feat(xtask): impl as_ast_kind method for each variant

[IWANABETHATGUY]

No description provided.

[Boshen]

What's the usage?

[graphite-app]

Your org has enabled the Graphite merge queue for merging into main

Add the label “0-merge” to the PR and Graphite will automatically add it to the merge queue when it’s ready to merge. Or use the label “hotfix” to add to the merge queue as a hot fix.

You must have a Graphite account and log in to Graphite in order to use the merge queue. Sign up using this link.

[IWANABETHATGUY]

What's the usage?

e.g.

oxc/crates/oxc_linter/src/rules/eslint/no_this_before_super.rs

Lines 136 to 150 in eda3d10

	fn is_wanted_node(node: &AstNode, ctx: &LintContext<'_>) -> Option<bool> {
	let parent = ctx.nodes().parent_node(node.id())?;
	if let AstKind::MethodDefinition(mdef) = parent.kind() {
	if matches!(mdef.kind, MethodDefinitionKind::Constructor) {
	let parent_2 = ctx.nodes().parent_node(parent.id())?;
	let parent_3 = ctx.nodes().parent_node(parent_2.id())?;
	if let AstKind::Class(c) = parent_3.kind() {
	let super_class = c.super_class.as_ref()?;
	return Some(!matches!(super_class, Expression::NullLiteral(_)));
	}
	}
	}
	Some(false)
	}

could be rewritten as:

    fn is_wanted_node(node: &AstNode, ctx: &LintContext<'_>) -> Option<bool> {
        let parent = ctx.nodes().parent_node(node.id())?;
        if let AstKind::MethodDefinition(mdef) = parent.kind() {
            if matches!(mdef.kind, MethodDefinitionKind::Constructor) {
                let parent_2 = ctx.nodes().parent_node(parent.id())?;
                let parent_3 = ctx.nodes().parent_node(parent_2.id())?;
                let c = parent_3.kind().as_class()?;
                let super_class = c.super_class.as_ref()?;
                return Some(!matches!(super_class, Expression::NullLiteral(_)));
            }
        }

        Some(false)
    }

[IWANABETHATGUY]

This could avoid written many nested levels if let #ty pattern match:
if let AstKind::xxx() =

[IWANABETHATGUY]

Also, this makes the code more easily combine with Option combinator

[codspeed-hq]

CodSpeed Performance Report

Merging #5491 will not alter performance

_{Comparing feat/ast-kind-as-ty-impl (cedf7a4) with main (a96866d)}

Summary

✅ 29 untouched benchmarks

[IWANABETHATGUY]

feel free to close it if you are not satisfied since it is not written by hand, 😄

[IWANABETHATGUY]

It is used in many place in our codebase:

[rzvxa]

LGTM!

I'm not sure if we need to return the inner values with the lifetime of a; But other than this thought everything looks good.

[IWANABETHATGUY]

LGTM!

I'm not sure if we need to return the inner values with the lifetime of a; But other than this thought everything looks good.

Ok, I omit the 'a lifetime

[rzvxa]

LGTM!
I'm not sure if we need to return the inner values with the lifetime of a; But other than this thought everything looks good.

Ok, I omit the 'a lifetime

Thanks, That wasn't really a change request. I just wanted to put my concern out there to see what you guys think about it.

So please feel free to revert it if you think having the lifetime can be helpful(I'm not sure how many of these pattern matchings are looking for an inner value with the lifetime so it might very well be necessary to simplify many of these).

[IWANABETHATGUY]

LGTM!
I'm not sure if we need to return the inner values with the lifetime of a; But other than this thought everything looks good.

Ok, I omit the 'a lifetime

Thanks, That wasn't really a change request. I just wanted to put my concern out there to see what you guys think about it.

So please feel free to revert it if you think having the lifetime can be helpful(I'm not sure how many of these pattern matchings are looking for an inner value with the lifetime so it might very well be necessary to simplify many of these).

I don't know it either, I think people could add it back if meets some scenario when found it is not convenient to use.

[Boshen]

AST codegen is getting too powerful 🦸

[graphite-app]

Merge activity

• Sep 5, 12:11 PM EDT: The merge label '0-merge' was detected. This PR will be added to the Graphite merge queue once it meets the requirements.
• Sep 5, 12:12 PM EDT: Boshen added this pull request to the Graphite merge queue.
• Sep 5, 12:15 PM EDT: The merge label '0-merge' was detected. This PR will be added to the Graphite merge queue once it meets the requirements.
• Sep 5, 12:33 PM EDT: Boshen merged this pull request with the Graphite merge queue.

[shulaoda]

This could avoid written many nested levels if let #ty pattern match: if let AstKind::xxx() =

But it seems that we can't stop the behavior of if let Some() =. Does this help with the following writing style?

oxc/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs

Lines 69 to 72 in cedf7a4

	fn run<'a>(possible_jest_node: &PossibleJestNode<'a, '_>, ctx: &LintContext<'a>) {
	let node = possible_jest_node.node;
	if let AstKind::CallExpression(call_expr) = node.kind() {
	if is_type_of_jest_fn_call(

[IWANABETHATGUY]

This could avoid written many nested levels if let #ty pattern match: if let AstKind::xxx() =

But it seems that we can't stop the behavior of if let Some() =. Does this help with the following writing style?

oxc/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs

Lines 69 to 72 in cedf7a4

	fn run<'a>(possible_jest_node: &PossibleJestNode<'a, '_>, ctx: &LintContext<'a>) {
	let node = possible_jest_node.node;
	if let AstKind::CallExpression(call_expr) = node.kind() {
	if is_type_of_jest_fn_call(

Yeah you could just:

diff --git a/crates/oxc_ast/src/generated/ast_kind.rs b/crates/oxc_ast/src/generated/ast_kind.rs
index 2c68cb958..e512d0034 100644
--- a/crates/oxc_ast/src/generated/ast_kind.rs
+++ b/crates/oxc_ast/src/generated/ast_kind.rs
@@ -696,7 +696,7 @@ impl<'a> AstKind<'a> {
         }
     }
 
-    pub fn as_call_expression(&self) -> Option<&CallExpression<'a>> {
+    pub fn as_call_expression(&self) -> Option<&'a CallExpression<'a>> {
         if let Self::CallExpression(v) = self {
             Some(*v)
         } else {
@@ -936,7 +936,7 @@ impl<'a> AstKind<'a> {
         }
     }
 
-    pub fn as_if_statement(&self) -> Option<&IfStatement<'a>> {
+    pub fn as_if_statement(&self) -> Option<&'a IfStatement<'a>> {
         if let Self::IfStatement(v) = self {
             Some(*v)
         } else {
diff --git a/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs b/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs
index 17be91c88..ce9be1882 100644
--- a/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs
+++ b/crates/oxc_linter/src/rules/vitest/no_conditional_tests.rs
@@ -66,30 +66,27 @@ impl Rule for NoConditionalTests {
     }
 }
 
-fn run<'a>(possible_jest_node: &PossibleJestNode<'a, '_>, ctx: &LintContext<'a>) {
+fn run<'a>(possible_jest_node: &PossibleJestNode<'a, '_>, ctx: &LintContext<'a>) -> Option<()> {
     let node = possible_jest_node.node;
-    if let AstKind::CallExpression(call_expr) = node.kind() {
-        if is_type_of_jest_fn_call(
-            call_expr,
-            possible_jest_node,
-            ctx,
-            &[
-                JestFnKind::General(JestGeneralFnKind::Describe),
-                JestFnKind::General(JestGeneralFnKind::Test),
-            ],
-        ) {
-            let if_statement_node = ctx
-                .nodes()
-                .iter_parents(node.id())
-                .find(|node| matches!(node.kind(), AstKind::IfStatement(_)));
+    let call_expr = node.kind().as_call_expression()?;
+    if is_type_of_jest_fn_call(
+        call_expr,
+        possible_jest_node,
+        ctx,
+        &[
+            JestFnKind::General(JestGeneralFnKind::Describe),
+            JestFnKind::General(JestGeneralFnKind::Test),
+        ],
+    ) {
+        let if_statement_node = ctx
+            .nodes()
+            .iter_parents(node.id())
+            .find(|node| matches!(node.kind(), AstKind::IfStatement(_)))?;
 
-            let Some(node) = if_statement_node else { return };
-
-            if let AstKind::IfStatement(if_statement) = node.kind() {
-                ctx.diagnostic(no_conditional_tests(if_statement.span));
-            }
-        }
+        let if_stmt = if_statement_node.kind().as_if_statement()?;
+        ctx.diagnostic(no_conditional_tests(if_stmt.span));
     }
+    None
 }
 
 #[test]

[IWANABETHATGUY]

I need to add the 'a back , or the code not gonna compile, 😢 @rzvxa

[rzvxa]

I need to add the 'a back , or the code not gonna compile, 😢 @rzvxa

Yes, it seems like it😓

I like this implementation(#5506) better than the one with the reference, What are your thoughts on it?