🚀 go-pugleaf

The documentation for imageto, including the example given, indicates that
the values given as -baseline arguments are scanlines numbered relative to
the topmost (first) scanline of the corresponding image row, e.g. given an
image,

       01234567890123456789
      0
      1         x
      2 x   x   x
      3         x
      4 x   x   x   xxxxx
      5 x   x   x   x x x
      6     x       x x x
      7   xx
      8
      9
     10 xxxxxxxxxxxxxxx

the default baselines should be "4,0".  When I analyzed an image and
provided the appropriate baseline specs, however, the resulting imageto
output font was not correct; glyph baselines were off by a significant
percentage of the total height (although consistently so within characters
from a particular image row).

Further experimentation and analysis of the print-guidelines output makes
it appear is if imageto is setting default baselines relative to the
*bottommost* (last) scanline of an image row; e.g. the baselines for the
above example need to be "1,0".  Furthermore, correcting the original
baselines based on this hypothesis yielded a correct font!

Dunno if this is a documentation bug or an imageto problem, but something's
clearly not right.  The actual speciman image and IFI file involved are
available if you'd like to see them.

pch