Files
KaTeX/test
Eddie Kohler 4480f2c987 Correct computation of TeX sizes. (#755)
* Em and mu sizes are relative to current font's quad metric,
  which might not be "1em" in CSS terms.
* We don't need the `emPerEx` metric, because we want to convert
  TeX `ex` to CSS `em`, not TeX `ex` to TeX `em`. The `xHeight`
  metric does that already: it is relative to font size.
* Mu sizes scale with the style, em and ex do not (empirically
  checked; agrees with rule 2 of TeXbook Appendix G).
  This corrects a bug in b866cd5224 (in which em scaled),
  and a bug in previous versions (in which mu did not scale).
* Correct rule sizes. Previous comment---"the sizes of rules are
  absolute"---was misleading. Rule sizes are NOT absolute---in
  \large size, a rule denominated in 'em' is larger. But the 'em'
  unit is not sensitive to styles. So now a 1em rule will be the
  same size in super/subscript, as it should be, but an 18mu rule
  will change size in super/subscript, as it should.

Note a TODO.
2017-07-09 17:23:00 -04:00
..
2017-01-13 22:37:17 -05:00
2014-09-12 13:30:30 -07:00
2017-01-13 22:37:17 -05:00