Merge pull request #213 from HigherOrderCO/feature/sc-441/add-string-…

…sugar-in-patterns

[sc-441] Add string patterns

cspell.json

@@ -42,6 +42,7 @@
     "namegen",
     "nams",
     "numop",
+    "nums",
     "oper",
     "opre",
     "oprune",

src/term/builtins.rs

@@ -1,4 +1,4 @@
-use super::{parser::parse_book, Book, Name, Pattern, Term};
+use super::{parser::parse_book, Book, Name, NumCtr, Pattern, Term};
 
 const BUILTINS: &str = include_str!(concat!(env!("CARGO_MANIFEST_DIR"), "/src/term/builtins.hvm"));
 
@@ -80,6 +80,7 @@ impl Pattern {
   pub fn encode_builtins(&mut self) {
     match self {
       Pattern::Lst(pats) => *self = Self::encode_list(std::mem::take(pats)),
+      Pattern::Str(str) => *self = Self::encode_str(str),
       Pattern::Ctr(_, pats) => {
         for pat in pats {
           pat.encode_builtins();
@@ -94,11 +95,20 @@ impl Pattern {
   }
 
   fn encode_list(elements: Vec<Pattern>) -> Pattern {
-    let lnil = Pattern::Var(Some(Name::from(LNIL)));
+    let lnil = Pattern::Ctr(Name::from(LNIL), vec![]);
 
     elements.into_iter().rfold(lnil, |acc, mut nxt| {
       nxt.encode_builtins();
       Pattern::Ctr(Name::from(LCONS), vec![nxt, acc])
     })
   }
+
+  fn encode_str(str: &str) -> Pattern {
+    let lnil = Pattern::Ctr(Name::from(SNIL), vec![]);
+
+    str.chars().rfold(lnil, |tail, head| {
+      let head = Pattern::Num(NumCtr::Num(head as u64));
+      Pattern::Ctr(Name::from(SCONS), vec![head, tail])
+    })
+  }
 }

src/term/check/ctrs_arities.rs

@@ -66,7 +66,7 @@ impl Pattern {
             to_check.push(el);
           }
         }
-        Pattern::Var(..) | Pattern::Num(..) => {}
+        Pattern::Var(..) | Pattern::Num(..) | Pattern::Str(_) => {}
       }
     }
     Ok(())

src/term/check/unbound_pats.rs

@@ -58,7 +58,7 @@ impl Pattern {
           check.push(snd);
         }
         Pattern::Lst(args) => args.iter().for_each(|arg| check.push(arg)),
-        Pattern::Var(_) | Pattern::Num(_) => {}
+        Pattern::Var(_) | Pattern::Num(_) | Pattern::Str(_) => {}
       }
     }
     unbounds

src/term/display.rs

@@ -111,6 +111,7 @@ impl fmt::Display for Pattern {
       Pattern::Num(num) => write!(f, "{num}"),
       Pattern::Tup(fst, snd) => write!(f, "({}, {})", fst, snd,),
       Pattern::Lst(pats) => write!(f, "[{}]", DisplayJoin(|| pats, ", ")),
+      Pattern::Str(str) => write!(f, "\"{str}\""),
     }
   }
 }

src/term/mod.rs

@@ -147,6 +147,7 @@ pub enum Pattern {
   Num(NumCtr),
   Tup(Box<Pattern>, Box<Pattern>),
   Lst(Vec<Pattern>),
+  Str(GlobalString),
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
@@ -640,6 +641,7 @@ impl Pattern {
           set.push(nam);
         }
         Pattern::Num(_) => {}
+        Pattern::Str(_) => {}
       }
     }
     let mut set = Vec::new();
@@ -664,6 +666,7 @@ impl Pattern {
       Pattern::Num(_) => Box::new([].iter()),
       Pattern::Tup(fst, snd) => Box::new([fst.as_ref(), snd.as_ref()].into_iter()),
       Pattern::Lst(els) => Box::new(els.iter()),
+      Pattern::Str(_) => Box::new([].iter()),
     }
   }
 
@@ -687,6 +690,7 @@ impl Pattern {
       Pattern::Num(NumCtr::Succ(num, _)) => Some(Name::new(format!("{num}+"))),
       Pattern::Tup(_, _) => Some(Name::new("(,)")),
       Pattern::Lst(_) => todo!(),
+      Pattern::Str(_) => todo!(),
     }
   }
 
@@ -702,11 +706,12 @@ impl Pattern {
   pub fn is_simple(&self) -> bool {
     match self {
       Pattern::Var(_) => true,
-      Pattern::Ctr(_, args) | Pattern::Lst(args) => args.iter().all(|arg| matches!(arg, Pattern::Var(_))),
+      Pattern::Ctr(_, args) => args.iter().all(|arg| matches!(arg, Pattern::Var(_))),
       Pattern::Num(_) => true,
       Pattern::Tup(fst, snd) => {
         matches!(fst.as_ref(), Pattern::Var(_)) && matches!(snd.as_ref(), Pattern::Var(_))
       }
+      Pattern::Lst(_) | Pattern::Str(_) => todo!(),
     }
   }
 
@@ -721,6 +726,7 @@ impl Pattern {
       Pattern::Num(NumCtr::Num(_)) => Type::Num,
       Pattern::Num(NumCtr::Succ(n, _)) => Type::NumSucc(*n),
       Pattern::Lst(..) => Type::Adt(builtins::LIST.into()),
+      Pattern::Str(..) => Type::Adt(builtins::STRING.into()),
     }
   }
 
@@ -736,11 +742,7 @@ impl Pattern {
       Pattern::Num(NumCtr::Succ(val, Some(Some(nam)))) => Term::add_num(Term::Var { nam: nam.clone() }, *val),
       Pattern::Num(NumCtr::Succ(_, Some(None))) => Term::Era,
       Pattern::Tup(fst, snd) => Term::Tup { fst: Box::new(fst.to_term()), snd: Box::new(snd.to_term()) },
-      Pattern::Lst(_) => {
-        let mut p = self.clone();
-        p.encode_builtins();
-        p.to_term()
-      }
+      Pattern::Lst(_) | Pattern::Str(_) => todo!(),
     }
   }
 

src/term/parser/parser.rs

@@ -405,26 +405,25 @@ where
       .map(Pattern::Lst)
       .boxed();
 
-    let num = any()
-      .filter(|t| matches!(t, Token::Num(_)))
-      .map(|t| {
-        let Token::Num(n) = t else { unreachable!() };
-        n
-      })
-      .labelled("<Num>");
+    let num_val = any().filter(|t| matches!(t, Token::Num(_))).map(|t| {
+      let Token::Num(n) = t else { unreachable!() };
+      n
+    });
 
-    let num_pat = num.map(|n| Pattern::Num(NumCtr::Num(n)));
+    let num = num_val.map(|n| Pattern::Num(NumCtr::Num(n))).labelled("<Num>");
 
-    let succ_pat = num
+    let succ = num_val
       .then_ignore(just(Token::Add))
       .then(name_or_era().or_not())
       .map(|(num, nam)| Pattern::Num(NumCtr::Succ(num, nam)))
       .labelled("<Num>+")
       .boxed();
 
-    let chr_pat = select!(Token::Char(c) => Pattern::Num(NumCtr::Num(c))).labelled("<Char>").boxed();
+    let chr = select!(Token::Char(c) => Pattern::Num(NumCtr::Num(c))).labelled("<Char>").boxed();
+
+    let str = select!(Token::Str(s) => Pattern::Str(s)).labelled("<String>").boxed();
 
-    choice((succ_pat, num_pat, chr_pat, var, ctr, list, tup))
+    choice((succ, num, chr, str, var, ctr, list, tup))
   })
 }
 

src/term/transform/desugar_implicit_match_binds.rs

@@ -52,7 +52,8 @@ impl Term {
                   *p = Some(Some(Name::new(format!("{nam}-{n}"))));
                 }
                 Pattern::Tup(_, _) => (),
-                Pattern::Lst(..) => unreachable!(),
+                Pattern::Lst(..) => (),
+                Pattern::Str(..) => (),
               }
             }
           }

src/term/transform/resolve_ctrs_in_pats.rs

@@ -36,6 +36,7 @@ impl Pattern {
         }
         Pattern::Var(None) => (),
         Pattern::Num(_) => (),
+        Pattern::Str(_) => (),
         Pattern::Tup(fst, snd) => {
           to_resolve.push(fst);
           to_resolve.push(snd);

tests/golden_tests/run_file/match_str.hvm

@@ -0,0 +1,9 @@
+(is_as "As") = 2
+(is_as "as") = 2
+(is_as "") = 1
+(is_as *) = 0
+
+map f (List.cons x xs) = (List.cons (f x) (map f xs))
+map f [] = []
+
+main = (map is_as ["As" "as" "" "Asd" "qwerty" "AAs"])

tests/golden_tests/simplify_matches/match_str.hvm

@@ -0,0 +1,6 @@
+(is_as "As") = 2
+(is_as "as") = 2
+(is_as "") = 1
+(is_as *) = 0
+
+main = *

tests/snapshots/compile_file__missing_pat.hvm.snap

@@ -2,5 +2,5 @@
 source: tests/golden_tests.rs
 input_file: tests/golden_tests/compile_file/missing_pat.hvm
 ---
-At tests/golden_tests/compile_file/missing_pat.hvm:2:3: found ':' expected '(', '#', '$', <Name>, '[', '{', 'λ', 'let', 'match', '*', '|', <Num>+, <Num>, or <Char>
+At tests/golden_tests/compile_file/missing_pat.hvm:2:3: found ':' expected '(', '#', '$', <Name>, '[', '{', 'λ', 'let', 'match', '*', '|', <Num>+, <Num>, <Char>, or <String>
   2 |   : *

tests/snapshots/run_file__match_str.hvm.snap

@@ -0,0 +1,5 @@
+---
+source: tests/golden_tests.rs
+input_file: tests/golden_tests/run_file/match_str.hvm
+---
+[2, 2, 1, 0, 0, 0]

tests/snapshots/simplify_matches__match_str.hvm.snap

@@ -0,0 +1,7 @@
+---
+source: tests/golden_tests.rs
+input_file: tests/golden_tests/simplify_matches/match_str.hvm
+---
+(is_as) = λa match a { (String.cons b c): (match b { 65: λd (match d { (String.cons f g): λh (match f { 115: λ* λj (match j { (String.cons * *): λ* 0; (String.nil): λ* 2 } *); *: λ* λ* 0 } h g); (String.nil): λ* 0 } *); 97: λhb (match hb { (String.cons jb kb): λlb (match jb { 115: λ* λnb (match nb { (String.cons * *): λ* 0; (String.nil): λ* 2 } *); *: λ* λ* 0 } lb kb); (String.nil): λ* 0 } *); *: λ* 0 } c); (String.nil): 1 }
+
+(main) = *