如何将包含编号列表的字符串拆分为多个字符串？

Question

我正在尝试获取一个包含编号列表的字符串，其中每个列表旁边都有数据项，并根据每个数字旁边的数据将它们拆分为多个字符串。我尝试使用正则表达式，但这会导致一些问题，因为某些数据是货币价值，例如 120,000.00 英镑。

示例数据为

1. Mr Test Test

2. £100,000.00

3. 5 Test Road, Test Street

测试输入：

string testInput = "1. Mr Test Test 2. £100,000 3. 5 Test Road"

但编号列表可能出现在字符串中的单独行上，因为字符串是使用PdfTextStripper in PDFBox 从PDF 中提取的

有没有办法可以准确地拆分它？

我最初使用 indexof 并依靠下一个数据项标题作为停止点，但数据并不总是相同（15 指向一个，25 在另一个上） .

期望的结果是：

string Name = "Mr Test Test";
string Money = "£100,000.00";
string Address = "5 Test Road, Test Street";

任何帮助将不胜感激

Answer 1

让我们实现一个简单的生成器；我们可以在 loop 中逐行查找：

代码：

private static IEnumerable<string> ParseListToLines(string value) {
  int start = 0;
  bool first = true;

  for (int index = 1; ; ++index) {
    string toFind = $"{index}.";

    int next = value.IndexOf(toFind, start);

    if (next < 0) {
      yield return value.Substring(start);

      break;
    }

    if (!first) // we don't return text before the 1st item
      yield return value.Substring(start, next - start);

    first = false;
    start = next + toFind.Length;
  }
}

演示：

 string testInput = "1. Mr Test Test 2. £100,000 3. 5 Test Road";

 string[] lines = ParseListToLines(testInput).ToArray();

 // string Name = lines[0].Trim();
 // string Money = lines[1].Trim();
 // string Address = lines[2].Trim();

 Console.Write(string.Join(Environment.NewLine, lines));

结果：

 Mr Test Test
 £100,000
 5 Test Road

编辑：更多带有详细测试的演示（在项目内和没有项目的新行；一位和两位数字标记；文本 - ** - 在初始 1. 标记之前):

  // Let's build multiline string...
  string testInput = "**\r\n" + string.Join(Environment.NewLine, Enumerable
    .Range(1, 12)
    .Select(i => $"{i,2}.String #{i} {(i < 3 ? "\r\n   Next Line" : "")}"));

  Console.WriteLine("Initial:");
  Console.WriteLine();

  Console.WriteLine(testInput);

  Console.WriteLine();
  Console.WriteLine("Parsed:");
  Console.WriteLine();

  // ... and parse it into lines
  string[] lines = ParseList(testInput)
    .Select(line => line.Trim())
    .Select((item, index) => $"line number {index} = \"{item}\"")
    .ToArray();      

  Console.WriteLine(string.Join(Environment.NewLine, lines));

结果：

Initial:

**              // This text - before initial "1." marker should be ingnored
 1.String #1    // 1st item contains multiline text
   Next Line    // 1st item continuation
 2.String #2    // Again, multiline marker
   Next Line
 3.String #3 
 4.String #4 
 5.String #5 
 6.String #6 
 7.String #7 
 8.String #8 
 9.String #9 
10.String #10   // two digit markers: "10.", "11.", "12."
11.String #11 
12.String #12 

Parsed:

line number 0 = "String #1    // 1st item is multiline one
   Next Line"
line number 1 = "String #2    // 2nd item is multiline as well
   Next Line"
line number 2 = "String #3"
line number 3 = "String #4"
line number 4 = "String #5"
line number 5 = "String #6"
line number 6 = "String #7"
line number 7 = "String #8"
line number 8 = "String #9"
line number 9 = "String #10"
line number 10 = "String #11"
line number 11 = "String #12"

编辑 2： 好吧，让我们再试一次：

string testInput = 
  "1. test 5. wrong 2. It's Correct 3. OK 4. 1. 2. 3. - all wrong 5. Corect Now;";

string[] report = ParseList(testInput)
  .Select(line => line.Trim())
  .ToArray();

Console.Write(string.Join((Environment.NewLine, report));

结果：

test 5. wrong
It's Correct
OK
1. 2. 3. - all wrong
Corect Now;