Here are some of the features I implemented the past few days:

- Added lexically scoped variables.
- Added support for selecting a calligraphic math font. The Termcat code +M is equivalent to the LaTeX code \mathcal{M}.
- It's now possible to use HTML tags as well as numerical and named entities in Termcat without escaping them. They 'fall through' and are outputted as raw HTML.
- Added special support for the unary minus operator: -x is syntactic sugar for -~ x. In contrast, for the binary operator the notation x ~-~ y is used (or x - y after setting up the necessary bindings).
- Like in LaTeX, whitespace is now more wide if it follows a colon or full stop (but not an acronym).
- It's now possible to write hé as h%'e and nǐ hǎo as n%3i h%3ao. (I should learn more about Pinyin and figure out if it would be possible to convert the standard notation ni3 hao3 automatically into nǐ hǎo!)
- Arbitrarily wide spaces can be inserted using the notation %_1.2em. Manual kerning is possible using the notation %+2px or %-.5ex.
- Added support for superscript (x^y), subscript (x_y) and fractions (x|y) in math mode. Similar to italics and other decorators, the outmost brackets in an expression like (1 + x)|(2 - y) are removed, so as to allow for spaces in the numerator and denominator.

I'm quite happy with my productivity as of late. It seems the foundations I created are holding up fairly well, and this means I can add new features easily.

Next week I will be working on adding user-defined functions (lamdas) to Termcat. That should be fun!