refactor(regex): remove symbol map lower bound

Author: hackwaly

bytes/internal/regex_engine/compile.mbt

@@ -27,7 +27,7 @@ pub fn compile(profile~ : Profile, ast : Pattern) -> Regex {
   }
   let symbol_map = @symbol_map.new()
   symbolize(profile~, symbol_map~, ast)
-  let (symbol_table, symbol_repr) = symbol_map.finalize(profile.lb, profile.ub)
+  let (symbol_table, symbol_repr) = symbol_map.finalize(profile.ub)
   let ctx = @automata.Context::new()
   let tc = TranslateContext::new(ctx, symbol_table)
   let (expr, pref) = translate(tc, ast)

bytes/internal/regex_engine/symbol_map/dense_table.mbt

@@ -13,9 +13,9 @@
 // limitations under the License.
 
 ///|
-priv struct DenseTable(@shared_types.Rechar, FixedArray[@shared_types.Rechar])
+priv struct DenseTable(FixedArray[@shared_types.Rechar])
 
 ///|
 impl Table for DenseTable with map(self, c) {
-  self.1.unsafe_get(c - self.0)
+  self.0.unsafe_get(c)
 }

bytes/internal/regex_engine/symbol_map/pkg.generated.mbti

@@ -12,7 +12,7 @@ pub fn new() -> SymbolMap
 
 // Types and methods
 type SymbolMap
-pub fn SymbolMap::finalize(Self, Int, Int) -> (&Table, ReadOnlyArray[Int])
+pub fn SymbolMap::finalize(Self, Int) -> (&Table, ReadOnlyArray[Int])
 pub fn SymbolMap::split(Self, @rechar_set.RecharSet) -> Unit
 
 // Type aliases

bytes/internal/regex_engine/symbol_map/symbol_map.mbt

@@ -40,45 +40,42 @@ pub fn SymbolMap::split(
 /// Function `finalize`.
 pub fn SymbolMap::finalize(
   self : SymbolMap,
-  lb : @shared_types.Rechar,
   ub : @shared_types.Rechar,
 ) -> (&Table, ReadOnlyArray[@shared_types.Rechar]) {
-  if ub - lb < 1024 {
-    let (table, repr) = self.finalize_dense(lb, ub)
+  if ub < 1024 {
+    let (table, repr) = self.finalize_dense(ub)
     (table as &Table, repr)
   } else {
-    let (table, repr) = self.finalize_sparse(lb, ub)
+    let (table, repr) = self.finalize_sparse(ub)
     (table as &Table, repr)
   }
 }
 
 ///|
 fn SymbolMap::finalize_dense(
   self : SymbolMap,
-  lb : @shared_types.Rechar,
   ub : @shared_types.Rechar,
 ) -> (DenseTable, ReadOnlyArray[@shared_types.Rechar]) {
   let repr = []
-  let table = FixedArray::make(ub - lb + 1, 0)
-  self.each_intervals(lb, ub, (lo, hi) => {
+  let table = FixedArray::make(ub + 1, 0)
+  self.each_intervals(0, ub, (lo, hi) => {
     let symbol = repr.length()
     repr.push(lo)
     for c in lo..<=hi {
-      table[c - lb] = symbol
+      table[c] = symbol
     }
   })
-  (DenseTable(lb, table), ReadOnlyArray::from_array(repr))
+  (DenseTable(table), ReadOnlyArray::from_array(repr))
 }
 
 ///|
 fn SymbolMap::finalize_sparse(
   self : SymbolMap,
-  lb : @shared_types.Rechar,
   ub : @shared_types.Rechar,
 ) -> (SparseTable, ReadOnlyArray[@shared_types.Rechar]) {
   let repr = []
   let entries = []
-  self.each_intervals(lb, ub, (lo, hi) => {
+  self.each_intervals(0, ub, (lo, hi) => {
     let symbol = repr.length()
     repr.push(lo)
     entries.push({ lo, hi, symbol })

bytes/internal/regex_engine/symbol_map/symbol_map_test.mbt

@@ -13,12 +13,12 @@
 // limitations under the License.
 
 ///|
-test "dense table respects non-zero lower bound" {
+test "dense table uses absolute indexes" {
   let symbol_map = @symbol_map.new()
   symbol_map.split(@shared_types.RecharSet::char(11))
-  let (table, repr) = symbol_map.finalize(10, 12)
+  let (table, repr) = symbol_map.finalize(12)
   assert_eq(repr.length(), 3)
-  assert_eq(repr[0], 10)
+  assert_eq(repr[0], 0)
   assert_eq(repr[1], 11)
   assert_eq(repr[2], 12)
   assert_eq(table.map(10), 0)

string/internal/regex_engine/compile.mbt

@@ -27,7 +27,7 @@ pub fn compile(profile~ : Profile, ast : Pattern) -> Regex {
   }
   let symbol_map = @symbol_map.new()
   symbolize(profile~, symbol_map~, ast)
-  let (symbol_table, symbol_repr) = symbol_map.finalize(profile.lb, profile.ub)
+  let (symbol_table, symbol_repr) = symbol_map.finalize(profile.ub)
   let ctx = @automata.Context::new()
   let tc = TranslateContext::new(ctx, symbol_table)
   let (expr, pref) = translate(tc, ast)

string/internal/regex_engine/symbol_map/pkg.generated.mbti

@@ -12,7 +12,7 @@ pub fn new() -> SymbolMap
 
 // Types and methods
 type SymbolMap
-pub fn SymbolMap::finalize(Self, Int, Int) -> (Table, ReadOnlyArray[Int])
+pub fn SymbolMap::finalize(Self, Int) -> (Table, ReadOnlyArray[Int])
 pub fn SymbolMap::split(Self, @rechar_set.RecharSet) -> Unit
 
 type Table

string/internal/regex_engine/symbol_map/symbol_map.mbt

@@ -40,32 +40,29 @@ pub fn SymbolMap::split(
 /// Function `finalize`.
 pub fn SymbolMap::finalize(
   self : SymbolMap,
-  lb : @shared_types.Rechar,
   ub : @shared_types.Rechar,
 ) -> (Table, ReadOnlyArray[@shared_types.Rechar]) {
-  // Most cases lb is 0
-  guard lb >= 0 else { panic() }
   let repr = []
   let table = Array::make(128, -1)
   if ub < 128 {
     self.each_intervals(0, ub, (lo, hi) => {
       let symbol = repr.length()
-      repr.push(Int::max(lo, lb))
+      repr.push(lo)
       for c in lo..<=hi {
         table[c] = symbol
       }
     })
   } else {
     self.each_intervals(0, 127, (lo, hi) => {
       let symbol = repr.length()
-      repr.push(Int::max(lo, lb))
+      repr.push(lo)
       for c in lo..<=hi {
         table[c] = symbol
       }
     })
     self.each_intervals(128, ub, (lo, hi) => {
       let symbol = repr.length()
-      repr.push(Int::max(lo, lb))
+      repr.push(lo)
       // (lo, hi, symbol)
       table.push(lo)
       table.push(hi)