如何在Fortran中获取派生类型组件的编号、名称和值

我不熟悉Fortran。我想开发一个子程序，将分配的变量信息打印到Fortran 95中的大型派生类型。为了简化，假设我们有一个派生类型声明和assignment，如下所示：

type SubjectType
    character(20) :: genre
    character(20) :: maindude
end type SubjectType

type BookType
    character(20) :: title
    character(20) :: author
    type(SubjectType) :: subject 

end type Booktype

type(Booktype) :: Book

Book%title = "Harry Potter"
Book%author = "JK Rowling"
Book%subject%genre = "Fantasy"
Book%subject%maindude = "Ron Weasley"

我希望我的程序的输出是一个文本文件，如下所示：

type SubjectType
    character(20) :: genre
    character(20) :: maindude
end type SubjectType

type BookType
    character(20) :: title
    character(20) :: author
    type(SubjectType) :: subject 

end type Booktype

type(Booktype) :: Book

Book%title = "Harry Potter"
Book%author = "JK Rowling"
Book%subject%genre = "Fantasy"
Book%subject%maindude = "Ron Weasley"

书名，哈利波特

图书%作者，JK罗琳

书籍%主题%流派，幻想

书籍%subject%main都德，罗恩·韦斯莱

为了实现这一目标，我相信这就是我需要做的：

• 确定每个级别的派生类型中的字段数。例如，

  图书类型中的字段数为3。
  Booktype%SubjectType
  中的字段数应为2

• 找到将字段“编号”与字段名称关联的方法（可能使用指针？）
• 循环遍历所有字段号并获取它们的名称和值

---

我的问题有两个。首先，我的方法是否正确/是否会导致预期结果？其次，我如何完成此过程的步骤1。也就是说，我如何在每个级别获得派生类型中的字段数？

我采用了一个非常基本的“键/值容器”类来完成任务，而不是像问题中建议的那样通过索引号来标识数据类型/字段。其思想是使用两个派生类型，这些派生类型足够通用，可以重用，以创建任意数量的数据字段来表示各种图书信息，并且可以轻松访问（和/或打印）

将上述所有代码放在一个模块中。我省略了几个助手函数，因为它们不是回答问题和提供有效解决方案所必需的。不过，您将在下面的示例程序中看到其中一个（

newBook

）。请注意，循环域打印所有现有数据，这些数据可能因书籍而异。另外，对于任何不在书本数据中的键，请求

valueFromKey

，将返回字符串“None”

! Example usage:
program main
    use BookModule
    implicit none
    integer :: i, j
    character(len=6) :: num
    type(BookCollection) :: MyBooks

    allocate(MyBooks%book(2))
    MyBooks%book(1) = newBook(keys=['title', 'author', 'date', 'genre', 'lead'], values=["Harry Potter", "JK Rowling", "1997", "Fantasy", "Ron Weasley"])
    MyBooks%book(2) = newBook(keys=['title', 'author', 'lead'], values=["1984", "George Orwell", "Winston Smith"])

    print *, "LOOP OVER BOOK COLLECTION"
    do i = 1, size(MyBooks%book)
        write(num, '(i6)') i
        print *, "Item ", adjustl(num)
        do j = 1, MyBooks%book(i)%size 
            print *, MyBooks%book(i)%type, ", ", MyBooks%book(i)%fields(j)%type, ", ", MyBooks%book(i)%fields(j)%key, ", ", MyBooks%book(i)%fields(j)%val
        enddo
        print *
    enddo

    print *, "GET FIELD VALUE FROM KEY"
    print *, " Title:  ", MyBooks%book(1)%valueFromKey('title')
    print *, " Author: ", MyBooks%book(1)%valueFromKey('author')
    print *, " Date:   ", MyBooks%book(1)%valueFromKey('date')
    print *
    print *, " Title:  ", MyBooks%book(2)%valueFromKey('title')
    print *, " Author: ", MyBooks%book(2)%valueFromKey('author')
    print *, " Date:   ", MyBooks%book(2)%valueFromKey('date')
end program main

示例输出：

 LOOP OVER BOOK COLLECTION
 Item 1
 BookType, book, title, Harry Potter
 BookType, book, author, JK Rowling
 BookType, book, date, 1997
 BookType, subject, genre, Fantasy
 BookType, subject, lead, Ron Weasley

 Item 2
 BookType, book, title, 1984
 BookType, book, author, George Orwell
 BookType, subject, lead, Winston Smith

 GET FIELD VALUE FROM KEY
  Title:  Harry Potter
  Author: JK Rowling
  Date:   1997

  Title:  1984
  Author: George Orwell
  Date:   None

---

1此处提到的基本“key/value container”类缺少任何类似于散列/映射/字典的功能；查找工作仅通过在数据成员上循环，直到找到键，然后获取相应的值。它实际上只适用于中小型数据集

使用派生类型输入/输出（DTIO）-Fortran 2003-。不要这样做，不要将派生类型视为数组，这是一条地狱之路。

 LOOP OVER BOOK COLLECTION
 Item 1
 BookType, book, title, Harry Potter
 BookType, book, author, JK Rowling
 BookType, book, date, 1997
 BookType, subject, genre, Fantasy
 BookType, subject, lead, Ron Weasley

 Item 2
 BookType, book, title, 1984
 BookType, book, author, George Orwell
 BookType, subject, lead, Winston Smith

 GET FIELD VALUE FROM KEY
  Title:  Harry Potter
  Author: JK Rowling
  Date:   1997

  Title:  1984
  Author: George Orwell
  Date:   None