feat: single thread mode

[andylokandy]

Signed-off-by: Andy Lok [email protected]

Closes #197

[coveralls]

Pull Request Test Coverage Report for Build 7774880861

• 0 of 19 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.04%) to 78.749%

---

Totals
Change from base Build 7772646372:	0.04%
Covered Lines:	1712
Relevant Lines:	2174

---

💛 - Coveralls

[dotansimha]

Using this PR along with wasm32 runtime, I get time not implemented on this platform errors 🤔 Not sure what's the source of it. @andylokandy does the code in Span is using SystemTime somewhere? (or std::time::Instant?)

[dotansimha]

maybe this one? https://github.com/tikv/minitrace-rust/blob/master/minitrace/src/collector/global_collector.rs#L211

EDIT: Yeah that's the one :) I can confirm changing it does not fail, checking the Span collecting now @andylokandy

[dotansimha]

@andylokandy ok with this commit dotansimha@ca4ab0d , it seems to work and it does not fail on the time issue.

I'm trying to use minitrace::flush but at the moment I only get the root Span reported and all the rest are ignored 🤔
My current structure is like that:

- root (Span::root)
	- `async fn` with #[trace]
	- `async fn` with #[trace] 
		- ... several `async fn`, some with `#[trace]` and some with `in_span`
	- sync `fn` with #[trace]

The macro seems to set #[trace] to use in_span, and i think it should be fine?

EDIT: Also using in_span explicitly doesn't work for the child Spans. I only get the root reported.

🤔

[andylokandy]

did you set_local_parent()? and drop all spans before flush?

[dotansimha]

did you set_local_parent()? and drop all spans before flush?

oops, I missed that 🤦 I added it now and it seems to work! 🎉

So I think the only fix needed for PR is dotansimha@ca4ab0d

[dotansimha]

@andylokandy can we get this released to Crates? 👀

Thank you