feat: add one-shot CLI mode#8
Merged
Merged
Conversation
|
Warning Rate limit exceeded
Your organization is not enrolled in usage-based pricing. Contact your admin to enable usage-based pricing to continue reviews beyond the rate limit, or try again in 4 minutes and 25 seconds. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (17)
📝 WalkthroughWalkthroughAdds a non-interactive one-shot CLI (clap) with global options, persistent TOML-backed device config and device-memory commands, model/label-based target resolution and timed BLE auto-connect, console entrypoints accepting a connected device, and refactors main/loader to route one-shot vs interactive flows. Changes
Sequence DiagramsequenceDiagram
participant User
participant CLI as "CLI Parser"
participant Config as "Config Manager"
participant BLE as "BLE Scanner"
participant Device as "IQOS Device"
participant Console as "Command Console"
User->>CLI: iqos --model <target> <command>
CLI->>Config: load config / resolve target
CLI->>BLE: start scan (timeout)
BLE->>CLI: discovered peripheral
CLI->>BLE: match peripheral (by model or address)
CLI->>Device: connect_and_discover()
Device->>CLI: read DeviceInfo (serial, model)
CLI->>Config: update default / save metadata
CLI->>Console: run one-shot command (connected device)
Console->>Device: send command
Device-->>Console: command response
Console-->>CLI: result
CLI->>User: print result, exit(code)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 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/cli.rs`:
- Around line 29-31: The Autostart enum variant (and the other five enum
variants that define an args: Vec<String> passthrough) are using
#[arg(value_name = "arg")] so clap treats hyphen-prefixed tokens as flags;
update each of those #[arg(...)] attributes to include allow_hyphen_values =
true and trailing_var_arg = true (e.g., #[arg(value_name = "arg",
allow_hyphen_values = true, trailing_var_arg = true)]) for Autostart and the
other variants that pass args to the registered command handler so
hyphen-prefixed values are accepted and all remaining tokens are collected.
In `@src/config.rs`:
- Around line 108-117: In config_file(), stop using XDG_CONFIG_HOME when it's
empty or a relative path: when std::env::var_os("XDG_CONFIG_HOME") returns a
value, convert it to a PathBuf and only use it if it's non-empty and
path.is_absolute() (otherwise ignore it and fall through to the home_dir or
temp_dir branches); update the logic in the config_file function to perform this
check before joining "iqos_cli/config.toml" so relative/empty XDG_CONFIG_HOME
values don't make you read/write in the CWD.
In `@src/main.rs`:
- Around line 353-356: The config persistence currently calls AppConfig::load()
and config.save() directly which can return errors and abort the REPL; change
this to non-fatal handling: call connected_device(...) as before, then attempt
AppConfig::load() inside a match/if let, and if loading fails log a warning and
create a default AppConfig, call update_default(&device) on the config, then
attempt config.save() but catch and log any save errors instead of propagating
them; ensure no Err from AppConfig::load or save escapes so the REPL always
continues after a successful BLE connection.
🪄 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: defaults
Review profile: CHILL
Plan: Pro
Run ID: ea45a2dd-7a01-4b7e-a75b-5c506aeaf354
⛔ Files ignored due to path filters (1)
Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (6)
Cargo.tomlsrc/cli.rssrc/config.rssrc/loader/mod.rssrc/loader/parser.rssrc/main.rs
Member
Author
|
review @coderabbitai @copilot |
|
🧠 Learnings used✅ Actions performedFull review triggered. |
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 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/config.rs`:
- Around line 77-87: save_device currently persists arbitrary labels that may
collide with model selector names used by resolve_target (e.g., "iluma" or
"iluma-i"), causing saved labels to be unreachable via --model; add validation
in save_device to reject any label that matches reserved model-selector names by
reusing a shared helper (e.g., introduce/use is_reserved_model_label or
validate_label) and check before inserting into self.devices, returning an error
or refusing to save; update SavedDevice creation path to only run after
validation and ensure the same helper is used by resolve_target to keep the set
of reserved names consistent.
In `@src/main.rs`:
- Around line 163-171: The current flow aborts on AppConfig::load and
config.save even for known-model targets; change it so AppConfig::load errors
are non-fatal when resolve_target indicates a known-model target (i.e., when
target identifies a built-in/registered model from model_arg) by falling back to
a default/empty AppConfig and logging a warning instead of returning ExitError;
likewise make apply_connection_memory + config.save non-fatal for the console
and registered-command one-shot paths (the code paths around connect_target,
apply_connection_memory, and the save call): on save failure, log a warning and
continue only when the target is a known-model and the current mode is console
or registered-command, but keep load/save errors fatal for default/label
resolution and for device save/remove commands. Use the symbols AppConfig::load,
resolve_target, connect_target, apply_connection_memory, and config.save to
locate and implement the conditional fallback and warnings.
🪄 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: defaults
Review profile: CHILL
Plan: Pro
Run ID: 9e2f1459-ed7c-4fba-b012-1e09a1db381e
📒 Files selected for processing (11)
.gitignorehistory.txtsrc/cli.rssrc/config.rssrc/loader/cmds/device.rssrc/loader/cmds/help.rssrc/loader/cmds/mod.rssrc/loader/iqoshelper.rssrc/loader/mod.rssrc/loader/parser.rssrc/main.rs
💤 Files with no reviewable changes (1)
- history.txt
✅ Files skipped from review due to trivial changes (3)
- src/loader/cmds/help.rs
- .gitignore
- src/loader/cmds/mod.rs
🚧 Files skipped from review as they are similar to previous changes (3)
- src/loader/mod.rs
- src/loader/parser.rs
- src/cli.rs
Member
Author
|
review @coderabbitai @copilot |
|
🧠 Learnings used✅ Actions performedFull review triggered. |
Contributor
There was a problem hiding this comment.
Pull request overview
Adds a non-interactive “one-shot” CLI mode to the existing IQOS interactive console, including persistent device memory (default device + labelled devices) stored in a TOML config, and reuses the existing command registry for executing device actions.
Changes:
- Introduces a clap-based CLI with global
--model/--timeout, one-shot command routing, and an auto-connected console mode. - Adds persistent
config.tomldevice memory (default + labelled devices) with list/save/remove flows. - Extends the interactive REPL to support
devicesubcommands and to remember the last connected device.
Reviewed changes
Copilot reviewed 12 out of 14 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| src/main.rs | Implements CLI entry path, target resolution, scan/connect flow, exit codes, and interactive flow integration with device memory. |
| src/cli.rs | Defines clap CLI surface, one-shot command mapping, global option normalization, and scan timeout resolution. |
| src/config.rs | Implements config load/save + device memory model and label validation. |
| src/model_selector.rs | Adds parsing/normalization for model selectors and “reserved label” detection with tests. |
| src/loader/parser.rs | Extends console execution to support device command and adds helpers for running with connected device / one-shot execution. |
| src/loader/mod.rs | Re-exports new console entry points. |
| src/loader/iqoshelper.rs | Adds completions for device subcommands and updates tests. |
| src/loader/cmds/device.rs | New REPL device command (list/save/remove) backed by config. |
| src/loader/cmds/mod.rs | Registers the new device command module. |
| src/loader/cmds/help.rs | Updates help output to include the new device command. |
| Cargo.toml | Adds clap/serde/toml dependencies for CLI parsing + config persistence. |
| Cargo.lock | Locks new transitive dependencies from clap/toml/serde additions. |
| .gitignore | Adds an explicit ignore for history.txt. |
| history.txt | Removes the committed history file. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (3)
.gitignore (1)
15-15: Redundant pattern; already covered by line 2.The
*.txtpattern on line 2 already ignores all.txtfiles, includinghistory.txt. The explicit entry here is redundant and can be removed for cleaner maintainability.🧹 Proposed fix to remove redundancy
-history.txt🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.gitignore at line 15, Remove the redundant "history.txt" entry from .gitignore because the existing "*.txt" pattern already covers it; locate the explicit "history.txt" line and delete it so only the general "*.txt" rule remains, keeping the file clean and maintainable.src/main.rs (1)
564-576: Fragile exit-code classification via error-message prefix matching.
classify_command_errorpattern-matches onanyhow::Error::to_string()prefixes. Any rewording in a command'sbail!/anyhow!call (or a new user-input error prefix added later) will silently misclassify exit codes, which the spec ties to user-observable behavior (2 = invalid args vs. 3 = device error). This is a growing maintenance hazard as the command registry expands.Consider introducing a typed error (e.g., a
CommandErrorenum withInvalidArgs/DeviceFailurevariants) returned by command handlers, or attach a sentinel viaanyhow::Error::downcast_refso classification is based on type rather than text. At minimum, centralize the usage/invalid-arg messages behind constants that both the bailing code andclassify_command_errorreference.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/main.rs` around lines 564 - 576, classify_command_error currently inspects the string from anyhow::Error to decide exit codes, which is fragile; change command handlers to return a typed error (e.g., introduce a CommandError enum with variants InvalidArgs and DeviceFailure) or attach a sentinel type that can be downcast from anyhow::Error; update classify_command_error to first attempt error.downcast_ref::<CommandError>() and map those variants to EXIT_INVALID_ARGUMENTS or EXIT_DEVICE_COMMAND_FAILED, and only fall back to string checks if no typed error is present; update existing bail!/anyhow! calls in command handlers to produce/convert to CommandError so classification is type-based rather than text-based.src/loader/cmds/device.rs (1)
24-35: Optional: validate label before mutatingconfigstate.
config.save_device(viavalidate_device_labelinsidesrc/config.rs) may reject the label afterupdate_defaulthas already mutated the in-memory config. It's harmless today becauseconfig.save()only runs on the success path, but reordering makes the intent clearer and future-proofs against accidental early saves:♻️ Proposed refactor
let mut config = AppConfig::load()?; - config.update_default(device); config.save_device(label.to_string(), device)?; + config.update_default(device); config.save()?;🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/loader/cmds/device.rs` around lines 24 - 35, Load the AppConfig first, then validate the provided label before mutating in-memory config: after calling AppConfig::load(), call the label validation routine (the existing validate_device_label logic exposed on AppConfig) with the incoming label and bail on failure, and only then call config.update_default(device), config.save_device(label.to_string(), device) and finally config.save(); this prevents update_default from mutating state when the label would later be rejected.
🤖 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/config.rs`:
- Around line 63-70: The current save_to function uses fs::write which creates
the file with default permissions; change it to create and write the file via
std::fs::OpenOptions so you can set restrictive permissions (owner read/write
only) on Unix. In save_to, after serializing with toml::to_string_pretty(self),
open the path with OpenOptions::new().write(true).create(true).truncate(true)
and, using std::os::unix::fs::OpenOptionsExt::mode(0o600) on Unix targets, write
the contents and flush; keep the existing context error messages. Ensure this is
done for the same save_to function and that Windows builds still compile (guard
Unix-specific imports/usage behind cfg(unix) if needed).
In `@src/main.rs`:
- Around line 195-216: The DeviceList and DeviceRemove branches currently map
AppConfig::load() and config.save() errors to EXIT_CONNECTION_FAILED; change
these to return EXIT_DEVICE_COMMAND_FAILED instead so pure-config I/O failures
are not reported as BLE connection errors—specifically update the map_err
closures around AppConfig::load() in OneShotCommand::DeviceList and
OneShotCommand::DeviceRemove, and the config.save() error handling in
DeviceRemove (and likewise the post-connect config.save() in
OneShotCommand::DeviceSave) to use ExitError::new(EXIT_DEVICE_COMMAND_FAILED,
... ) instead of EXIT_CONNECTION_FAILED.
---
Nitpick comments:
In @.gitignore:
- Line 15: Remove the redundant "history.txt" entry from .gitignore because the
existing "*.txt" pattern already covers it; locate the explicit "history.txt"
line and delete it so only the general "*.txt" rule remains, keeping the file
clean and maintainable.
In `@src/loader/cmds/device.rs`:
- Around line 24-35: Load the AppConfig first, then validate the provided label
before mutating in-memory config: after calling AppConfig::load(), call the
label validation routine (the existing validate_device_label logic exposed on
AppConfig) with the incoming label and bail on failure, and only then call
config.update_default(device), config.save_device(label.to_string(), device) and
finally config.save(); this prevents update_default from mutating state when the
label would later be rejected.
In `@src/main.rs`:
- Around line 564-576: classify_command_error currently inspects the string from
anyhow::Error to decide exit codes, which is fragile; change command handlers to
return a typed error (e.g., introduce a CommandError enum with variants
InvalidArgs and DeviceFailure) or attach a sentinel type that can be downcast
from anyhow::Error; update classify_command_error to first attempt
error.downcast_ref::<CommandError>() and map those variants to
EXIT_INVALID_ARGUMENTS or EXIT_DEVICE_COMMAND_FAILED, and only fall back to
string checks if no typed error is present; update existing bail!/anyhow! calls
in command handlers to produce/convert to CommandError so classification is
type-based rather than text-based.
🪄 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: defaults
Review profile: CHILL
Plan: Pro
Run ID: 93c53d40-d769-4b8f-b65e-7df270983259
⛔ Files ignored due to path filters (1)
Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (13)
.gitignoreCargo.tomlhistory.txtsrc/cli.rssrc/config.rssrc/loader/cmds/device.rssrc/loader/cmds/help.rssrc/loader/cmds/mod.rssrc/loader/iqoshelper.rssrc/loader/mod.rssrc/loader/parser.rssrc/main.rssrc/model_selector.rs
💤 Files with no reviewable changes (1)
- history.txt
Member
Author
|
CodeRabbit review follow-up for commit ac56870:
|
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
src/loader/parser.rs (1)
37-53: Fragile prefix-string classification ties exit codes to error message text.
CommandError::classifyandis_invalid_argument_messagedistinguish "argument" vs "device" failures by scanning the first few characters of the error'sDisplaystring. Per the issue spec, this mapping drives exit codes2vs3, so:
- Any future command-module error that happens to start with
Usage:,Invalid option:, etc. will be reported as an argument error even if it was produced deep inside a BLE/device call.- Conversely, an argument-validation message that doesn't start with one of the five hardcoded prefixes silently degrades to
DeviceFailure(exit3).- The prefixes are also duplicated at the
main.rsside (classify_command_errorfalls back tois_invalid_argument_message), so every command now has an implicit contract on its error wording.Consider having individual command handlers return
CommandError::invalid_arguments(...)directly (viaanyhow::Error::from(CommandError::...)) soclassify_command_errorinmain.rscan rely purely ondowncast_ref::<CommandError>()and drop the string-matching fallback.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/loader/parser.rs` around lines 37 - 53, Replace fragile prefix-based classification by making CommandError the canonical source: update classify (fn classify(error: anyhow::Error) -> Self) to first try error.downcast_ref::<CommandError>() (or error.downcast::<CommandError>()) and return the concrete CommandError if present; only then (for compatibility) fall back to is_invalid_argument_message(message) as a last resort. Add/ensure a constructor like CommandError::invalid_arguments(...) (or CommandError::InvalidArguments(...) conversion helper) that command handlers use to return argument errors directly (convert to anyhow::Error where needed). Also update command handlers to return CommandError::invalid_arguments(...) for argument validation failures so classify_command_error in main.rs can rely on downcast_ref::<CommandError>() and you can eventually remove the brittle is_invalid_argument_message checks.src/config.rs (1)
66-93: Non-atomic save can corruptconfig.tomlon crash/interrupt.
save_toopens the target path withtruncate(true)and then streamscontents+flush(). If the process is killed (SIGINT, panic, power loss) between the truncate and the final flush, the previous contents — every[devices.<label>]entry plus[default]— are already gone and the file is left empty or partial. Given this file is the single source of truth for user-labelled devices, the blast radius is larger than usual.Consider the standard write-temp + rename pattern so the old file survives until the new one is fully written:
♻️ Proposed refactor
- let contents = toml::to_string_pretty(self)?; - let mut options = OpenOptions::new(); - options.write(true).create(true).truncate(true); - - #[cfg(unix)] - { - options.mode(0o600); - } - - let mut file = options - .open(&path) - .with_context(|| format!("failed to write {}", path.display()))?; - - #[cfg(unix)] - file.set_permissions(fs::Permissions::from_mode(0o600)) - .with_context(|| format!("failed to set permissions on {}", path.display()))?; - - file.write_all(contents.as_bytes()) - .with_context(|| format!("failed to write {}", path.display()))?; - file.flush() - .with_context(|| format!("failed to write {}", path.display())) + let contents = toml::to_string_pretty(self)?; + let tmp_path = path.with_extension("toml.tmp"); + + let mut options = OpenOptions::new(); + options.write(true).create(true).truncate(true); + #[cfg(unix)] + { + options.mode(0o600); + } + + let mut file = options + .open(&tmp_path) + .with_context(|| format!("failed to write {}", tmp_path.display()))?; + + #[cfg(unix)] + file.set_permissions(fs::Permissions::from_mode(0o600)) + .with_context(|| format!("failed to set permissions on {}", tmp_path.display()))?; + + file.write_all(contents.as_bytes()) + .with_context(|| format!("failed to write {}", tmp_path.display()))?; + file.sync_all() + .with_context(|| format!("failed to sync {}", tmp_path.display()))?; + drop(file); + + fs::rename(&tmp_path, &path) + .with_context(|| format!("failed to replace {}", path.display()))🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/config.rs` around lines 66 - 93, The save_to function currently truncates the target (OpenOptions::truncate(true)) and streams contents directly which can leave config.toml empty/partial on crash; change save_to to write to a temporary file in the same directory (e.g. create a temp PathBuf next to path), write the serialized contents to that temp file (use OpenOptions without truncating the real file), set the file mode/permissions on unix (the same 0o600 logic currently applied to options/file), flush and sync data (fsync) on the temp file, then atomically rename (fs::rename) the temp file to path so the original file remains intact until the new file is fully persisted; ensure error contexts remain (with_context) and remove the temp file on failure.
🤖 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/config.rs`:
- Around line 101-146: validate_device_label currently only trims for emptiness
but returns the original label, causing padded labels to be stored; change
validate_device_label(&str) to return a Result<String> with the trimmed label
and update callers to use that trimmed String (not the original) when calling
save_device(label, ...), update_saved_device_metadata(label, ...),
remove_device(label), and any confirmation prints (e.g., the DeviceSave flow in
run_one_shot). Ensure is_reserved_model_label/parse_device_model behavior
remains unchanged and that save_device, update_saved_device_metadata, and
remove_device continue to accept the label types they currently use (convert to
&str or clone as needed).
In `@src/main.rs`:
- Around line 387-405: In find_matching_peripheral, per-peripheral backend
errors from central.peripheral(&addr).await and peripheral.properties().await
are currently propagated via ? which aborts the whole scan; change those awaits
to handle errors locally (use match/if let Err(...) { continue } or
map_err/log_and_continue) so a transient failure on one addr skips that
peripheral and the loop keeps draining events until timeout or a match; keep the
existing return Ok(Some((peripheral, discovered))) path and ensure the loop
still returns Ok(None) on timeout and preserves the outer anyhow::Error result
type.
---
Nitpick comments:
In `@src/config.rs`:
- Around line 66-93: The save_to function currently truncates the target
(OpenOptions::truncate(true)) and streams contents directly which can leave
config.toml empty/partial on crash; change save_to to write to a temporary file
in the same directory (e.g. create a temp PathBuf next to path), write the
serialized contents to that temp file (use OpenOptions without truncating the
real file), set the file mode/permissions on unix (the same 0o600 logic
currently applied to options/file), flush and sync data (fsync) on the temp
file, then atomically rename (fs::rename) the temp file to path so the original
file remains intact until the new file is fully persisted; ensure error contexts
remain (with_context) and remove the temp file on failure.
In `@src/loader/parser.rs`:
- Around line 37-53: Replace fragile prefix-based classification by making
CommandError the canonical source: update classify (fn classify(error:
anyhow::Error) -> Self) to first try error.downcast_ref::<CommandError>() (or
error.downcast::<CommandError>()) and return the concrete CommandError if
present; only then (for compatibility) fall back to
is_invalid_argument_message(message) as a last resort. Add/ensure a constructor
like CommandError::invalid_arguments(...) (or
CommandError::InvalidArguments(...) conversion helper) that command handlers use
to return argument errors directly (convert to anyhow::Error where needed). Also
update command handlers to return CommandError::invalid_arguments(...) for
argument validation failures so classify_command_error in main.rs can rely on
downcast_ref::<CommandError>() and you can eventually remove the brittle
is_invalid_argument_message checks.
🪄 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: defaults
Review profile: CHILL
Plan: Pro
Run ID: c0e970f2-9635-4896-9fc0-3abaa84875a7
📒 Files selected for processing (4)
src/config.rssrc/loader/cmds/device.rssrc/loader/parser.rssrc/main.rs
🚧 Files skipped from review as they are similar to previous changes (1)
- src/loader/cmds/device.rs
Member
Author
|
Handled the CodeRabbit review-body nitpicks in 6865ad5 as well:
Verified with:
|
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
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
Verification
Related issue: #5
Summary by CodeRabbit
New Features
Documentation
Chores