regex - 如何在`StringCvt.scanString（RE.findcompiledComment）输入`中访问len和pos

Question

背景：我正在尝试使用正则表达式来解析一种语言的评论，其中开头//：

structure Main =
struct
structure RE = RegExpFn(
    structure P = AwkSyntax
    structure E = ThompsonEngine
)
val regexes = [
    ("[a-zA-z@= ]* *//.*",   fn match => ("comment", match)),
    ("[0-9]*",      fn match => ("2nd", match)),
    ("1tom|2jerry", fn match => ("3rd", match))
]
fun main () =
    let
        val input = "@=abs //sdfasdfdfa sdf as"
        val comment = "[a-zA-z@= ]* *//"
        val compiledComment = RE.compileString comment
    in
        (* #1 StringCvt.scanString (RE.match regexes) input *)
        (* #2 StringCvt.scanString (RE.find compiledComment) input *)
        (* #3 case ... of ... *)
    end

    end

input是我的测试用例，我希望修剪//sdfasdfdfa sdf as和保留@=abs。

以下是我的一些试验：

• 让我们StringCvt.scanString (RE.find compiledComment) input的fun main返回值：

---

- Main.main();
[autoloading]
[autoloading done]
val it = SOME (Match ({len=8,pos=0},[])) : StringCvt.cs Main.RE.match option

---

• 设为StringCvt.scanString (RE.match regexes) input返回值：

---

- Main.main();
[autoloading]
[autoloading done]
val it = SOME ("comment",Match ({len=#,pos=#},[]))
  : (string * StringCvt.cs Main.RE.match) option

---

这两种情况告诉我StringCvt.scanString (RE.find compiledComment) input是我想要的，因为它的值包含{len=8,pos=0},[])，可以用来修剪所有评论。但我对它的价值和类型有点困惑：val it = SOME (Match ({len=8,pos=0},[])) : StringCvt.cs Main.RE.match option. 我怎样才能访问len和pos这里？为什么StringCvt.cs并且Main.RE.match只按空间分割？

谷歌搜索 sml 的文档后，我在下面包含了所有信息：

#+BEGIN_SRC sml
StringCvt.scanString (RE.match regexes) input
val it = SOME (
        "comment"  ,           Match             ({len=#,pos=#},[]))
 :      (string    *           StringCvt.cs      Main.RE.match)
         option



StringCvt.scanString (RE.find compiledComment) input
val it = SOME (
        Match                      ({len=8,pos=0},[]))
 :      StringCvt.cs               Main.RE.match
        option



val find : regexp ->
(char,'a) StringCvt.reader -> ({pos : 'a, len : int} option MatchTree.match_tree,'a) StringCvt.reader

val scanString :
((char, cs) reader -> ('a, cs) reader) -> string -> 'a option

val match : (string * ({pos : 'a, len : int} option MatchTree.match_tree -> 'b)) list -> (char,'a) StringCvt.reader -> ('b,'a) StringCvt.reader

#+END_SRC
type cs
The abstract type of the character stream used by scanString. A value of this type represents the state of a character stream. The concrete type is left unspecified to allow implementations a choice of representations. Typically, cs will be an integer index into a string.

IIUC, 的类型Match应该是StringCvt.cs, 的类型({len=8,pos=0},[]))应该({len=#,pos=#},[]))是Main.RE.match. 然后我开始模式匹配：

let
...
in 
case StringCvt.scanString (RE.find compiledComment) input of
            NONE => ""
         |  SOME (
                StringCvt.cs ({len = b, pos = a}, _)) => String.substring (input a b)

不幸的，

main.sml:23.19-23.39 Error: non-constructor applied to argument in pattern
main.sml:23.92 Error: unbound variable or constructor: a
main.sml:23.94 Error: unbound variable or constructor: b
main.sml:23.86-23.95 Error: operator is not a function [tycon mismatch]
  operator: string
  in expression:
    input <errorvar>
[autoloading failed: unable to load module(s)]
stdIn:1.2-1.11 Error: unbound structure: Main in path Main.main

似乎我不能StringCvt.cs用于模式，因为它不是构造函数。然后我尝试使用wildcard：

case StringCvt.scanString (RE.find compiledComment) input of
    NONE => ""
 |  SOME (_ ({len = b, pos = a}, _)) => String.substring (input a b)

,

main.sml:23.19 Error: non-constructor applied to argument in pattern

那么，这里的构造函数Match是必须的吗？我无法再深入挖掘。你有什么想法？提前致谢

标签： regexpattern-matchingsmlsmlnj

Answer 1

解决了：

case StringCvt.scanString (RE.find compiledComment) input
         of NONE => ""
          | SOME match =>
            let
                val {pos, len} = MatchTree.root match
            in
                String.substring (input, 0, pos)
            end