Incomplete Implementation for Windows OS in svzerodsolver.cpp: Reads input files but does not save path for output.csv

Description

On Windows machines, the code for reading the input files works, but that path is not properly handled with the implementation of the if/else statements:

if (argc == 3) {
output_file_name = argv[2];

} else {
// If output file is not provided, default is <path to .json>+"output.csv"
std::size_t end_of_path = input_file_name.rfind("/");

if (end_of_path == std::string::npos) {
  end_of_path = input_file_name.rfind("\\");  // For Windows paths (?)  **<-- will find input file**

  // If <path to .json> is still not found, use current directory
  if (end_of_path == std::string::npos) {
    output_file_path = ".";
  }
} else {
  output_file_path = input_file_name.substr(0, end_of_path); **<-- skips because if {} was evaluated (missing path)**
}

output_file_name = output_file_path + "/output.csv";
std::cout << "[svzerodsolver] Output will be written to '"
          << output_file_name << "'." << std::endl;
;

}

Reproduction

Any code that relies on svzerodsolver.cpp implementation (originally found from 0d simulations in svVascularize gui).

OS: Windows 11

Expected behavior

Adding one short else statement fixed this issue, allowing the output file to save properly on my machine:

if (end_of_path == std::string::npos) {
  end_of_path = input_file_name.rfind("\\");  

  // If <path to .json> is still not found, use current directory
  if (end_of_path == std::string::npos) {
    output_file_path = ".";
  }
  **else {
    output_file_path = input_file_name.substr(0, end_of_path); <-- new code
  }** 
} 
else {
  output_file_path = input_file_name.substr(0, end_of_path);
}

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct and Contributing Guidelines

Start in svzerodsolver.cpp at the output-file path handling for the input JSON argument, and reproduce the behavior with a Windows-style path. Verify that an omitted output argument produces output.csv beside the input file, while a bare filename still uses the current directory. Confirm the result on Windows or with path-focused tests.