使用Sprache解析文件时出现异常“解析失败:到达输入的意外结尾；预期=”

Question

我想解析下面的文件，

first=The_First_Step
{
    {
        value=First.Value,
    }
}

second=The_Second_Step
{
    {
        another = Second_Value,
        more =  Yet.More,
    }
}

我把语法写成，

public static NGSection ParseSections(string ngServerConfig)
{
    return Sections.End().Parse(ngServerConfig);
}

internal static Parser<string> ValueText = Parse.LetterOrDigit.AtLeastOnce().Text().Token();

internal static Parser<string> Identifier = Parse.AnyChar.AtLeastOnce().Text().Token();

internal static Parser<Config> Config =
      from id in Identifier
      from equal in Parse.Char('=').Token()
      from value in ValueText
      from comma in Parse.Char(',').Token()
      select new Config(id, value);

internal static Parser<Section> Section =
      from id in Identifier
      from equal in Parse.Char('=').Token()
      from title in ValueText
      from lbracket in Parse.Char('{').Token()
      from inbracket in Parse.Char('{').Token()
      from configs in Config.AtLeastOnce()
      from outbracket in Parse.Char('}').Token()
      from rbracket in Parse.Char('}').Token()
      select new Section(id, title, configs);

internal static Parser<NGSection> Sections =
     from sections in Section.AtLeastOnce()
     select new NGSection(sections);

我得到了异常，因为

分析失败:意外到达输入末尾；预期=(第13行，第2列)；最近使用: ore }}

任何线索都会有帮助。

Answer 1

两个问题:首先，在您的示例中，值可能包含_或.，因此LetterOrDigit不会涵盖它。应该是：

internal static Parser<string> ValueText =
  Parse.LetterOrDigit.Or(Parse.Chars('_', '.')).AtLeastOnce().Text().Token();

接下来，Identifier解析器的AnyChar太贪婪了；您需要排除=，否则它将被视为标识符的一部分：

internal static Parser<string> Identifier =
  Parse.CharExcept('=').AtLeastOnce().Text().Token();