too low precision for double data points

[DNKpp]

Bug category

• [ ] bug - compilation error
• [ ] bug - compilation warning
• [ ] bug - runtime error
• [ ] bug - runtime warning
• [x] bug - logic error

Describe the bug

The precision used for converting the double data values from binary to textual representation seems to unreasonable low. That's at least the case in the line::data_string function, where it is constantly set to 10 (see line.cpp#L166). This leads to data points not showing up in plots with logarithmic axes scaling.

Locally I bumped this straight to 16 (std::numeric_limits<double>::digits10yields 15),, which works as intended (at least for me).

Steps to Reproduce Well, create a loglog plot with very small double data values.

Output

imprecise output

That's part of the cmd dump, sent to gnuplot

fixed output

That's part of the cmd dump, sent to gnuplot

Platform

• [ ] cross-platform issue - linux
• [x] cross-platform issue - windows
• [ ] cross-platform issue - macos

Environment Details:

• OS: Windows
• OS Version: 11
• Compiler: msvc
• Compiler version: 17.9.0

Additional context