add chat message reactions#19
Conversation
|
Important Review skippedAuto incremental reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the WalkthroughThis update introduces a comprehensive set of new CLI commands for managing message reactions within chat rooms, including adding, removing, and subscribing to reactions. The README and help documentation are updated accordingly. Dependency versions are bumped, and minor improvements are made to typing and reaction-related logic for consistency and clarity. Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant CLI
participant ChatClient
participant AblyRealtime
participant Room
User->>CLI: ably rooms messages reactions add ROOMID MSGSERIAL REACTION
CLI->>AblyRealtime: Initialize connection
CLI->>ChatClient: Initialize chat client
ChatClient->>Room: Get room handle
CLI->>Room: Attach and subscribe to status
CLI->>Room: Add reaction to message
Room-->>CLI: Reaction add confirmation
CLI->>User: Output success or error
CLI->>Room: Release room
CLI->>AblyRealtime: Close connection
sequenceDiagram
participant User
participant CLI
participant ChatClient
participant AblyRealtime
participant Room
User->>CLI: ably rooms messages reactions subscribe ROOMID [--raw]
CLI->>AblyRealtime: Initialize connection
CLI->>ChatClient: Initialize chat client
ChatClient->>Room: Get room handle (with raw option if set)
CLI->>Room: Attach and subscribe to status
CLI->>Room: Subscribe to reactions (raw or summary)
Room-->>CLI: Emit reaction events
CLI->>User: Display reaction updates
User->>CLI: Ctrl+C (SIGINT)
CLI->>Room: Unsubscribe and release
CLI->>AblyRealtime: Close connection
Suggested reviewers
Poem
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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🔭 Outside diff range comments (1)
src/commands/rooms/typing/start.ts (1)
342-357: 🛠️ Refactor suggestion
process.exit()prevents Ocliffinally()from runningCalling
process.exit(0)short-circuits Node’s event-loop and may skip thefinallyhook you carefully implemented for clean-up. Allow the promise to resolve naturally and let Oclif invokefinally().- resolve(); - process.exit(0); // Explicitly exit to ensure process terminates + resolve(); // Oclif will exit once run() settles and finally() completes
🧹 Nitpick comments (15)
src/commands/rooms/messages/index.ts (2)
11-11: Add matching examples forremoveandsubscribefor parityNow that the
reactionscommand-group offers three sub-commands (add,remove,subscribe), consider extending the examples list so that users discover the whole surface area from--helpoutput.'$ ably rooms messages reactions add my-room "message-id" "👍"', + '$ ably rooms messages reactions remove my-room "message-id" "👍"', + "$ ably rooms messages reactions subscribe my-room",
26-28: Minor ordering / grouping nitAll other help lines are sorted alphabetically by sub-command (
get,send,subscribe). Movingreactionsbelowsubscribekeeps the list alphabetised and easier to scan.src/commands/rooms/messages/reactions/index.ts (2)
1-12: Expose global flags so users don’t lose functionalityOther top-level commands inherit the shared
globalFlags(API key, JSON output, etc.). Adding the same here will keep UX consistent:import { Command } from "@oclif/core"; export default class MessagesReactionsIndex extends Command { + static override flags = { + ...require("../../../chat-base-command.js").ChatBaseCommand.globalFlags, + };
13-25: Let oclif generate sub-command help automaticallyInstead of maintaining a hard-coded list that can drift, you can rely on oclif’s built-in help by throwing
new Help({ config: this.config }).showHelp(this, []);or simply doing nothing and allowingoclifto fall back to its default behaviour.This reduces maintenance and eliminates duplication with the examples array.
src/commands/rooms/typing/start.ts (1)
142-151: Rename log tag to matchkeystroke()semanticsThe surrounding log messages still refer to “start typing”, but we now use
keystroke(). Consider renaming the tag ("keystrokeAttempt","keystroked") so logs remain semantically accurate.README.md (5)
3513-3516: Expand parent command examples for full reactions lifecycle
You’ve added thereactions addexample under theably rooms messagesexamples. For consistency and to demonstrate the full lifecycle in the parent group, consider includingremoveandsubscribeexamples here as well.Example diff:
## EXAMPLES $ ably rooms messages send my-room "Hello world!" $ ably rooms messages subscribe my-room $ ably rooms messages get my-room - $ ably rooms messages reactions add my-room "message-id" "👍" + $ ably rooms messages reactions add my-room "message-id" "👍" + $ ably rooms messages reactions remove my-room "message-id" "👍" + $ ably rooms messages reactions subscribe my-room
3571-3577: Specify language for fenced code blocks
The usage block forably rooms messages reactionsis fenced without a language spec. For improved markdown linting and syntax highlighting, add a language (e.g.,shorsh-session).- ``` + ```sh USAGE $ ably rooms messages reactions<details> <summary>🧰 Tools</summary> <details> <summary>🪛 markdownlint-cli2 (0.17.2)</summary> 3571-3571: Fenced code blocks should have a language specified null (MD040, fenced-code-language) </details> </details> --- `3592-3600`: **Add language spec to `add` command usage block** The `reactions add` usage block is missing a language identifier on its fenced code. Use `sh` for consistency with other CLI examples. ```diff - ``` + ```sh USAGE $ ably rooms messages reactions add ROOMID MESSAGESERIAL REACTION [--access-token <value>] ...<details> <summary>🧰 Tools</summary> <details> <summary>🪛 markdownlint-cli2 (0.17.2)</summary> 3592-3592: Fenced code blocks should have a language specified null (MD040, fenced-code-language) </details> </details> --- `3643-3652`: **Add language spec to `remove` command usage block** The `reactions remove` code block should also include a language spec. ```diff - ``` + ```sh USAGE $ ably rooms messages reactions remove ROOMID MESSAGESERIAL REACTION [--access-token <value>] ...<details> <summary>🧰 Tools</summary> <details> <summary>🪛 markdownlint-cli2 (0.17.2)</summary> 3643-3643: Fenced code blocks should have a language specified null (MD040, fenced-code-language) </details> </details> --- `3683-3690`: **Add language spec to `subscribe` command usage block** Similarly, add a language to the `reactions subscribe` usage code fence for proper linting. ```diff - ``` + ```sh USAGE $ ably rooms messages reactions subscribe ROOMID [--access-token <value>] ...<details> <summary>🧰 Tools</summary> <details> <summary>🪛 markdownlint-cli2 (0.17.2)</summary> 3683-3683: Fenced code blocks should have a language specified null (MD040, fenced-code-language) </details> </details> </blockquote></details> <details> <summary>src/commands/rooms/messages/reactions/remove.ts (1)</summary><blockquote> `171-185`: **No validation for unknown message serial format** If `messageSerial` is malformed the SDK will throw an opaque error. Adding a quick format/emptiness check gives earlier feedback and a clearer message to the user. ```ts if (!/^[\w-]+:\d+$/.test(messageSerial)) { this.error(`Invalid messageSerial "${messageSerial}". Expected format "<id>:<index>".`); }src/commands/rooms/messages/reactions/add.ts (2)
241-256: Default reaction parameters can be simplifiedBuilding
reactionParamspiecemeal leads to three conditional blocks and repeated casts.
A flatter expression reduces complexity:-const reactionParams: { … } = { name: reaction }; - -if (flags.type) { - reactionParams.type = flags.type as MessageReactionType; -} -if (flags.type === "multiple" && flags.count) { - reactionParams.count = flags.count; -} -if (this.metadataObj) { - reactionParams.metadata = this.metadataObj; -} +const reactionParams = { + name: reaction, + ...(flags.type && { type: flags.type as MessageReactionType }), + ...(flags.type === "multiple" && flags.count && { count: flags.count }), + ...(this.metadataObj && { metadata: this.metadataObj }), +};
278-288:resultDataomitsmetadatafor empty objects
...(this.metadataObj && { metadata: this.metadataObj })filters out{}.
If callers intentionally pass an empty object it disappears from the output.
UsemetadataObj !== nullto preserve intent.src/commands/rooms/messages/reactions/subscribe.ts (2)
224-258: Guard console output against undefined fields
event.reaction.clientIdand.namemay beundefined; interpolating them directly yields"undefined"strings.- `${chalk.blue(event.reaction.clientId || "Unknown")}` + `${chalk.blue(event.reaction.clientId ?? "anonymous")}`Apply similar guards for
name&messageSerialto keep output tidy.
48-75: Duplicated cleanup logic across commands
finallyblocks in all three reaction commands repeat nearly identical resource-release code. Extracting a helper (e.g.,withCleanup(async () => { … })) or adding adispose()method toChatBaseCommandwould cut ~40 lines per file and reduce future maintenance.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
⛔ Files ignored due to path filters (1)
pnpm-lock.yamlis excluded by!**/pnpm-lock.yaml
📒 Files selected for processing (12)
README.md(3 hunks)package.json(1 hunks)src/commands/rooms/messages/index.ts(2 hunks)src/commands/rooms/messages/reactions/add.ts(1 hunks)src/commands/rooms/messages/reactions/index.ts(1 hunks)src/commands/rooms/messages/reactions/remove.ts(1 hunks)src/commands/rooms/messages/reactions/subscribe.ts(1 hunks)src/commands/rooms/reactions/send.ts(1 hunks)src/commands/rooms/reactions/subscribe.ts(1 hunks)src/commands/rooms/typing/start.ts(7 hunks)src/commands/rooms/typing/subscribe.ts(2 hunks)test/unit/services/config-manager.test.ts(1 hunks)
🧰 Additional context used
🪛 markdownlint-cli2 (0.17.2)
README.md
3571-3571: Fenced code blocks should have a language specified
null
(MD040, fenced-code-language)
3592-3592: Fenced code blocks should have a language specified
null
(MD040, fenced-code-language)
3643-3643: Fenced code blocks should have a language specified
null
(MD040, fenced-code-language)
3683-3683: Fenced code blocks should have a language specified
null
(MD040, fenced-code-language)
⏰ Context from checks skipped due to timeout of 90000ms (4)
- GitHub Check: e2e-tests
- GitHub Check: test
- GitHub Check: test
- GitHub Check: e2e-tests
🔇 Additional comments (11)
src/commands/rooms/reactions/send.ts (1)
133-133: Update aligned with API simplification.The removal of the explicit empty
reactionsoption from the room retrieval call reflects the dependency upgrade to@ably/chatv0.7.0, which no longer requires these empty options.test/unit/services/config-manager.test.ts (1)
11-12: Import statement refactoring improves clarity.Separating the named import
expectfrom the namespace importchaifollows better import organization practices while maintaining the same functionality.package.json (1)
70-70:Details
❓ Verification inconclusive
Dependency version upgrade to support new chat reaction features.
Upgrading
@ably/chatfrom v0.5.1 to v0.7.0 is necessary to support the new chat message reaction functionality being implemented in this PR.
🏁 Script executed:
#!/bin/bash # Check for breaking changes or notable features in the new version echo "Checking for new features in @ably/chat v0.7.0 compared to v0.5.1" gh api graphql -f query=' { repository(owner: "ably", name: "ably-chat-js") { releases(first: 10) { nodes { name tagName description } } } }' | jq '.data.repository.releases.nodes[] | select(.tagName == "v0.7.0" or .tagName == "v0.6.0")'Length of output: 428
🏁 Script executed:
#!/bin/bash set -e # List published releases for @ably/chat echo "Releases (via GitHub API):" gh api repos/ably/ably-chat-js/releases --jq '.[].tag_name' # List all tags in the repo echo "" echo "Tags (via GitHub API):" gh api repos/ably/ably-chat-js/tags --jq '.[].name' # Locate the changelog file in the repo changelog=$(git ls-files | grep -i CHANGELOG.md || true) if [ -n "$changelog" ]; then echo "" echo "Found changelog at: $changelog" echo "" echo "Changelog entries around v0.7.0:" # Print lines from v0.7.0 section to the next version heading sed -n '/## *\[?0\.7\.0\]*/,/## *\[?[0-9]\+\.[0-9]\+\.[0-9]\]*/p' "$changelog" || echo "No entry for v0.7.0 found" else echo "" echo "No CHANGELOG.md file found in repository" fiLength of output: 717
🏁 Script executed:
#!/bin/bash set -e echo "Fetching release notes for @ably/chat v0.7.0 from GitHub:" gh api repos/ably/ably-chat-js/releases/tags/0.7.0 --jq '.body' echo "" echo "Filtering for mentions of 'reaction' in v0.7.0 release notes:" gh api repos/ably/ably-chat-js/releases/tags/0.7.0 --jq '.body' | grep -i reaction -C2 || echo "No 'reaction' mentions found"Length of output: 1738
Upgrade @ably/chat to v0.7.0 for experimental message reactions
- package.json (line 70):
- "@ably/chat": "^0.5.1", + "@ably/chat": "^0.7.0",- v0.7.0 adds experimental message reactions (PR #473), which your new feature relies on.
- Note: v0.7.0 also includes minor breaking changes:
• Single-channel architecture
• Room options restructuring
• Typing API changes (typing.keystroke()replacestyping.start())
•onDiscontinuitynow scoped at the room level- Please verify that all affected code paths have been updated to accommodate these changes.
src/commands/rooms/reactions/subscribe.ts (1)
112-112: Update aligned with API simplification.The removal of explicit options from
this.chatClient.rooms.get(roomId)is consistent with similar changes in other files and reflects the updated API in@ably/chatv0.7.0.src/commands/rooms/typing/subscribe.ts (2)
101-101: Update aligned with API simplification.Removal of the explicit typing options from the room retrieval call aligns with similar changes across the codebase and the updated API patterns in
@ably/chatv0.7.0.
167-167: Type annotation simplification.Removing the explicit type annotation for the typing event parameter allows the TypeScript compiler to infer the type automatically, which is now possible with the updated library version.
src/commands/rooms/typing/start.ts (2)
8-16: Comments and constants contradict each otherThe block comment states heartbeats are throttled to once every 10 s, yet the
KEYSTROKE_INTERVALis set to 450 ms. Either the comment is stale or we are sending a redundant 20× flood of keystrokes.Update the documentation or increase the interval (e.g. 9 000 ms) to match the library’s throttle period and save unnecessary network traffic.
165-177: 🛠️ Refactor suggestionHigh-frequency
setIntervalcan leak rejected promisesAt 450 ms we create ~133 Promises per minute; unhandled rejections could accumulate if the connection drops. Consider:
• Moving the interval to ≥9 000 ms (see first comment)
• Using an async IIFE inside the timer withawait+ try/catch to avoid silent rejections.- this.typingIntervalId = setInterval(() => { - room.typing.keystroke().catch((error: Error) => { … }); - }, KEYSTROKE_INTERVAL); + this.typingIntervalId = setInterval(async () => { + try { + await room.typing.keystroke(); + } catch (error) { + this.logCliEvent(flags, "typing", "keystrokeError", …); + } + }, KEYSTROKE_INTERVAL);Likely an incorrect or invalid review comment.
README.md (2)
153-156: New reaction commands listed correctly in commands overview
Therooms messages reactionssubcommands are inserted in the commands list following the existing pattern (get, reactions group, send, subscribe). The indentation and link anchors follow the established style.
3567-3570: Document new message-reactions command group
The## \ably rooms messages reactions`` section correctly introduces the new command group. The heading level and style match existing sections.src/commands/rooms/messages/reactions/remove.ts (1)
183-185: Confirmdeleteis the correct API callIn SDK versions prior to
@ably/chat@0.7.0the method was namedremove, notdelete.
If users install the CLI with an older lock-file they will face a runtime failure (room.messages.reactions.delete is not a function).Ask npm/yarn to respect the
^0.7.0minimum or guard the call:if (typeof room.messages.reactions.delete !== "function") { this.error("Reaction deletion requires @ably/chat >= 0.7.0"); }
| this.ablyClient.connection.on( | ||
| (stateChange: Ably.ConnectionStateChange) => { | ||
| this.logCliEvent( | ||
| flags, | ||
| "connection", | ||
| stateChange.current, | ||
| `Realtime connection state changed to ${stateChange.current}`, | ||
| { reason: stateChange.reason }, | ||
| ); | ||
| }, | ||
| ); | ||
|
|
There was a problem hiding this comment.
🛠️ Refactor suggestion
Listener should be detached to avoid memory leaks
You attach a permanent connection.on listener to the underlying ablyClient but never remove it.
After repeated CLI invocations in the same long–lived process (tests, REPL, or in-process usage), listeners will accumulate and trigger the Node.js MaxListenersExceededWarning.
this.ablyClient.connection.on(
(stateChange: Ably.ConnectionStateChange) => {
/* … */
},
);
+// ⬆ keep the unsubscribe handle and call `off` once finishedConsider capturing the returned listener function (or call once) and explicitly remove it in your cleanup / finally block, similar to how you handle room status subscriptions.
Committable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In src/commands/rooms/messages/reactions/remove.ts around lines 88 to 99, the
connection.on listener on ablyClient is attached permanently without removal,
causing potential memory leaks and MaxListenersExceededWarning. Modify the code
to capture the listener function returned by connection.on or use
connection.once instead, and ensure this listener is explicitly removed or
cleaned up in the appropriate cleanup or finally block, similar to how room
status subscriptions are handled.
| // Validate count for Multiple type | ||
| if (flags.type === "multiple" && flags.count !== undefined && flags.count <= 0) { | ||
| const errorMsg = "Count must be a positive integer for Multiple type reactions"; | ||
| this.logCliEvent(flags, "reaction", "invalidCount", errorMsg, { | ||
| error: errorMsg, | ||
| count: flags.count, | ||
| }); | ||
| if (this.shouldOutputJson(flags)) { | ||
| this.log( | ||
| this.formatJsonOutput( | ||
| { error: errorMsg, roomId, success: false }, | ||
| flags, | ||
| ), | ||
| ); | ||
| } else { | ||
| this.error(errorMsg); | ||
| } | ||
| return; | ||
| } |
There was a problem hiding this comment.
Enforce count when type === "multiple"
The current validation only rejects non-positive counts but allows a missing count, resulting in undefined being sent which the SDK rejects.
-if (flags.type === "multiple" && flags.count !== undefined && flags.count <= 0) {
+if (flags.type === "multiple" && (flags.count === undefined || flags.count <= 0)) {📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| // Validate count for Multiple type | |
| if (flags.type === "multiple" && flags.count !== undefined && flags.count <= 0) { | |
| const errorMsg = "Count must be a positive integer for Multiple type reactions"; | |
| this.logCliEvent(flags, "reaction", "invalidCount", errorMsg, { | |
| error: errorMsg, | |
| count: flags.count, | |
| }); | |
| if (this.shouldOutputJson(flags)) { | |
| this.log( | |
| this.formatJsonOutput( | |
| { error: errorMsg, roomId, success: false }, | |
| flags, | |
| ), | |
| ); | |
| } else { | |
| this.error(errorMsg); | |
| } | |
| return; | |
| } | |
| // Validate count for Multiple type | |
| if (flags.type === "multiple" && (flags.count === undefined || flags.count <= 0)) { | |
| const errorMsg = "Count must be a positive integer for Multiple type reactions"; | |
| this.logCliEvent(flags, "reaction", "invalidCount", errorMsg, { | |
| error: errorMsg, | |
| count: flags.count, | |
| }); | |
| if (this.shouldOutputJson(flags)) { | |
| this.log( | |
| this.formatJsonOutput( | |
| { error: errorMsg, roomId, success: false }, | |
| flags, | |
| ), | |
| ); | |
| } else { | |
| this.error(errorMsg); | |
| } | |
| return; | |
| } |
🤖 Prompt for AI Agents
In src/commands/rooms/messages/reactions/add.ts around lines 122 to 140, the
validation for the "multiple" type reactions only rejects counts that are
non-positive but does not reject missing counts, allowing undefined to be sent.
Update the condition to also check if the count is undefined or not a positive
integer, and reject the input if count is missing or invalid by logging the
error and stopping execution as currently done.
| try { | ||
| // Parse metadata if provided | ||
| if (flags.metadata) { | ||
| try { | ||
| this.metadataObj = JSON.parse(flags.metadata); | ||
| this.logCliEvent( | ||
| flags, | ||
| "reaction", | ||
| "metadataParsed", | ||
| "Metadata parsed successfully", | ||
| { metadata: this.metadataObj }, | ||
| ); | ||
| } catch (error) { | ||
| const errorMsg = `Invalid metadata JSON: ${error instanceof Error ? error.message : String(error)}`; | ||
| this.logCliEvent(flags, "reaction", "metadataParseError", errorMsg, { | ||
| error: errorMsg, | ||
| roomId, | ||
| }); | ||
| if (this.shouldOutputJson(flags)) { | ||
| this.log( | ||
| this.formatJsonOutput( | ||
| { error: errorMsg, roomId, success: false }, | ||
| flags, | ||
| ), | ||
| ); | ||
| } else { | ||
| this.error(errorMsg); | ||
| } | ||
| return; | ||
| } | ||
| } | ||
|
|
There was a problem hiding this comment.
Parsed metadata never reaches the server
metadataObj is parsed and logged but later not attached to the reaction sent to Ably; users will believe the data was stored although it’s silently dropped.
const reactionParams: {
name: string;
type?: MessageReactionType;
count?: number;
+ metadata?: Record<string, unknown>;
} = {
name: reaction,
};
@@
if (flags.type === "multiple" && flags.count) {
reactionParams.count = flags.count;
}
+if (this.metadataObj) {
+ reactionParams.metadata = this.metadataObj;
+}📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| try { | |
| // Parse metadata if provided | |
| if (flags.metadata) { | |
| try { | |
| this.metadataObj = JSON.parse(flags.metadata); | |
| this.logCliEvent( | |
| flags, | |
| "reaction", | |
| "metadataParsed", | |
| "Metadata parsed successfully", | |
| { metadata: this.metadataObj }, | |
| ); | |
| } catch (error) { | |
| const errorMsg = `Invalid metadata JSON: ${error instanceof Error ? error.message : String(error)}`; | |
| this.logCliEvent(flags, "reaction", "metadataParseError", errorMsg, { | |
| error: errorMsg, | |
| roomId, | |
| }); | |
| if (this.shouldOutputJson(flags)) { | |
| this.log( | |
| this.formatJsonOutput( | |
| { error: errorMsg, roomId, success: false }, | |
| flags, | |
| ), | |
| ); | |
| } else { | |
| this.error(errorMsg); | |
| } | |
| return; | |
| } | |
| } | |
| const reactionParams: { | |
| name: string; | |
| type?: MessageReactionType; | |
| count?: number; | |
| metadata?: Record<string, unknown>; | |
| } = { | |
| name: reaction, | |
| }; | |
| if (flags.type === "multiple" && flags.count) { | |
| reactionParams.count = flags.count; | |
| } | |
| if (this.metadataObj) { | |
| reactionParams.metadata = this.metadataObj; | |
| } |
🤖 Prompt for AI Agents
In src/commands/rooms/messages/reactions/add.ts between lines 90 and 121, the
parsed metadata is stored in this.metadataObj and logged but never included in
the reaction payload sent to the server. To fix this, ensure that the parsed
metadata is attached to the reaction object or request payload before sending it
to Ably, so the metadata is actually transmitted and stored as intended.
| process.exit(0); | ||
| }; |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Avoid forcing process.exit(0) inside an oclif command
oclif handles process termination; calling process.exit bypasses its error handling, can cut off pending I/O (e.g. flushed logs) and complicates test harnesses.
Simply resolving the promise is sufficient:
- resolve();
-
- process.exit(0);
+ resolve(); // let oclif regain control📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| process.exit(0); | |
| }; | |
| resolve(); // let oclif regain control | |
| }; |
🤖 Prompt for AI Agents
In src/commands/rooms/messages/reactions/subscribe.ts around lines 522 to 523,
remove the call to process.exit(0) inside the oclif command. Instead of forcing
process termination, allow the command to complete naturally by resolving the
promise or returning from the function, so that oclif can handle process exit
and cleanup properly.
splindsay-92
force-pushed
the
upgrade-chat-js
branch
2 times, most recently
from
f4814a9 to
8f810ff
Compare
|
Tests are failing and this needs a rebase @vladvelici given there were build issues on the commit you based off. |
vladvelici
force-pushed
the
add-chat-message-reactions
branch
from
0dcef6f to
d630cf8
Compare
vladvelici
force-pushed
the
add-chat-message-reactions
branch
from
d630cf8 to
883e807
Compare
|
Thanks @vladvelici |
2 participants
Summary by CodeRabbit
New Features
Documentation
Chores
Style