Misalignements in output of free with LANG=fr_FR.UTF-8

For reference, here’s the output of LANG=en_US.UTF-8 free on my system:

               total        used        free      shared  buff/cache   available
Mem:        14272892     3280112     8881508      111944     2111272    10613568
Swap:       16776512           0    16776512

Now, here’s the output when using LANG=fr_FR.UTF-8 free:

               total       utilisé      libre     partagé tamp/cache   disponible
Mem:        14272892     3299744     8864056      111728     2109092    10594196
Partition d'échange:   16776512           0    16776512

Should be something like:

                           total     utilisé       libre     partagé  tamp/cache  disponible
Mem:                    14272892     3299744     8864056      111728     2109092    10594196
Partition d'échange:    16776512           0    16776512

I see two problems:

• The header labels are not properly aligned with their columns
• The size of the labels of the first column doesn’t seem to be taken into account at all