feat(ecmascript): String Objects (#112)

* feat(ecmascript): String Objects

* fix clippy

* fix typo

* fix missing file

* stiilll fix wrong file

* feat(ecmascript): RegExp Objects

* chore: Move RegExpHeapData into ecmascript

* chore: Rename vestigal heap/object.rs into object_entry.rs and remove unused functions

Author: aapoalas

nova_vm/src/builtin_strings

@@ -9,6 +9,12 @@
 [object String]
 [object Undefined]
 [Symbol.hasInstance]
+[Symbol.iterator]
+[Symbol.match]
+[Symbol.matchAll]
+[Symbol.replace]
+[Symbol.search]
+[Symbol.split]
 [Symbol.toPrimitive]
 abs
 acos
@@ -42,7 +48,11 @@ caller
 cause
 cbrt
 ceil
+charAt
+charCodeAt
 clz32
+codePointAt
+concat
 constructor
 copyWithin
 cos
@@ -55,11 +65,14 @@ defineProperties
 defineProperty
 description
 done
+dotAll
 E
+endsWith
 entries
 EPSILON
 Error
 EvalError
+exec
 exp
 expm1
 false
@@ -69,20 +82,33 @@ find
 findIndex
 findLast
 findLastIndex
+flags
 flat
 flatMap
 Float32Array
 Float64Array
 floor
 for
 freeze
+fromCharCode
+fromCodePoint
 fromEntries
 fround
 function
 Function
 get [Symbol.species]
 get byteLength
 get description
+get dotAll
+get flags
+get global
+get hasIndices
+get ignoreCase
+get multiline
+get source
+get stricky
+get unicode
+get unicodeSets
 getDate
 getDay
 getFullYear
@@ -106,14 +132,18 @@ getUTCMilliseconds
 getUTCMinutes
 getUTCMonth
 getUTCSeconds
+global
 globalThis
 groupBy
+hasIndices
 hasInstance
 hasOwn
 hasOwnProperty
 hypot
+ignoreCase
 imul
 includes
+indexOf
 Infinity
 Int16Array
 Int32Array
@@ -128,12 +158,15 @@ isNaN
 isPrototypeOf
 isSafeInteger
 isSealed
+isWellFormed
 iterator
 keyFor
 keys
+lastIndexOf
 length
 LN10
 LN2
+localeCompare
 log
 log10
 LOG10E
@@ -152,16 +185,20 @@ message
 min
 MIN_SAFE_INTEGER
 MIN_VALUE
+multiline
 name
 NaN
 NEGATIVE_INFINITY
 next
+normalize
 now
 null
 number
 Number
 object
 Object
+padEnd
+padStart
 parse
 parseFloat
 parseInt
@@ -175,9 +212,13 @@ prototype
 Proxy
 random
 RangeError
+raw
 ReferenceError
 RegExp
+RegExp String Iterator
+repeat
 replace
+replaceAll
 round
 seal
 search
@@ -202,26 +243,35 @@ SharedArrayBuffer
 sign
 sin
 sinh
+slice
+source
 species
 split
 sqrt
 SQRT1_2
 SQRT2
+startsWith
+stricky
 string
 String
+String Iterator
 symbol
 Symbol
 SyntaxError
 tan
 tanh
+test
 toDateString
 toExponential
 toFixed
 toISOString
 toJSON
 toLocaleDateString
+toLocaleLowerCase
 toLocaleString
 toLocaleTimeString
+toLocaleUpperCase
+toLowerCase
 toPrecision
 toPrimitive
 toReversed
@@ -230,7 +280,12 @@ toSpliced
 toString
 toStringTag
 toTimeString
+toUpperCase
 toUTCString
+toWellFormed
+trim
+trimEnd
+trimStart
 true
 trunc
 TypeError
@@ -239,6 +294,8 @@ Uint32Array
 Uint8Array
 Uint8ClampedArray
 undefined
+unicode
+unicodeSets
 unscopables
 URIError
 utc

nova_vm/src/ecmascript/builders/builtin_function_builder.rs

@@ -19,7 +19,7 @@ use super::property_builder::{self, PropertyBuilder};
 pub struct NoPrototype;
 
 #[derive(Clone, Copy)]
-pub struct CreatorPrototype(Object);
+pub struct CreatorPrototype(Option<Object>);
 
 #[derive(Default, Clone, Copy)]
 pub struct NoLength;
@@ -162,7 +162,30 @@ impl<'agent, L, N, B, Pr> BuiltinFunctionBuilder<'agent, NoPrototype, L, N, B, P
             this: self.this,
             object_index,
             realm: self.realm,
-            prototype: CreatorPrototype(prototype),
+            prototype: CreatorPrototype(Some(prototype)),
+            length: self.length,
+            name: self.name,
+            behaviour: self.behaviour,
+            properties: self.properties,
+        }
+    }
+
+    #[must_use]
+    pub fn with_null_prototype(
+        self,
+    ) -> BuiltinFunctionBuilder<'agent, CreatorPrototype, L, N, B, Pr> {
+        let object_index = if self.object_index.is_none() {
+            self.agent.heap.objects.push(None);
+            Some(ObjectIndex::last(&self.agent.heap.objects))
+        } else {
+            self.object_index
+        };
+        BuiltinFunctionBuilder {
+            agent: self.agent,
+            this: self.this,
+            object_index,
+            realm: self.realm,
+            prototype: CreatorPrototype(None),
             length: self.length,
             name: self.name,
             behaviour: self.behaviour,
@@ -518,7 +541,7 @@ impl<'agent>
         assert!(slot.is_none());
         *slot = Some(ObjectHeapData {
             extensible: true,
-            prototype: Some(prototype.0),
+            prototype: prototype.0,
             keys,
             values,
         });

nova_vm/src/ecmascript/builtins.rs

@@ -14,6 +14,8 @@ pub mod error;
 pub(crate) mod fundamental_objects;
 pub(crate) mod numbers_and_dates;
 pub mod ordinary;
+pub(crate) mod regexp;
+pub(crate) mod text_processing;
 
 pub(crate) use arguments::*;
 pub(crate) use array::abstract_operations::*;

nova_vm/src/ecmascript/builtins/fundamental_objects/function_objects/function_prototype.rs

@@ -204,6 +204,7 @@ impl FunctionPrototype {
             Some(this_object_index),
         )
         .with_property_capacity(8)
+        .with_null_prototype()
         // 10.2.4 AddRestrictedFunctionProperties ( F, realm )
         .with_property(|builder| {
             builder

nova_vm/src/ecmascript/builtins/numbers_and_dates/date_objects/date_prototype.rs

@@ -577,7 +577,7 @@ impl DatePrototype {
         let date_constructor = intrinsics.date();
 
         OrdinaryObjectBuilder::new_intrinsic_object(agent, realm, this)
-            .with_property_capacity(7)
+            .with_property_capacity(45)
             .with_constructor_property(date_constructor)
             .with_builtin_function_property::<DatePrototypeGetDate>()
             .with_builtin_function_property::<DatePrototypeGetDay>()

nova_vm/src/ecmascript/builtins/regexp.rs

@@ -0,0 +1,7 @@
+use crate::heap::indexes::ObjectIndex;
+
+#[derive(Debug, Clone, Copy)]
+pub struct RegExpHeapData {
+    pub(crate) object_index: ObjectIndex,
+    // _regex: RegExp,
+}

nova_vm/src/ecmascript/builtins/text_processing.rs

@@ -0,0 +1,2 @@
+pub(crate) mod regexp_objects;
+pub(crate) mod string_objects;

nova_vm/src/ecmascript/builtins/text_processing/regexp_objects.rs

@@ -0,0 +1,3 @@
+pub(crate) mod regexp_constructor;
+pub(crate) mod regexp_prototype;
+pub(crate) mod regexp_string_iterator_prototype;

nova_vm/src/ecmascript/builtins/text_processing/regexp_objects/regexp_constructor.rs

@@ -0,0 +1,75 @@
+use crate::ecmascript::builders::builtin_function_builder::BuiltinFunctionBuilder;
+use crate::ecmascript::builtins::ArgumentsList;
+use crate::ecmascript::builtins::Behaviour;
+use crate::ecmascript::builtins::Builtin;
+use crate::ecmascript::execution::Agent;
+use crate::ecmascript::execution::JsResult;
+use crate::ecmascript::execution::RealmIdentifier;
+use crate::ecmascript::types::IntoFunction;
+use crate::ecmascript::types::IntoObject;
+use crate::ecmascript::types::Object;
+use crate::ecmascript::types::String;
+use crate::ecmascript::types::Value;
+use crate::ecmascript::types::BUILTIN_STRING_MEMORY;
+use crate::heap::WellKnownSymbolIndexes;
+
+pub struct RegExpConstructor;
+
+impl Builtin for RegExpConstructor {
+    const BEHAVIOUR: Behaviour = Behaviour::Constructor(Self::behaviour);
+    const LENGTH: u8 = 1;
+    const NAME: String = BUILTIN_STRING_MEMORY.RegExp;
+}
+
+struct RegExpSpecies;
+impl Builtin for RegExpSpecies {
+    const BEHAVIOUR: Behaviour = Behaviour::Regular(RegExpConstructor::species);
+    const LENGTH: u8 = 0;
+    const NAME: String = BUILTIN_STRING_MEMORY.get__Symbol_species_;
+}
+impl RegExpConstructor {
+    fn behaviour(
+        _agent: &mut Agent,
+        _this_value: Value,
+        _arguments: ArgumentsList,
+        _new_target: Option<Object>,
+    ) -> JsResult<Value> {
+        todo!();
+    }
+
+    fn species(
+        _agent: &mut Agent,
+        _this_value: Value,
+        _arguments: ArgumentsList,
+    ) -> JsResult<Value> {
+        todo!();
+    }
+
+    pub(crate) fn create_intrinsic(agent: &mut Agent, realm: RealmIdentifier) {
+        let intrinsics = agent.get_realm(realm).intrinsics();
+        let regexp_prototype = intrinsics.reg_exp_prototype();
+        let this = intrinsics.reg_exp();
+        let this_object_index = intrinsics.reg_exp_base_object();
+
+        BuiltinFunctionBuilder::new_intrinsic_constructor::<RegExpConstructor>(
+            agent,
+            realm,
+            this,
+            Some(this_object_index),
+        )
+        .with_property_capacity(2)
+        .with_prototype_property(regexp_prototype.into_object())
+        .with_property(|builder| {
+            builder
+                .with_key(WellKnownSymbolIndexes::Species.into())
+                .with_getter(|agent| {
+                    BuiltinFunctionBuilder::new::<RegExpSpecies>(agent, realm)
+                        .build()
+                        .into_function()
+                })
+                .with_enumerable(false)
+                .build()
+        })
+        .build();
+    }
+}