the outbound reply fails; move the
bot.set_message_reaction call to run only after the outgoing send succeeds—i.e.,
after the successful await of the send function (the call around send_message at
lines near where send_message is awaited), so wrap or relocate the logic that
checks msg.metadata.get("telegram_message_id") and calls
bot.set_message_reaction(ChatId(chat_id), MessageId(mid)).reaction(...) to
execute only when the send_message future returns Ok; keep the same parsing of
mid_str.parse::<i32>() and error handling but perform it post-send to ensure the
✅ reaction is only applied on successful send.
Owner
|
Thanks for this! #432 has been merged, so this needs a rebase. A few things to address before merge:
|
React with eyes emoji when a message is received to give instant visual feedback, then replace with a checkmark when the response is sent. Reactions are best-effort — errors are silently ignored since the bot may lack reaction permissions in some groups. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add `telegram.reactions` config flag (default: true) for opt-out - Move 👀 after /model and /persona early returns so commands don't get stuck with eyes reaction - Move ✅ after successful send_message so failed sends aren't marked as complete - Restore removed log line before bot initialization - Add debug logging on reaction failures instead of silent ignore - Propagate telegram_message_id in tool for_user outbound paths Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
stuartbowness
force-pushed
the
telegram-reaction-indicator-v2
branch
from
3692f96 to
430bdd8
Compare
Contributor
Author
|
All four items addressed:
CI is green (build, clippy, fmt, tests all pass). Ready for re-review. |
Owner
|
Merged! Great turnaround @stuartbowness — all four review items addressed cleanly. The config gate and debug logging are exactly right. Nice work! |
7 tasks
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
2 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.
Summary
Depends on #432. Replaces #430 (rebased to resolve conflicts from merged history).
Test plan
🤖 Generated with Claude Code
Summary by CodeRabbit
New Features
Chores