I don't thinki double->float warnings are at all useful. They might be useful if both VC++ and GCC stopped them from happening when you are simply assigning a constant to a float, or check the number of digits in the constant to make a guess as to whether the user really is assuming higher accuracy than a float.
On Monday, February 22, 2021 at 12:55:45 PM UTC+1 Albrecht Schlosser wrote:
(2) Use a new file just for the nanosvg implementation (maybe
fl_nanosvg.cxx?) and use the #pragma before the inclusion of nanosvg
headers (no restore needed [1]).
That is a convenient solution but it would require some glue code
between FLTK and nanosvg to be put in practice because FLTK
uses one struct and 4 functions (struct NSVGimage, nsvgDelete(),
nsvgParse(), nsvgRasterizeXY() and nsvgCreateRasterizer()) defined
by nanosvg.h and nanosvgrast.h. Moreover, this would require to duplicate
in FLTK the declaration of "struct NSVGimage" instead of using that
found in nanosvg.h. Thus, option (1) may be more convenient.
Comments are owned by the poster. All other content is copyright 1998-2025 by Bill Spitzak and others. This project is hosted by The FLTK Team. Please report site problems to 'erco@seriss.com'.
