Support empty-bodied computation expressions #17352
Conversation
❗ Release notes required
|
|
Haven't reviewed it yet, but thinking out loud: How would this play with resumable state machines. I assume it should be fine since it's just a transformation essentially. And second, orthogonal comment - this feature will encourage more CE use, which my itself is not bad, but might lead to more nested CEs, which will involve more checking = being slower. We really need to do something with it. I made an attempt or two in the past, but change surface was bigger than I anticipated. |
That's a good call-out. I'll add some tests to make sure. |
|
/azp run |
|
Azure Pipelines successfully started running 2 pipeline(s). |
|
Cool stuff! Hope to give it a review soon. Can be a nice and intuitive addition to the language. |
...iler.ComponentTests/Conformance/Expressions/ComputationExpressions/ComputationExpressions.fs
Show resolved
Hide resolved
* The warning caused by dotnet#12038 distracted from the test's intent. Wrapping the code in a function removes the warning.
There was a problem hiding this comment.
Nice work, great testing. I am not a CE guru (yet) so letting @vzarytovskii to make a final review here.
…oll/fsharp into empty-bodied-cexprs
Description
Examples
Checklist
Important
Shoutout to @edgarfgp @fabulous-dev for sponsoring my work on this.