feat: add cancellation_token method to RunningService

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM

[lucasfernog-crabnebula]

Motivation and Context

currently both RunningService::waiting and RunningService::cancel needs ownership of the RunningService instance
this makes it impossible to poll waiting() while other async task can cancel it

A cancellation_token() function is solves this, cloning the tokio_util::sync::CancellationToken
so we can dispatch a cancel from a separate task

How Has This Been Tested?

• run a service
• call cancellation_token() to grab its cancellation token
• spawn a separate task to run RunningService::waiting() while other task calls cancel()

Breaking Changes

No breaking change

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

[4t145]

Perhaps we can simply add a cancellation_token field on it, while remain the grop guard? So user can clone the cancellation token.

[lucasfernog-crabnebula]

but DropGuard can't be cloned, unless you want to change how cancellation is handled (if you clone a cancellation token and keep a copy in the RunningService the drop will never be triggered - and currently it also needs an owned value)

[4t145]

I believe it still can be triggered, you can see the source code. https://docs.rs/tokio-util/0.7.15/src/tokio_util/sync/cancellation_token/guard.rs.html#22-26

[lucasfernog-crabnebula]

@4t145 pushed

[lucasfernog-crabnebula]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

[4t145]

I explicitly added the cancellation token as part of the service struct because i don't want to call .disarm() on the drop guard - i want to preserve the "cancel on drop" functionality

alternatively we could also implement drop on our own

That's exactly what I meant

[4t145]

LGTM