feat(users/Profpatsch/netencode): Add initial Haskell parser r/5632

A simple categorical parser that does not implement Monad, and does
not contain an `m` and some rudementary error message handling.

In the future I’d probably want to wrap everything in an additional
`m`, so that subparsers can somehow use `Selective` to throw errors
from within `m` that contain the parsing context if at all possible.
Hard to do without Monad, I have to say. Not even stuff like `StateT`
works without the inner `m` implementing `Monad`.

Change-Id: I1366eda606ddfb019637b09c82d8b0e30bd4e318
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7797
Tested-by: BuildkiteCI
Reviewed-by: Profpatsch <mail@profpatsch.de>

1 files changed, 6 insertions, 1 deletions

diff --git a/users/Profpatsch/netencode/netencode.cabal b/users/Profpatsch/netencode/netencode.cabal
index 1bd1d6052a..4e418d6dd8 100644
--- a/users/Profpatsch/netencode/netencode.cabal
+++ b/users/Profpatsch/netencode/netencode.cabal
@@ -5,7 +5,9 @@ author:             Profpatsch
 maintainer:         mail@profpatsch.de
 
 library
-    exposed-modules:          Netencode
+    exposed-modules:
+        Netencode,
+        Netencode.Parse
 
     build-depends:
         base >=4.15 && <5,
@@ -16,5 +18,8 @@ library
         data-fix,
         bytestring,
         attoparsec,
+        text,
+        semigroupoids,
+        selective
 
     default-language: Haskell2010