added the check for negative sec

[Cosmicoppai]

 RUBY_VERSION
=> "3.2.3"
irb(main):006> require 'timeout'
=> true

# Case 1: Negative timeout with quick execution

irb(main):007* Timeout.timeout(-5) do
irb(main):008*   puts "This should not execute"
irb(main):009> end
This should not execute
=> nil

# Case 2: Negative timeout with sleep

irb(main):010* Timeout.timeout(-5) do
irb(main):011*   sleep(10)
irb(main):012> end
C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:43:in `rescue in handle_timeout': execution expired (Timeout::Error)
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:40:in `handle_timeout'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:195:in `timeout'
        from (irb):10:in `<main>'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/irb-1.12.0/exe/irb:9:in `<top (required)>'
        from C:/Ruby32-x64/bin/irb:25:in `load'
        from C:/Ruby32-x64/bin/irb:25:in `<main>'
(irb):11:in `sleep': execution expired (Timeout::ExitException)
        from (irb):11:in `block in <top (required)>'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:186:in `block in timeout'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:41:in `handle_timeout'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/timeout-0.4.1/lib/timeout.rb:195:in `timeout'
        from (irb):10:in `<main>'
        from C:/Ruby32-x64/lib/ruby/gems/3.2.0/gems/irb-1.12.0/exe/irb:9:in `<top (required)>'
        from C:/Ruby32-x64/bin/irb:25:in `load'
        from C:/Ruby32-x64/bin/irb:25:in `<main>'

 Timeout::VERSION
=> "0.4.1"

Potential issues with current behaviour:

• Inconsistency: The behaviour differs based on the block's content, which may not be immediately apparent
• Silent failure: The negative timeout is silently ignored, potentially masking logical errors in the calling code.
• Unexpected source of error: one might expect the timeout method to validate its input, rather than relying on methods called within the block to catch invalid time values.

I suggest this change for the consistent behaviour regardless of code-block as well as the clear indication of the source of the error

[eregon]

Thanks for the PR, could you add tests for this?

[eregon]

Could you also update the docs?

[jeremyevans]

Thank you for working on this. I agree we should raise for negative numbers, as you can make the argument that the timeout has already expired. It's arguable whether we should raise Timeout::Error (or klass) in this case, instead of ArgumentError, but I'm not opposed to using ArgumentError. I do have a couple inline requests for changes.

[jeremyevans]

Sorry I missed approving this a few days ago. Thanks for the reminder!

[Cosmicoppai]

@eregon whenever you find time, I’d greatly appreciate your review on the updated changes, thank you for your guidance!

[eregon]

LGTM, thanks for the review @jeremyevans