Re: [ft] How to get nominal width of monospaced fonts

[Sam Varshavchik]

Werner LEMBERG writes:

> > I'm trying to determine the nominal width of monospaced fonts, in
> > pixels, that is the normal horizontal advance of rendered glyphs
> > when FT_FACE_FLAG_FIXED_WIDTH is set.
>
> This is not sufficient information.  Are you interested in the
> *linear* (i.e., unhinted) advance width or the hinted one?  The former
> can be a fractional pixel value, while the latter is always returned
> as full pixels for TrueType fonts.[*]

I'm looking for a metric that corresponds to what I get after an  
FT_Load_Glyph(), and looking at face->glyph->advance.[xy], for monospaced  
Truetype fonts like Liberation Mono. I'm trying to answer the following  
question: here's a monospaced font where, hopefully, all glyphs have the  
same advance.x for rendering. Can I obtain this value without actually  
rendering each glyph and seeing what I get.

> > I thought, at first, to look at face->size->metrics.max_advance, but
> > if I do that I find that it's always several pixels more than what I
> > eventually get after FT_LOAD_RENDERing each glyph, and computing
> > face->glyph->advance.x >> 6.
>
> This might be due to hinting.

I was seeing differences that were more than two pixels. I would think that  
hinting-related differences would not translate to more than a pixel's worth  
of a difference.

> > For example, using Liberation Mono: with a height of 16 pixels, I
> > get 11 in metrics.max_advance, but the actual advance width, based
> > on advance.x, is 10.  With a height of 21 pixels, I get a
> > max_advance of 15, and the actual advance based on advance.x is 13.
> > Is it possible to get the advance width, of fixed width fonts,
> > without actually rendering some sample glyphs and seeing what it is?
>
> If hinting is involved, no.  Exception: The font contains an `hdmx'
> table that lists advance widths for some ppem value.

That pretty much answers my question. In the interim, I am using a heuristic  
approach of loading a sample set of glyphs from each face, and averaging out  
their advance. So far, this gives me a metric that ends up working out  
correctly, I was just wondering if I reinvented some wheel, somewhere.  
Seemed like a basic font property that gets shouted from the rooftops.  
Thanks.