Based on the unobfuscated source, this is a lookup table of some kind for operators, encoded in a string.

  ++   -> tokl=11 tokc=0x1
  --   -> tokl=11 tokc=0xff
  *    -> tokl=1 tokc=0xc1af
  /    -> tokl=1 tokc=0xf9f7
  %    -> tokl=1 tokc=0xf9f7
  +    -> tokl=2 tokc=0xc801
  -    -> tokl=2 tokc=0xd8f7
  <<   -> tokl=3 tokc=0xe0d3
  >>   -> tokl=3 tokc=0xf8d3
  ...
  !    -> tokl=2 tokc=0x4
  *    -> tokl=0 tokc=0x0

For example

  *@R<^1c
  ^^       operator (padded with @ when one character)
    ^^^^   tokc (variable length)
        ^  tokl

T is being assigned a string value that looks like line noise. I assume it's Perl.

It's funny to see the Perl comparison when APL-family languages normally have code that looks like that:

https://github.com/Co-dfns/Co-dfns/tree/master/cmp

Rust is more unreadable to me than Perl, in all fairness.

Haha. Thanks I needed that Monday morning laugh.

It's part of the tokenizer. t contains data in the format of [char 0][char 1][base 64 encoded token id …], whereas the id is as long as the entry is smaller than 'b'. So it's ('++', `#m`), ('--', `%am`), ('*@', 'R<^1c'), and so on. '@' is used as a wildcard. As you can see this part parses operators.

Looks like a type definition in Rust.

If it doesn't end with .unwrap() then it isn't Rust.

true :(

This made me smile :-)

It's looking for operators. That string encodes the operator to look for, and the values of tokl and tokc that will be used if that operator is found.

The first entry looks for "++", and if found sets tokc to 1 ('#' - 'b' + 64) and tokl to 11 ('m' - 'b').

The second entry looks for "--" and if found sets tokc to 255 and tokl to 11.

The third entry looks for "*" and if found sets tokc to 12693263 and tokl to 1. (12693263 is 0xc1af0f, which is an x86 IMUL instruction).

I would like to look at the code after preprocessing, it is too hard to analyze the code with all those #define statements.

First thought would be "opcodes", but on a closer look the operators in there seem to be relevant. Maybe this encodes their meaning/relevant opcode and precedence?

From the website:

``` binary operators, by decreasing priority order: '*' '/' '%', '+' '-', '>>' '<<', '<' '<=' '>' '>=', '==' '!=', '&', '^', '|', '=', '&&', '||'. ```

The unary operators ++ and -- have a higher precedence, so they come first. Then there are "*@R<^1c", "/@%[_[H3c", "%@%[_[H3c" (same as for '/', though), "[email protected]#d" and so.

I did not have time to read the source (my compile finished a minute ago), but maybe you can now figure out the rest? :)

//edit: ha, edflsafoiewq and xashor we're a bit faster and more thorough. But nice to see we agree on the overall idea.

Isn't it obvious? /s