line height redux

Re: our discussion of lineHeight on today's call, the link to the long
treatment I gave on this subject is [1].

[1] http://lists.w3.org/Archives/Public/public-tt/2013Aug/0057.html

Repeating the results found there:

CONCLUSION: In TTML1, the height of a line area rectangle generated by a
paragraph (<p/>) element is the greater of (a) the computed value of the
lineHeight property that applies to the paragraph and (b) the sum of the
maximum altitude and maximum depth of each font that applies to the line
area's child inline areas.

This conclusion is based on the fact that we have specified use of the CSS
compatible "line-height" line stacking strategy when mapping to XSL-FO.

Note that in this conclusion, the computed value of the lineHeight property
of a paragraph is a parameter in determining the actual used height trait
of a given line area generated by the paragraph.

Note further that if the specified value of tts:lineHeight is "normal" or
if the initial value of "normal" applied (due to inheritance in the absence
of a specified tts:lineHeight), then the computed value of the lineHeight
property of the paragraph will already be at least 125% as large as the
largest font size of any of its descendant (inline) elements. As a
consequence, the above CONCLUSION reduces to:

CONCLUSION: In TTML1, the height of a line area rectangle generated by a
paragraph (<p/>) element with a specified or inherited lineHeight of *normal*
is the computed value of the lineHeight property that applies to the
paragraph.

This is why today I said that if one uses (or inherits) "normal", then only
one height applies to all lines.

This subject is particularly confusing due to the multiple uses of the
phrase "line height", e.g.,

   - content attribute tts:lineHeight
   - initial value of tts:lineHeight property
   - computed value of tts:lineHeight property
   - line stacking strategy of "line-height"
   - the height of a generated line area (aka line box)

G.