Fridtjof Siebert

Inner feature inheriting from generic outer feature results in endless loop or recursion in front end

3

This small example ``` > cat a.fz a is b : a is ``` causes an endless loop or recursion in the front end: ``` > timeout 3s ../build/bin/fz a...

This example from #302 ``` loop_example => for i := 1, i + 1 while i ./build//bin/fz loop.fz /home/fridi/fuzion/clean/fuzion.1/loop.fz:6:5: error 1: Duplicate feature declaration i := i + 1 Feature...

`x.this.type` should report an error if `x` is not an outer feature.

2

see this: ``` > cat test_wrong_this.fz test_wrong_this is a is say b.this.type b is say a.this.type c : a is d : b is a b c d unit >...

base library: support for date and time

3

This can be extremely complex due to many special cases in time zones

`no targets for call` during DFA

1

This example inspired by [this Twitter post](https://x.com/ChShersh/status/1804890794689479139) ``` semigroup(T type) is append(a,b T) T => abstract i32_semigroup : semigroup i32 is redef append(a,b i32) => a+b Seq_semigroup(A type) : semigroup...

C backend creates non-functions that do not return a result

2

This small example ``` b := (String.type.from_marray ((marray Any).type.new 1 nil)) _ := type_of b ``` provokes this error in the code created for a type parameter (which is probably...

`Sequence` cleanup: `fold` vs `reduce`, more powerful `scan`

5

`foldf` is redundant with `reduce`, and `fold*` generally provides the same functionality like `reduce*`, so we should think of better names, e.g., - `reduce(R type, init R, f (R,T) ->...

front end mixes up inherited and outer type parameters and causes Incompatible types error

2

This example is similar to the one in #935, but it crashes the interpreter: ``` inherit_outer is eimer(T type) is v T is abstract map(B type, f T -> B)...

AoC issue: Poor DFA performance

2

The Fuzion Advent-of-Code solution from [12 Dec](https://github.com/fridis/fuzion_aoc/blob/main/2023/12/part2.fz) ``` dec12_part2 is input := (io.stdin.with ()->io.buffered.read_lines).val.filter !="" binomial(n, k i64) => for res := i64 1, res * (n+1-i) / i i...

DFA slow, likely for effects

1

Running `tests/catch_postcondition` with the JVM backend takes 24s on my machine, 93% of this time is spent during DFA.