Re: [問題] Wrapping/Widening + Var-args

[tkcn (小安)]

※ 引述《bennylu (減肥)》之銘言：
: 在設計overloading method時, 如下:
:         method(5);
:         ..
:         void method(long i) {}
:         void method(Integer i) {}
:         void method(int... i) {}
: 我們知道優先權順序是widening > wrapping > var-args,
: 但是如果將上述三者做一些組合,
:         method(5);
:         ..
:         void method(long... i) {}
:         void method(Integer... i) {}
: 照理說應該是 widening + var-args > wrapping + var-args,
: 但是compiler抱怨說 reference to "method" is ambiguous,
: 所以想來問一下各位先進, compiler在挑選overloaded method時詳細的演算法,
: 先謝謝啦

這不能算是演算法，只能算是規則吧。

根據 Java 1.5 Tiger - A Developer's Notebook

method resolution is a three-pass process:

1. The compiler attempts to locate the correct method
   without any boxing, unboxing, or vararg invocations.
   This will find any method that would have been
   invoked under Java 1.4 rules.

2. If the first pass fails, the compiler tries method
   resolution again, this time allowing boxing and
   unboxing conversions.
   Methods with varargs are not considered in this pass.

3. If the second pass fails, the compiler tries method
   resolution one last time, allowing boxing and unboxing,
   and also considers vararg methods.

--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 140.122.183.195
※ 編輯: tkcn            來自: 140.122.183.195      (09/08 00:13)
※ 編輯: tkcn            來自: 140.122.183.195      (09/08 00:14)
推 bennylu:謝謝學長(誤), 不過似乎沒有提到為何會ambiguous
→ ogamenewbie:單看字面就是進第3階段, 然後多重符合?
→ ogamenewbie:還是你確定那個 wid啥 > 的那個規則在第3階段也適用?
→ tkcn:我的認知就是 2f 那樣，也就是你的 "照理說" 是錯的。
→ bennylu:同意
推 dream1124:好問題 推