Scientific `d` notation not read in C++ [closed]

问题:

I need to read a data file in which numbers are written in a format like this:

1.0d-05

C++ doesn't seem to recognize this type of scientific notation! Any ideas on how I could read/convert those types of numbers?

I need numbers (i.e. double / float) not strings. Maybe there is already a class / header to manage this format, but I could not find it.

回答1:

Files produced by Fortran programs report double precision numbers (in scientific notation) using the letter D instead of E.

So your options are:

1. Preprocess the Fortran data file (a simple Search and Replace is enough).

2. Use something like:

   #include <iostream>
   #include <sstream>
   #include <string>
   #include <vector>
   
   int main()
   {
     std::istringstream input("+1.234000D-5 -2.345600D+0 +3.456700D-2");
   
     std::vector<double> result;
   
     std::string s;
     while (input >> s)
     {
       auto e(s.find_first_of("Dd"));
       if (e != std::string::npos)
         s[e] = 'E';
   
       result.push_back(std::stod(s));
     }
   
     for (auto d : result)
       std::cout << std::fixed << d << std::endl;
   
     return 0;
   }
   

Also:

• take a look at Reading ASCII numbers using "D" instead of "E" for scientific notation using C (it's C but you get the idea);
• double check the input file since Fortran can drop the 'E' / 'D' (see For three digit exponents Fortran drops the 'E' in the output and How to read FORTRAN formatted numbers in C++).