Katherine
Katherine
`aalt` is probably not essential, but ought to be present.
Is this okay? I know some fonts do this on purpose (Zapfino is an extreme example) but I don't think it's intentional for Crimson. Should the Em square be increased?
For Crimson-Roman, the greek glyphs are drawn with a rounded, handwritten script style. These look lovely, of course, but don't go with the style for the rest of the font....
TODO
* there's no reason the data bus can't be 16 bit * really there's no reason why the data can't be analog, too. i could extend the ALU with operations...
RFC7230 defines the extension # which provides lists: ``` 1#element => element *( OWS "," OWS element ) and: #element => [ 1#element ] and for n >= 1 and...
The AST is inconvenient; it has nesting of a forced hierarchy of particular types. In retrospect I think I'd prefer to construct rrd nodes directly. And now the tnode tree...
Here I'm introducing Cargo to the build. It's driven from the existing makefiles. This marks the start of porting to Rust by a piecemeal approach, so we'll have a hybrid...
For the following retest case: ``` ^abc$ +abc -qqq\nabc -abc\nzzz -qqq\nabc\nzzz ``` I believe the IR-generated C is correct: ``` ; bmake -r CC=clang DEBUG=1 ASAN=1 && ./build/bin/retest -O1 -l...
Currently we produce `$unsupported` for various things (e.g. lookahead, lookbehind in pcre) and then error about it. My suggestion instead is that we do lex these correctly, and then error...
This helps a little for rendering out to more natural regexp syntax. For example: ``` ; ./build/bin/re -r pcre -bpl pcre '(?:a|)(?:|xyz)' a?(?:xyz)? ```