在Smalltalk Pharo 2中無法多次查看鍵盤事件

Question

我想看一下鍵盤事件，根據Sensor的文檔，我可以執行此操作而無需使用peekKeyboardEvent從隊列中刪除事件，但是它似乎不起作用。

這有效：

"Show that a single event can be checked multiple times"
Transcript clear; show: 'Type something... '; flush.
(Delay forSeconds: 2) wait.
5 timesRepeat: [  
    Transcript show: (Sensor peekEvent); cr
]

輸出：

Type something... #(2 48243801 5 2 8 0 0 1)
#(2 48243801 5 2 8 0 0 1)
#(2 48243801 5 2 8 0 0 1)
#(2 48243801 5 2 8 0 0 1)
#(2 48243801 5 2 8 0 0 1)

但這不是：

"Show that a single keyboard event can be checked multiple times"
Transcript clear; show: 'Type something... '; flush.
(Delay forSeconds: 2) wait.
5 timesRepeat: [  
    Transcript show: (Sensor peekKeyboardEvent); cr
]

輸出：

Type something... #(2 48205144 97 0 0 97 0 1)
nil
nil
nil
nil

另一個問題：為什么Transcript flush不會導致輸出立即顯示？ 它僅在腳本運行后出現。

Answer 1

首先，pharo是一個快速移動的目標，因此最好告訴您哪個版本。

找到答案的最好方法是瀏覽代碼。 我將在當前的開發pharo 3.0中對此進行展示
如果瀏覽peekKeyboardEvent的實現者（選擇它，然后選擇Alt + m），則會在InputEventSensor中找到一個版本：

peekKeyboardEvent
    "Allows for use of old Sensor protocol to get at the keyboard,
    as when running kbdTest or the InterpreterSimulator in Morphic"

    ^eventQueue findFirst: [:buf | self isKbdEvent: buf]

如果分析inst var對eventQueue的引用

initialize
        "Initialize the receiver"
        super initialize.
        eventQueue := WaitfreeQueue new.
        ...snip...

然后瀏覽WaitfreeQueue（選擇它，然后選擇Alt + b）

findFirst: aBlock
    "Note, this method only for backward compatibility. It duplicating the semantics of #nextOrNilSuchThat: completely.
    Use #nextOrNilSuchThat: instead "

    ^ self nextOrNilSuchThat: aBlock

然后：

nextOrNilSuchThat: aBlock
    "Fetch an object from queue that satisfies aBlock, skipping (but not removing) any intermediate objects.
    If no object has been found, answer <nil> and leave me intact.

    NOTA BENE:  aBlock can contain a non-local return (^).
    Found item is removed from queue    .

    If queue currently in the middle of extraction by other process, don't wait and return <nil> immediately"
    ...snip...

您可以信任評論，也可以自己在代碼中驗證，這種方法消耗事件而不是偷看。

因此，似乎確實不贊成使用這種輪詢方式，並且失去了對pharo的支持