我可以将生成的树存储在一个Scala Def宏中，然后将其传递给另一棵

Question

说我有一个表达式，我想捕获它的树（不求值！），然后稍后再将其传递给另一个宏：

// capture a tree from some arbitrary expression
val capturedTree = captureMacro( some(expression(tree())) )

// plug the tree into here
val result = processMacro(capturedTree, otherStuff)

有没有办法做到这一点？我尝试过将capturedTree制成精致的类型，希望能保留最初的树，但那没有发生：

trait WithCapturedTree { def tree:Any }
class ReturnTreeMacro(val c:MacroContext) {
    import c.universe._

    def run(expression: Tree) =
        q"new WithCapturedTree {val tree = ${expression}}"
    }
def returnTree(expression:String):WithCapturedTree = macro ReturnTreeMacro.run

reify { returnTree("foo"+"bar") }
// returns Expr[WithCapturedTree{val tree: java.lang.String}](returnTree("foobar"))
// I.e. the result of "foo"+"bar" has already been evaluated!

有没有办法使这种方法起作用？有更好的解决方案吗？