CSY/reowolf Changeset - 19b752d30f04 · Centrum Wiskunde & Informatica (CWI)

Changeset - 19b752d30f04

Parent rev.

Child rev.

[Not reviewed]

0 5 0

MH - 4 years ago 2021-04-02 14:56:59
contact@maxhenger.nl

cleanup old NamespacedIdentifier

5 files changed with 71 insertions and 209 deletions:

src/protocol/ast.rs

138

src/protocol/lexer.rs

src/protocol/parser/symbol_table.rs

src/protocol/parser/utils.rs

src/protocol/parser/visitor_linker.rs

0 comments (0 inline, 0 general)

src/protocol/ast.rs

➞

Show inline comments

@@ @@ -536,815 +536,695 @@ pub struct PragmaModule { @@
     pub position: InputPosition,
     pub value: Vec<u8>,
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct PragmaOld {
     pub this: PragmaId,
     // Phase 1: parser
     pub position: InputPosition,
     pub value: Vec<u8>,
+}
 impl SyntaxElement for PragmaOld {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum Import {
     Module(ImportModule),
     Symbols(ImportSymbols)
+}
 impl Import {
     pub(crate) fn as_module(&self) -> &ImportModule {
         match self {
             Import::Module(m) => m,
             _ => panic!("Unable to cast 'Import' to 'ImportModule'")
+        }
+    }
     pub(crate) fn as_symbols(&self) -> &ImportSymbols {
         match self {
             Import::Symbols(m) => m,
             _ => panic!("Unable to cast 'Import' to 'ImportSymbols'")
+        }
+    }
+}
 impl SyntaxElement for Import {
     fn position(&self) -> InputPosition {
         match self {
             Import::Module(m) => m.position,
             Import::Symbols(m) => m.position
+        }
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ImportModule {
     pub this: ImportId,
     // Phase 1: parser
     pub position: InputPosition,
     pub module_name: Vec<u8>,
     pub alias: Identifier,
     // Phase 2: module resolving
     pub module_id: Option<RootId>,
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct AliasedSymbol {
     // Phase 1: parser
     pub position: InputPosition,
     pub name: Identifier,
     pub alias: Identifier,
     // Phase 2: symbol resolving
     pub definition_id: Option<DefinitionId>,
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ImportSymbols {
     pub this: ImportId,
     // Phase 1: parser
     pub position: InputPosition,
     pub module_name: Vec<u8>,
     // Phase 2: module resolving
     pub module_id: Option<RootId>,
     // Phase 1&2
     // if symbols is empty, then we implicitly import all symbols without any
     // aliases for them. If it is not empty, then symbols are explicitly
     // specified, and optionally given an alias.
     pub symbols: Vec<AliasedSymbol>,
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct Identifier {
     pub position: InputPosition,
     pub value: Vec<u8>
+}
 impl PartialEq for Identifier {
     fn eq(&self, other: &Self) -> bool {
         return self.value == other.value
+    }
+}
 impl PartialEq<NamespacedIdentifier> for Identifier {
     fn eq(&self, other: &NamespacedIdentifier) -> bool {
         return self.value == other.value
 impl Display for Identifier {
     fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
         // A source identifier is in ASCII range.
         write!(f, "{}", String::from_utf8_lossy(&self.value))
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum NamespacedIdentifierPart {
     // Regular identifier
     Identifier{start: u16, end: u16},
     // Polyargs associated with a preceding identifier
     PolyArgs{start: u16, end: u16},
+}
 impl NamespacedIdentifierPart {
     pub(crate) fn is_identifier(&self) -> bool {
         match self {
             NamespacedIdentifierPart::Identifier{..} => true,
             NamespacedIdentifierPart::PolyArgs{..} => false,
+        }
+    }
     pub(crate) fn as_identifier(&self) -> (u16, u16) {
         match self {
             NamespacedIdentifierPart::Identifier{start, end} => (*start, *end),
             NamespacedIdentifierPart::PolyArgs{..} => {
                 unreachable!("Tried to obtain {:?} as Identifier", self);
+            }
+        }
+    }
     pub(crate) fn as_poly_args(&self) -> (u16, u16) {
         match self {
             NamespacedIdentifierPart::PolyArgs{start, end} => (*start, *end),
             NamespacedIdentifierPart::Identifier{..} => {
                 unreachable!("Tried to obtain {:?} as PolyArgs", self)
+            }
+        }
+    }
+}
 /// An identifier with optional namespaces and polymorphic variables. Note that
 /// we allow each identifier to be followed by polymorphic arguments during the
 /// parsing phase (e.g. Foo<A,B>::Bar<C,D>::Qux). But in our current language
 /// implementation we can only have valid namespaced identifier that contain one
 /// set of polymorphic arguments at the appropriate position.
 /// TODO: @tokens Reimplement/rename once we have a tokenizer
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
-pub struct NamespacedIdentifier2 {
+pub struct NamespacedIdentifier {
     pub position: InputPosition,
     pub value: Vec<u8>, // Full name as it resides in the input source
     pub poly_args: Vec<ParserTypeId>, // All poly args littered throughout the namespaced identifier
     pub parts: Vec<NamespacedIdentifierPart>, // Indices into value/poly_args
+}
-impl NamespacedIdentifier2 {
+impl NamespacedIdentifier {
     /// Returns the identifier value without any of the specific polymorphic
     /// arguments.
     pub fn strip_poly_args(&self) -> Vec<u8> {
         debug_assert!(!self.parts.is_empty() && self.parts[0].is_identifier());
         let mut result = Vec::with_capacity(self.value.len());
         let mut iter = self.iter();
         let (first_ident, _) = iter.next().unwrap();
         result.extend(first_ident);
         for (ident, _) in iter.next() {
             result.push(b':');
             result.push(b':');
             result.extend(ident);
+        }
         result
+    }
     /// Returns an iterator of the elements in the namespaced identifier
     pub fn iter(&self) -> NamespacedIdentifier2Iter {
         return NamespacedIdentifier2Iter{
     pub fn iter(&self) -> NamespacedIdentifierIter {
         return NamespacedIdentifierIter{
             identifier: self,
             element_idx: 0
+        }
+    }
     pub fn get_poly_args(&self) -> Option<&[ParserTypeId]> {
         let has_poly_args = self.parts.iter().any(|v| !v.is_identifier());
         if has_poly_args {
             Some(&self.poly_args)
         } else {
             None
+        }
+    }
     // Check if two namespaced identifiers match eachother when not considering
     // the polymorphic arguments
     pub fn matches_namespaced_identifier(&self, other: &Self) -> bool {
         let mut iter_self = self.iter();
         let mut iter_other = other.iter();
         loop {
             let val_self = iter_self.next();
             let val_other = iter_other.next();
             if val_self.is_some() != val_other.is_some() {
                 // One is longer than the other
                 return false;
+            }
             if val_self.is_none() {
                 // Both are none
                 return true;
+            }
             // Both are something
             let (val_self, _) = val_self.unwrap();
             let (val_other, _) = val_other.unwrap();
             if val_self != val_other { return false; }
+        }
+    }
     // Check if the namespaced identifier matches an identifier when not
     // considering the polymorphic arguments
     pub fn matches_identifier(&self, other: &Identifier) -> bool {
         let mut iter = self.iter();
         let (first_ident, _) = iter.next().unwrap();
         if first_ident != other.value {
             return false;
+        }
         if iter.next().is_some() {
             return false;
+        }
         return true;
+    }
+}
 /// Iterator over elements of the namespaced identifier. The element index will
 /// only ever be at the start of an identifier element.
 #[derive(Debug)]
 pub struct NamespacedIdentifier2Iter<'a> {
     identifier: &'a NamespacedIdentifier2,
 pub struct NamespacedIdentifierIter<'a> {
     identifier: &'a NamespacedIdentifier,
     element_idx: usize,
+}
-impl<'a> Iterator for NamespacedIdentifier2Iter<'a> {
+impl<'a> Iterator for NamespacedIdentifierIter<'a> {
     type Item = (&'a [u8], Option<&'a [ParserTypeId]>);
     fn next(&mut self) -> Option<Self::Item> {
         match self.get(self.element_idx) {
             Some((ident, poly)) => {
                 self.element_idx += 1;
                 if poly.is_some() {
                     self.element_idx += 1;
+                }
                 Some((ident, poly))
             },
             None => None
+        }
+    }
+}
-impl<'a> NamespacedIdentifier2Iter<'a> {
+impl<'a> NamespacedIdentifierIter<'a> {
     /// Returns number of parts iterated over, may not correspond to number of
     /// times one called `next()` because returning an identifier with
     /// polymorphic arguments increments the internal counter by 2.
     pub fn num_returned(&self) -> usize {
         return self.element_idx;
+    }
     pub fn num_remaining(&self) -> usize {
         return self.identifier.parts.len() - self.element_idx;
+    }
     pub fn returned_section(&self) -> &[u8] {
         if self.element_idx == 0 { return &self.identifier.value[0..0]; }
         let last_idx = match &self.identifier.parts[self.element_idx - 1] {
             NamespacedIdentifierPart::Identifier{end, ..} => *end,
             NamespacedIdentifierPart::PolyArgs{end, ..} => *end,
         };
         return &self.identifier.value[..last_idx as usize];
+    }
     /// Returns a specific element from the namespaced identifier
     pub fn get(&self, idx: usize) -> Option<<Self as Iterator>::Item> {
         if idx >= self.identifier.parts.len() {
             return None
+        }
         let cur_part = &self.identifier.parts[idx];
         let next_part = self.identifier.parts.get(idx + 1);
         let (ident_start, ident_end) = cur_part.as_identifier();
         let poly_slice = match next_part {
             Some(part) => match part {
                 NamespacedIdentifierPart::Identifier{..} => None,
                 NamespacedIdentifierPart::PolyArgs{start, end} => Some(
                     &self.identifier.poly_args[*start as usize..*end as usize]
                 ),
             },
             None => None
         };
         Some((
             &self.identifier.value[ident_start as usize..ident_end as usize],
             poly_slice
         ))
+    }
     /// Returns the previously returend index into the parts array of the
     /// identifier.
     pub fn prev_idx(&self) -> Option<usize> {
         if self.element_idx == 0 {
             return None;
         };
         if self.identifier.parts[self.element_idx - 1].is_identifier() {
             return Some(self.element_idx - 1);
+        }
         // Previous part had polymorphic arguments, so the one before that must
         // be an identifier (if well formed)
         debug_assert!(self.element_idx >= 2 && self.identifier.parts[self.element_idx - 2].is_identifier());
         return Some(self.element_idx - 2)
+    }
     /// Returns the previously returned result from `next()`
     pub fn prev(&self) -> Option<<Self as Iterator>::Item> {
         match self.prev_idx() {
             None => None,
             Some(idx) => self.get(idx)
+        }
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct NamespacedIdentifier {
     pub position: InputPosition,
     pub num_namespaces: u8,
     pub value: Vec<u8>,
+}
 impl NamespacedIdentifier {
     pub(crate) fn iter(&self) -> NamespacedIdentifierIter {
         NamespacedIdentifierIter{
             value: &self.value,
             cur_offset: 0,
             num_returned: 0,
             num_total: self.num_namespaces
+        }
+    }
+}
 impl PartialEq for NamespacedIdentifier {
     fn eq(&self, other: &Self) -> bool {
         return self.value == other.value
+    }
+}
 impl PartialEq<Identifier> for NamespacedIdentifier {
     fn eq(&self, other: &Identifier) -> bool {
         return self.value == other.value;
+    }
+}
 // TODO: Just keep ref to NamespacedIdentifier
 pub(crate) struct NamespacedIdentifierIter<'a> {
     value: &'a Vec<u8>,
     cur_offset: usize,
     num_returned: u8,
     num_total: u8,
+}
 impl<'a> NamespacedIdentifierIter<'a> {
     pub(crate) fn num_returned(&self) -> u8 {
         return self.num_returned;
+    }
     pub(crate) fn num_remaining(&self) -> u8 {
         return self.num_total - self.num_returned
+    }
     pub(crate) fn returned_section(&self) -> &[u8] {
         // Offset always includes the two trailing ':' characters
         let end = if self.cur_offset >= 2 { self.cur_offset - 2 } else { self.cur_offset };
         return &self.value[..end]
+    }
+}
 impl<'a> Iterator for NamespacedIdentifierIter<'a> {
     type Item = &'a [u8];
     fn next(&mut self) -> Option<Self::Item> {
         if self.cur_offset >= self.value.len() {
             debug_assert_eq!(self.num_returned, self.num_total);
             None
         } else {
             debug_assert!(self.num_returned < self.num_total);
             let start = self.cur_offset;
             let mut end = start;
             while end < self.value.len() - 1 {
                 if self.value[end] == b':' && self.value[end + 1] == b':' {
                     self.cur_offset = end + 2;
                     self.num_returned += 1;
                     return Some(&self.value[start..end]);
+                }
                 end += 1;
+            }
             // If NamespacedIdentifier is constructed properly, then we cannot
             // end with "::" in the value, so
             debug_assert!(end == 0 || (self.value[end - 1] != b':' && self.value[end] != b':'));
             debug_assert_eq!(self.num_returned + 1, self.num_total);
             self.cur_offset = self.value.len();
             self.num_returned += 1;
             return Some(&self.value[start..]);
+        }
+    }
+}
 impl Display for Identifier {
     fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
         // A source identifier is in ASCII range.
         write!(f, "{}", String::from_utf8_lossy(&self.value))
+    }
+}
 /// TODO: @types Remove the Message -> Byte hack at some point...
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum ParserTypeVariant {
     // Basic builtin
     Message,
     Bool,
     Byte,
     Short,
     Int,
     Long,
     String,
     // Literals (need to get concrete builtin type during typechecking)
     IntegerLiteral,
     Inferred,
     // Complex builtins
     Array(ParserTypeId), // array of a type
     Input(ParserTypeId), // typed input endpoint of a channel
     Output(ParserTypeId), // typed output endpoint of a channel
     Symbolic(SymbolicParserType), // symbolic type (definition or polyarg)
+}
 impl ParserTypeVariant {
     pub(crate) fn supports_polymorphic_args(&self) -> bool {
         use ParserTypeVariant::*;
         match self {
             Message | Bool | Byte | Short | Int | Long | String | IntegerLiteral | Inferred => false,
             _ => true
+        }
+    }
+}
 /// ParserType is a specification of a type during the parsing phase and initial
 /// linker/validator phase of the compilation process. These types may be
 /// (partially) inferred or represent literals (e.g. a integer whose bytesize is
 /// not yet determined).
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ParserType {
     pub this: ParserTypeId,
     pub pos: InputPosition,
     pub variant: ParserTypeVariant,
+}
 /// SymbolicParserType is the specification of a symbolic type. During the
 /// parsing phase we will only store the identifier of the type. During the
 /// validation phase we will determine whether it refers to a user-defined type,
 /// or a polymorphic argument. After the validation phase it may still be the
 /// case that the resulting `variant` will not pass the typechecker.
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct SymbolicParserType {
     // Phase 1: parser
-    pub identifier: NamespacedIdentifier2,
+    pub identifier: NamespacedIdentifier,
     // Phase 2: validation/linking (for types in function/component bodies) and
     //  type table construction (for embedded types of structs/unions)
     pub poly_args2: Vec<ParserTypeId>, // taken from identifier or inferred
     pub variant: Option<SymbolicParserTypeVariant>
+}
 /// Specifies whether the symbolic type points to an actual user-defined type,
 /// or whether it points to a polymorphic argument within the definition (e.g.
 /// a defined variable `T var` within a function `int func<T>()`
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum SymbolicParserTypeVariant {
     Definition(DefinitionId),
     // TODO: figure out if I need the DefinitionId here
     PolyArg(DefinitionId, usize), // index of polyarg in the definition
+}
 /// ConcreteType is the representation of a type after resolving symbolic types
 /// and performing type inference
 #[derive(Debug, Clone, Copy, Eq, PartialEq, serde::Serialize, serde::Deserialize)]
 pub enum ConcreteTypePart {
     // Markers for the use of polymorphic types within a procedure's body that
     // refer to polymorphic variables on the procedure's definition. Different
     // from markers in the `InferenceType`, these will not contain nested types.
     Marker(usize),
     // Special types (cannot be explicitly constructed by the programmer)
     Void,
     // Builtin types without nested types
     Message,
     Bool,
     Byte,
     Short,
     Int,
     Long,
     String,
     // Builtin types with one nested type
     Array,
     Slice,
     Input,
     Output,
     // User defined type with any number of nested types
     Instance(DefinitionId, usize),
+}
 #[derive(Debug, Clone, Eq, PartialEq, serde::Serialize, serde::Deserialize)]
 pub struct ConcreteType {
     pub(crate) parts: Vec<ConcreteTypePart>
+}
 impl Default for ConcreteType {
     fn default() -> Self {
         Self{ parts: Vec::new() }
+    }
+}
 impl ConcreteType {
     pub(crate) fn has_marker(&self) -> bool {
         self.parts
             .iter()
             .any(|p| {
                 if let ConcreteTypePart::Marker(_) = p { true } else { false }
             })
+    }
+}
 // TODO: Remove at some point
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 pub enum PrimitiveType {
     Unassigned,
     Input,
     Output,
     Message,
     Boolean,
     Byte,
     Short,
     Int,
     Long,
     Symbolic(PrimitiveSymbolic)
+}
 // TODO: @cleanup, remove PartialEq implementations
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct PrimitiveSymbolic {
     // Phase 1: parser
     pub(crate) identifier: NamespacedIdentifier, // TODO: @remove at some point, also remove NSIdent itself
     // Phase 2: typing
     pub(crate) definition: Option<DefinitionId>
+}
 impl PartialEq for PrimitiveSymbolic {
     fn eq(&self, other: &Self) -> bool {
         self.identifier == other.identifier
+    }
+}
 impl Eq for PrimitiveSymbolic{}
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 pub struct Type {
     pub primitive: PrimitiveType,
     pub array: bool,
+}
 #[allow(dead_code)]
 impl Type {
     pub const UNASSIGNED: Type = Type { primitive: PrimitiveType::Unassigned, array: false };
     pub const INPUT: Type = Type { primitive: PrimitiveType::Input, array: false };
     pub const OUTPUT: Type = Type { primitive: PrimitiveType::Output, array: false };
     pub const MESSAGE: Type = Type { primitive: PrimitiveType::Message, array: false };
     pub const BOOLEAN: Type = Type { primitive: PrimitiveType::Boolean, array: false };
     pub const BYTE: Type = Type { primitive: PrimitiveType::Byte, array: false };
     pub const SHORT: Type = Type { primitive: PrimitiveType::Short, array: false };
     pub const INT: Type = Type { primitive: PrimitiveType::Int, array: false };
     pub const LONG: Type = Type { primitive: PrimitiveType::Long, array: false };
     pub const INPUT_ARRAY: Type = Type { primitive: PrimitiveType::Input, array: true };
     pub const OUTPUT_ARRAY: Type = Type { primitive: PrimitiveType::Output, array: true };
     pub const MESSAGE_ARRAY: Type = Type { primitive: PrimitiveType::Message, array: true };
     pub const BOOLEAN_ARRAY: Type = Type { primitive: PrimitiveType::Boolean, array: true };
     pub const BYTE_ARRAY: Type = Type { primitive: PrimitiveType::Byte, array: true };
     pub const SHORT_ARRAY: Type = Type { primitive: PrimitiveType::Short, array: true };
     pub const INT_ARRAY: Type = Type { primitive: PrimitiveType::Int, array: true };
     pub const LONG_ARRAY: Type = Type { primitive: PrimitiveType::Long, array: true };
+}
 impl Display for Type {
     fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
         match &self.primitive {
             PrimitiveType::Unassigned => {
                 write!(f, "unassigned")?;
+            }
             PrimitiveType::Input => {
                 write!(f, "in")?;
+            }
             PrimitiveType::Output => {
                 write!(f, "out")?;
+            }
             PrimitiveType::Message => {
                 write!(f, "msg")?;
+            }
             PrimitiveType::Boolean => {
                 write!(f, "boolean")?;
+            }
             PrimitiveType::Byte => {
                 write!(f, "byte")?;
+            }
             PrimitiveType::Short => {
                 write!(f, "short")?;
+            }
             PrimitiveType::Int => {
                 write!(f, "int")?;
+            }
             PrimitiveType::Long => {
                 write!(f, "long")?;
+            }
             PrimitiveType::Symbolic(data) => {
                 // Type data is in ASCII range.
                 if let Some(id) = &data.definition {
                     write!(
                         f, "Symbolic({}, id: {})",
                         String::from_utf8_lossy(&data.identifier.value),
                         id.index
                     )?;
                 } else {
                     write!(
                         f, "Symbolic({}, id: Unresolved)",
                         String::from_utf8_lossy(&data.identifier.value)
                     )?;
+                }
+            }
+        }
         if self.array {
             write!(f, "[]")
         } else {
             Ok(())
+        }
+    }
+}
 type LiteralCharacter = Vec<u8>;
 type LiteralInteger = i64; // TODO: @int_literal
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum Literal {
     Null, // message
     True,
     False,
     Character(LiteralCharacter),
     Integer(LiteralInteger),
     Struct(LiteralStruct),
+}
 impl Literal {
     pub(crate) fn as_struct(&self) -> &LiteralStruct {
         if let Literal::Struct(literal) = self{
             literal
         } else {
             unreachable!("Attempted to obtain {:?} as Literal::Struct", self)
+        }
+    }
     pub(crate) fn as_struct_mut(&mut self) -> &mut LiteralStruct {
         if let Literal::Struct(literal) = self{
             literal
         } else {
             unreachable!("Attempted to obtain {:?} as Literal::Struct", self)
+        }
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct LiteralStructField {
     // Phase 1: parser
     pub(crate) identifier: Identifier,
     pub(crate) value: ExpressionId,
     // Phase 2: linker
     pub(crate) field_idx: usize, // in struct definition
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct LiteralStruct {
     // Phase 1: parser
-    pub(crate) identifier: NamespacedIdentifier2,
+    pub(crate) identifier: NamespacedIdentifier,
     pub(crate) fields: Vec<LiteralStructField>,
     // Phase 2: linker
     pub(crate) poly_args2: Vec<ParserTypeId>, // taken from identifier
     pub(crate) definition: Option<DefinitionId>
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct LiteralEnum {
     // Phase 1: parser
     pub(crate) identifier: NamespacedIdentifier2,
     pub(crate) poly_args: Vec<ParserTypeId>,
     pub(crate) identifier: NamespacedIdentifier,
     // Phase 2: linker
     pub(crate) poly_args2: Vec<ParserTypeId>,
     pub(crate) definition: Option<DefinitionId>,
     pub(crate) variant_idx: usize,
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum Method {
     Get,
     Put,
     Fires,
     Create,
     Symbolic(MethodSymbolic)
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct MethodSymbolic {
-    pub(crate) identifier: NamespacedIdentifier2,
+    pub(crate) identifier: NamespacedIdentifier,
     pub(crate) definition: Option<DefinitionId>
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum Field {
     Length,
     Symbolic(FieldSymbolic),
+}
 impl Field {
     pub fn is_length(&self) -> bool {
         match self {
             Field::Length => true,
             _ => false,
+        }
+    }
     pub fn as_symbolic(&self) -> &FieldSymbolic {
         match self {
             Field::Symbolic(v) => v,
             _ => unreachable!("attempted to get Field::Symbolic from {:?}", self)
+        }
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct FieldSymbolic {
     // Phase 1: Parser
     pub(crate) identifier: Identifier,
     // Phase 3: Typing
     pub(crate) definition: Option<DefinitionId>,
     pub(crate) field_idx: usize,
+}
 #[derive(Debug, Clone, Copy, serde::Serialize, serde::Deserialize)]
 pub enum Scope {
     Definition(DefinitionId),
     Regular(BlockStatementId),
     Synchronous((SynchronousStatementId, BlockStatementId)),
+}
 impl Scope {
     pub fn is_block(&self) -> bool {
         match &self {
             Scope::Definition(_) => false,
             Scope::Regular(_) => true,
             Scope::Synchronous(_) => true,
+        }
+    }
     pub fn to_block(&self) -> BlockStatementId {
         match &self {
             Scope::Regular(id) => *id,
             Scope::Synchronous((_, id)) => *id,
             _ => panic!("unable to get BlockStatement from Scope")
+        }
+    }
+}
 pub trait VariableScope {
     fn parent_scope(&self, h: &Heap) -> Option<Scope>;
     fn get_variable(&self, h: &Heap, id: &Identifier) -> Option<VariableId>;
+}
 impl VariableScope for Scope {
     fn parent_scope(&self, h: &Heap) -> Option<Scope> {
         match self {
             Scope::Definition(def) => h[*def].parent_scope(h),
             Scope::Regular(stmt) => h[*stmt].parent_scope(h),
             Scope::Synchronous((stmt, _)) => h[*stmt].parent_scope(h),
+        }
+    }
     fn get_variable(&self, h: &Heap, id: &Identifier) -> Option<VariableId> {
         match self {
             Scope::Definition(def) => h[*def].get_variable(h, id),
             Scope::Regular(stmt) => h[*stmt].get_variable(h, id),
             Scope::Synchronous((stmt, _)) => h[*stmt].get_variable(h, id),
+        }
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub enum Variable {
     Parameter(Parameter),
     Local(Local),
+}
 impl Variable {
     pub fn identifier(&self) -> &Identifier {
         match self {
             Variable::Parameter(var) => &var.identifier,
             Variable::Local(var) => &var.identifier,
+        }
+    }
     pub fn is_parameter(&self) -> bool {
         match self {
             Variable::Parameter(_) => true,
             _ => false,
@@ @@ -2594,109 +2474,109 @@ pub struct SlicingExpression { @@
     pub position: InputPosition,
     pub subject: ExpressionId,
     pub from_index: ExpressionId,
     pub to_index: ExpressionId,
     // Phase 2: linker
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for SlicingExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct SelectExpression {
     pub this: SelectExpressionId,
     // Phase 1: parser
     pub position: InputPosition,
     pub subject: ExpressionId,
     pub field: Field,
     // Phase 2: linker
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for SelectExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ArrayExpression {
     pub this: ArrayExpressionId,
     // Phase 1: parser
     pub position: InputPosition,
     pub elements: Vec<ExpressionId>,
     // Phase 2: linker
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for ArrayExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct CallExpression {
     pub this: CallExpressionId,
     // Phase 1: parser
     pub position: InputPosition,
     pub method: Method,
     pub arguments: Vec<ExpressionId>,
     pub poly_args: Vec<ParserTypeId>,
     // Phase 2: linker
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for CallExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct LiteralExpression {
     pub this: LiteralExpressionId,
     // Phase 1: parser
     pub position: InputPosition,
     pub value: Literal,
     // Phase 2: linker
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for LiteralExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
 pub struct VariableExpression {
     pub this: VariableExpressionId,
     // Phase 1: parser
     pub position: InputPosition,
-    pub identifier: NamespacedIdentifier2,
+    pub identifier: NamespacedIdentifier,
     // Phase 2: linker
     pub declaration: Option<VariableId>,
     pub parent: ExpressionParent,
     // Phase 3: type checking
     pub concrete_type: ConcreteType,
+}
 impl SyntaxElement for VariableExpression {
     fn position(&self) -> InputPosition {
         self.position
+    }
+}

src/protocol/lexer.rs

➞

Show inline comments

 use crate::protocol::ast::*;
 use crate::protocol::inputsource::*;
 const MAX_LEVEL: usize = 128;
 const MAX_NAMESPACES: u8 = 8; // only three levels are supported at the moment
 macro_rules! debug_log {
     ($format:literal) => {
         enabled_debug_print!(true, "lexer", $format);
     };
     ($format:literal, $($args:expr),*) => {
         enabled_debug_print!(true, "lexer", $format, $($args),*);
     };
+}
 macro_rules! debug_line {
     ($source:expr) => {
+        {
             let mut buffer = String::with_capacity(128);
             for idx in 0..buffer.capacity() {
                 let next = $source.lookahead(idx);
                 if next.is_none() || Some(b'\n') == next { break; }
                 buffer.push(next.unwrap() as char);
+            }
             buffer
+        }
     };
+}
 fn is_vchar(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= 0x21 && c <= 0x7E
     } else {
         false
+    }
+}
 fn is_wsp(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c == b' ' || c == b'\t'
     } else {
         false
+    }
+}
 fn is_ident_start(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= b'A' && c <= b'Z' || c >= b'a' && c <= b'z'
     } else {
         false
+    }
+}
 fn is_ident_rest(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= b'A' && c <= b'Z' || c >= b'a' && c <= b'z' || c >= b'0' && c <= b'9' || c == b'_'
     } else {
         false
+    }
+}
 fn is_constant(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= b'0' && c <= b'9' || c == b'\''
     } else {
         false
+    }
+}
 fn is_integer_start(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= b'0' && c <= b'9'
     } else {
         false
+    }
+}
 fn is_integer_rest(x: Option<u8>) -> bool {
     if let Some(c) = x {
         c >= b'0' && c <= b'9'
             || c >= b'a' && c <= b'f'
             || c >= b'A' && c <= b'F'
             || c == b'x'
             || c == b'o'
     } else {
         false
+    }
+}
 fn lowercase(x: u8) -> u8 {
     if x >= b'A' && x <= b'Z' {
         x - b'A' + b'a'
     } else {
+        x
+    }
+}
-fn identifier_as_namespaced(identifier: Identifier) -> NamespacedIdentifier2 {
+fn identifier_as_namespaced(identifier: Identifier) -> NamespacedIdentifier {
     let identifier_len = identifier.value.len();
     debug_assert!(identifier_len < u16::max_value() as usize);
-    NamespacedIdentifier2{
+    NamespacedIdentifier{
         position: identifier.position,
         value: identifier.value,
         poly_args: Vec::new(),
         parts: vec![
             NamespacedIdentifierPart::Identifier{start: 0, end: identifier_len as u16}
         ],
+    }
+}
 pub struct Lexer<'a> {
     source: &'a mut InputSource,
     level: usize,
+}
 impl Lexer<'_> {
     pub fn new(source: &mut InputSource) -> Lexer {
         Lexer { source, level: 0 }
+    }
     fn error_at_pos(&self, msg: &str) -> ParseError2 {
         ParseError2::new_error(self.source, self.source.pos(), msg)
+    }
     fn consume_line(&mut self) -> Result<Vec<u8>, ParseError2> {
         let mut result: Vec<u8> = Vec::new();
         let mut next = self.source.next();
         while next.is_some() && next != Some(b'\n') && next != Some(b'\r') {
             if !(is_vchar(next) || is_wsp(next)) {
                 return Err(self.error_at_pos("Expected visible character or whitespace"));
+            }
             result.push(next.unwrap());
             self.source.consume();
             next = self.source.next();
+        }
         if next.is_some() {
             self.source.consume();
+        }
         if next == Some(b'\r') && self.source.next() == Some(b'\n') {
             self.source.consume();
+        }
         Ok(result)
+    }
     fn consume_whitespace(&mut self, expected: bool) -> Result<(), ParseError2> {
         let mut found = false;
         let mut next = self.source.next();
         while next.is_some() {
             if next == Some(b' ')
                 || next == Some(b'\t')
                 || next == Some(b'\r')
                 || next == Some(b'\n')
+            {
                 self.source.consume();
                 next = self.source.next();
                 found = true;
                 continue;
+            }
             if next == Some(b'/') {
                 next = self.source.lookahead(1);
                 if next == Some(b'/') {
                     self.source.consume(); // slash
                     self.source.consume(); // slash
                     self.consume_line()?;
                     next = self.source.next();
                     found = true;
                     continue;
+                }
                 if next == Some(b'*') {
                     self.source.consume(); // slash
                     self.source.consume(); // star
                     next = self.source.next();
                     while next.is_some() {
                         if next == Some(b'*') {
                             next = self.source.lookahead(1);
                             if next == Some(b'/') {
                                 self.source.consume(); // star
                                 self.source.consume(); // slash
                                 break;
+                            }
+                        }
                         self.source.consume();
                         next = self.source.next();
+                    }
                     next = self.source.next();
                     found = true;
                     continue;
+                }
+            }
             break;
+        }
         if expected && !found {
             Err(self.error_at_pos("Expected whitespace"))
         } else {
             Ok(())
+        }
+    }
     fn consume_any_chars(&mut self) {
         if !is_ident_start(self.source.next()) { return }
         self.source.consume();
@@ @@ -348,473 +348,459 @@ impl Lexer<'_> { @@
         if data_len == 1 {
             debug_assert!(data[0] >= b'0' && data[0] <= b'9');
             return Ok((data[0] - b'0') as i64);
         } else {
             // TODO: Fix, u64 should be supported as well
             let parsed = if data[1] == b'b' {
                 let data = String::from_utf8_lossy(&data[2..]);
                 i64::from_str_radix(&data, 2)
             } else if data[1] == b'o' {
                 let data = String::from_utf8_lossy(&data[2..]);
                 i64::from_str_radix(&data, 8)
             } else if data[1] == b'x' {
                 let data = String::from_utf8_lossy(&data[2..]);
                 i64::from_str_radix(&data, 16)
             } else {
                 // Assume decimal
                 let data = String::from_utf8_lossy(&data);
                 i64::from_str_radix(&data, 10)
             };
             if let Err(_err) = parsed {
                 return Err(ParseError2::new_error(&self.source, position, "Invalid integer constant"));
+            }
             Ok(parsed.unwrap())
+        }
+    }
     // Statement keywords
     // TODO: Clean up these functions
     fn has_statement_keyword(&self) -> bool {
         self.has_keyword(b"channel")
             || self.has_keyword(b"skip")
             || self.has_keyword(b"if")
             || self.has_keyword(b"while")
             || self.has_keyword(b"break")
             || self.has_keyword(b"continue")
             || self.has_keyword(b"synchronous")
             || self.has_keyword(b"return")
             || self.has_keyword(b"assert")
             || self.has_keyword(b"goto")
             || self.has_keyword(b"new")
+    }
     fn has_type_keyword(&self) -> bool {
         self.has_keyword(b"in")
             || self.has_keyword(b"out")
             || self.has_keyword(b"msg")
             || self.has_keyword(b"boolean")
             || self.has_keyword(b"byte")
             || self.has_keyword(b"short")
             || self.has_keyword(b"int")
             || self.has_keyword(b"long")
             || self.has_keyword(b"auto")
+    }
     fn has_builtin_keyword(&self) -> bool {
         self.has_keyword(b"get")
             || self.has_keyword(b"fires")
             || self.has_keyword(b"create")
             || self.has_keyword(b"length")
+    }
     fn has_reserved(&self) -> bool {
         self.has_statement_keyword()
             || self.has_type_keyword()
             || self.has_builtin_keyword()
             || self.has_keyword(b"let")
             || self.has_keyword(b"struct")
             || self.has_keyword(b"enum")
             || self.has_keyword(b"true")
             || self.has_keyword(b"false")
             || self.has_keyword(b"null")
+    }
     // Identifiers
     fn has_identifier(&self) -> bool {
         if self.has_statement_keyword() || self.has_type_keyword() || self.has_builtin_keyword() {
             return false;
+        }
         let next = self.source.next();
         is_ident_start(next)
+    }
     fn consume_identifier(&mut self) -> Result<Identifier, ParseError2> {
         if self.has_statement_keyword() || self.has_type_keyword() || self.has_builtin_keyword() {
             return Err(self.error_at_pos("Expected identifier"));
+        }
         let position = self.source.pos();
         let value = self.consume_ident()?;
         Ok(Identifier{ position, value })
+    }
     fn consume_identifier_spilled(&mut self) -> Result<(), ParseError2> {
         if self.has_statement_keyword() || self.has_type_keyword() || self.has_builtin_keyword() {
             return Err(self.error_at_pos("Expected identifier"));
+        }
         self.consume_ident()?;
         Ok(())
+    }
     fn has_namespaced_identifier(&self) -> bool {
         self.has_identifier()
+    }
     fn consume_namespaced_identifier(&mut self) -> Result<NamespacedIdentifier, ParseError2> {
         if self.has_reserved() {
             return Err(self.error_at_pos("Encountered reserved keyword"));
+        }
         let position = self.source.pos();
         let mut ns_ident = self.consume_ident()?;
         let mut num_namespaces = 1;
         while self.has_string(b"::") {
             self.consume_string(b"::")?;
             if num_namespaces >= MAX_NAMESPACES {
                 return Err(self.error_at_pos("Too many namespaces in identifier"));
+            }
             let new_ident = self.consume_ident()?;
             ns_ident.extend(b"::");
             ns_ident.extend(new_ident);
             num_namespaces += 1;
+        }
         Ok(NamespacedIdentifier{
             position,
             value: ns_ident,
             num_namespaces,
         })
+    }
     fn consume_namespaced_identifier_spilled(&mut self) -> Result<(), ParseError2> {
         // TODO: @performance
         if self.has_reserved() {
             return Err(self.error_at_pos("Encountered reserved keyword"));
+        }
         self.consume_ident()?;
         while self.has_string(b"::") {
             self.consume_string(b"::")?;
             self.consume_ident()?;
+        }
         Ok(())
+    }
     fn consume_namespaced_identifier2(&mut self, h: &mut Heap) -> Result<NamespacedIdentifier2, ParseError2> {
     fn consume_namespaced_identifier(&mut self, h: &mut Heap) -> Result<NamespacedIdentifier, ParseError2> {
         if self.has_reserved() {
             return Err(self.error_at_pos("Encountered reserved keyword"));
+        }
         // Consumes a part of the namespaced identifier, returns a boolean
         // indicating whether polymorphic arguments were specified.
         // TODO: Continue here: if we fail to properly parse the polymorphic
         //  arguments, assume we have reached the end of the namespaced
         //  identifier and are instead dealing with a less-than operator. Ugly?
         //  Yes. Needs tokenizer? Yes.
         fn consume_part(
-            l: &mut Lexer, h: &mut Heap, ident: &mut NamespacedIdentifier2,
+            l: &mut Lexer, h: &mut Heap, ident: &mut NamespacedIdentifier,
             backup_pos: &mut InputPosition
         ) -> Result<(), ParseError2> {
             // Consume identifier
             if !ident.value.is_empty() {
                 ident.value.extend(b"::");
+            }
             let ident_start = ident.value.len();
             ident.value.extend(l.consume_ident()?);
             ident.parts.push(NamespacedIdentifierPart::Identifier{
                 start: ident_start as u16,
                 end: ident.value.len() as u16
             });
             // Maybe consume polymorphic args.
             *backup_pos = l.source.pos();
             l.consume_whitespace(false)?;
             match l.consume_polymorphic_args(h, true)? {
                 Some(args) => {
                     let poly_start = ident.poly_args.len();
                     ident.poly_args.extend(args);
                     ident.parts.push(NamespacedIdentifierPart::PolyArgs{
                         start: poly_start as u16,
                         end: ident.poly_args.len() as u16,
                     });
                     *backup_pos = l.source.pos();
                 },
                 None => {}
             };
             Ok(())
+        }
-        let mut ident = NamespacedIdentifier2{
+        let mut ident = NamespacedIdentifier{
             position: self.source.pos(),
             value: Vec::new(),
             poly_args: Vec::new(),
             parts: Vec::new(),
         };
         // Keep consume parts separted by "::". We don't consume the trailing
         // whitespace, hence we keep a backup position at the end of the last
         // valid part of the namespaced identifier (i.e. the last ident, or the
         // last encountered polymorphic arguments).
         let mut backup_pos = self.source.pos();
         consume_part(self, h, &mut ident, &mut backup_pos)?;
         self.consume_whitespace(false)?;
         while self.has_string(b"::") {
             self.consume_string(b"::")?;
             self.consume_whitespace(false)?;
             consume_part(self, h, &mut ident, &mut backup_pos)?;
             self.consume_whitespace(false)?;
+        }
         self.source.seek(backup_pos);
         Ok(ident)
+    }
     fn consume_namespaced_identifier_spilled(&mut self) -> Result<(), ParseError2> {
         if self.has_reserved() {
             return Err(self.error_at_pos("Encountered reserved keyword"));
+        }
         debug_log!("consume_nsident2_spilled: {}", debug_line!(self.source));
         fn consume_part_spilled(l: &mut Lexer, backup_pos: &mut InputPosition) -> Result<(), ParseError2> {
             l.consume_ident()?;
             *backup_pos = l.source.pos();
             l.consume_whitespace(false)?;
             match l.maybe_consume_poly_args_spilled_without_pos_recovery() {
                 Ok(true) => { *backup_pos = l.source.pos(); },
                 Ok(false) => {},
                 Err(_) => { return Err(l.error_at_pos("Failed to parse poly args (spilled)")) },
+            }
             Ok(())
+        }
         let mut backup_pos = self.source.pos();
         consume_part_spilled(self, &mut backup_pos)?;
         self.consume_whitespace(false)?;
         while self.has_string(b"::") {
             self.consume_string(b"::")?;
             self.consume_whitespace(false)?;
             consume_part_spilled(self, &mut backup_pos)?;
             self.consume_whitespace(false)?;
+        }
         self.source.seek(backup_pos);
         Ok(())
+    }
     // Types and type annotations
     /// Consumes a type definition. When called the input position should be at
     /// the type specification. When done the input position will be at the end
     /// of the type specifications (hence may be at whitespace).
     fn consume_type2(&mut self, h: &mut Heap, allow_inference: bool) -> Result<ParserTypeId, ParseError2> {
         // Small helper function to convert in/out polymorphic arguments. Not
         // pretty, but return boolean is true if the error is due to inference
         // not being allowed
         let reduce_port_poly_args = |
             heap: &mut Heap,
             port_pos: &InputPosition,
             args: Vec<ParserTypeId>,
         | -> Result<ParserTypeId, bool> {
             match args.len() {
 => if allow_inference {
                     Ok(heap.alloc_parser_type(|this| ParserType{
                         this,
                         pos: port_pos.clone(),
                         variant: ParserTypeVariant::Inferred
                     }))
                 } else {
                     Err(true)
                 },
 => Ok(args[0]),
                 _ => Err(false)
+            }
         };
         // Consume the type
         debug_log!("consume_type2: {}", debug_line!(self.source));
         let pos = self.source.pos();
         let parser_type_variant = if self.has_keyword(b"msg") {
             self.consume_keyword(b"msg")?;
             ParserTypeVariant::Message
         } else if self.has_keyword(b"boolean") {
             self.consume_keyword(b"boolean")?;
             ParserTypeVariant::Bool
         } else if self.has_keyword(b"byte") {
             self.consume_keyword(b"byte")?;
             ParserTypeVariant::Byte
         } else if self.has_keyword(b"short") {
             self.consume_keyword(b"short")?;
             ParserTypeVariant::Short
         } else if self.has_keyword(b"int") {
             self.consume_keyword(b"int")?;
             ParserTypeVariant::Int
         } else if self.has_keyword(b"long") {
             self.consume_keyword(b"long")?;
             ParserTypeVariant::Long
         } else if self.has_keyword(b"str") {
             self.consume_keyword(b"str")?;
             ParserTypeVariant::String
         } else if self.has_keyword(b"auto") {
             if !allow_inference {
                 return Err(ParseError2::new_error(
                         &self.source, pos,
                         "Type inference is not allowed here"
                 ));
+            }
             self.consume_keyword(b"auto")?;
             ParserTypeVariant::Inferred
         } else if self.has_keyword(b"in") {
             // TODO: @cleanup: not particularly neat to have this special case
             //  where we enforce polyargs in the parser-phase
             self.consume_keyword(b"in")?;
             let poly_args = self.consume_polymorphic_args(h, allow_inference)?.unwrap_or_default();
             let poly_arg = reduce_port_poly_args(h, &pos, poly_args)
                 .map_err(|infer_error|  {
                     let msg = if infer_error {
                         "Type inference is not allowed here"
                     } else {
                         "Type 'in' only allows for 1 polymorphic argument"
                     };
                     ParseError2::new_error(&self.source, pos, msg)
                 })?;
             ParserTypeVariant::Input(poly_arg)
         } else if self.has_keyword(b"out") {
             self.consume_keyword(b"out")?;
             let poly_args = self.consume_polymorphic_args(h, allow_inference)?.unwrap_or_default();
             let poly_arg = reduce_port_poly_args(h, &pos, poly_args)
                 .map_err(|infer_error| {
                     let msg = if infer_error {
                         "Type inference is not allowed here"
                     } else {
                         "Type 'out' only allows for 1 polymorphic argument, but {} were specified"
                     };
                     ParseError2::new_error(&self.source, pos, msg)
                 })?;
             ParserTypeVariant::Output(poly_arg)
         } else {
             // Must be a symbolic type
-            let identifier = self.consume_namespaced_identifier2(h)?;
+            let identifier = self.consume_namespaced_identifier(h)?;
             ParserTypeVariant::Symbolic(SymbolicParserType{identifier, variant: None, poly_args2: Vec::new()})
         };
         // If the type was a basic type (not supporting polymorphic type
         // arguments), then we make sure the user did not specify any of them.
         let mut backup_pos = self.source.pos();
         if !parser_type_variant.supports_polymorphic_args() {
             self.consume_whitespace(false)?;
             if let Some(b'<') = self.source.next() {
                 return Err(ParseError2::new_error(
                     &self.source, self.source.pos(),
                     "This type does not allow polymorphic arguments"
                 ));
+            }
             self.source.seek(backup_pos);
+        }
         let mut parser_type_id = h.alloc_parser_type(|this| ParserType{
             this, pos, variant: parser_type_variant
         });
         // If we're dealing with arrays, then we need to wrap the currently
         // parsed type in array types
         self.consume_whitespace(false)?;
         while let Some(b'[') = self.source.next() {
             let pos = self.source.pos();
             self.source.consume();
             self.consume_whitespace(false)?;
             if let Some(b']') = self.source.next() {
                 // Type is wrapped in an array
                 self.source.consume();
                 parser_type_id = h.alloc_parser_type(|this| ParserType{
                     this, pos, variant: ParserTypeVariant::Array(parser_type_id)
                 });
                 backup_pos = self.source.pos();
                 // In case we're dealing with another array
                 self.consume_whitespace(false)?;
             } else {
                 return Err(ParseError2::new_error(
                     &self.source, pos,
                     "Expected a closing ']'"
                 ));
+            }
+        }
         self.source.seek(backup_pos);
         Ok(parser_type_id)
+    }
     /// Attempts to consume a type without returning it. If it doesn't encounter
     /// a well-formed type, then the input position is left at a "random"
     /// position.
     fn maybe_consume_type_spilled_without_pos_recovery(&mut self) -> bool {
         // Consume type identifier
         debug_log!("maybe_consume_type_spilled_...: {}", debug_line!(self.source));
         if self.has_type_keyword() {
             self.consume_any_chars();
         } else {
-            let ident = self.consume_namespaced_identifier();
+            let ident = self.consume_namespaced_identifier_spilled();
             if ident.is_err() { return false; }
+        }
         // Consume any polymorphic arguments that follow the type identifier
         let mut backup_pos = self.source.pos();
         if self.consume_whitespace(false).is_err() { return false; }
         match self.maybe_consume_poly_args_spilled_without_pos_recovery() {
             Ok(true) => backup_pos = self.source.pos(),
             Ok(false) => {},
             Err(()) => return false
+        }
         // Consume any array specifiers. Make sure we always leave the input
         // position at the end of the last array specifier if we do find a
         // valid type
         if self.consume_whitespace(false).is_err() { return false; }
         while let Some(b'[') = self.source.next() {
             self.source.consume();
             if self.consume_whitespace(false).is_err() { return false; }
             if self.source.next() != Some(b']') { return false; }
             self.source.consume();
             backup_pos = self.source.pos();
             if self.consume_whitespace(false).is_err() { return false; }
+        }
         self.source.seek(backup_pos);
         return true;
+    }
     fn maybe_consume_type_spilled(&mut self) -> bool {
         let backup_pos = self.source.pos();
         if !self.maybe_consume_type_spilled_without_pos_recovery() {
             self.source.seek(backup_pos);
             return false;
+        }
         return true;
+    }
     /// Attempts to consume polymorphic arguments without returning them. If it
     /// doesn't encounter well-formed polymorphic arguments, then the input
     /// position is left at a "random" position. Returns a boolean indicating if
     /// the poly_args list was present.
     fn maybe_consume_poly_args_spilled_without_pos_recovery(&mut self) -> Result<bool, ()> {
         debug_log!("maybe_consume_poly_args_spilled_...: {}", debug_line!(self.source));
         self.consume_comma_separated_spilled_without_pos_recovery(
             b'<', b'>', |lexer| {
                 lexer.maybe_consume_type_spilled_without_pos_recovery()
             })
+    }
     /// Consumes polymorphic arguments and its delimiters if specified. If
     /// polyargs are present then the args are consumed and the input position
     /// will be placed after the polyarg list. If polyargs are not present then
     /// the input position will remain unmodified and an empty vector will be
     /// returned.
     ///
     /// Polymorphic arguments represent the specification of the parametric
     /// types of a polymorphic type: they specify the value of the polymorphic
     /// type's polymorphic variables.
     fn consume_polymorphic_args(&mut self, h: &mut Heap, allow_inference: bool) -> Result<Option<Vec<ParserTypeId>>, ParseError2> {
         self.consume_comma_separated(
             h, b'<', b'>', "Expected the end of the polymorphic argument list",
             |lexer, heap| lexer.consume_type2(heap, allow_inference)
+        )
+    }
     /// Consumes polymorphic variables. These are identifiers that are used
     /// within polymorphic types. The input position may be at whitespace. If
     /// polymorphic variables are present then the whitespace, wrapping
     /// delimiters and the polymorphic variables are consumed. Otherwise the
     /// input position will stay where it is. If no polymorphic variables are
     /// present then an empty vector will be returned.
     fn consume_polymorphic_vars(&mut self, h: &mut Heap) -> Result<Vec<Identifier>, ParseError2> {
         let backup_pos = self.source.pos();
         match self.consume_comma_separated(
             h, b'<', b'>', "Expected the end of the polymorphic variable list",
             |lexer, _heap| lexer.consume_identifier()
         )? {
             Some(poly_vars) => Ok(poly_vars),
             None => {
                 self.source.seek(backup_pos);
                 Ok(vec!())
+            }
+        }
+    }
     // Parameters
     fn consume_parameter(&mut self, h: &mut Heap) -> Result<ParameterId, ParseError2> {
         let parser_type = self.consume_type2(h, false)?;
         self.consume_whitespace(true)?;
         let position = self.source.pos();
         let identifier = self.consume_identifier()?;
         let id =
             h.alloc_parameter(|this| Parameter { this, position, parser_type, identifier });
         Ok(id)
+    }
     fn consume_parameters(&mut self, h: &mut Heap) -> Result<Vec<ParameterId>, ParseError2> {
         match self.consume_comma_separated(
             h, b'(', b')', "Expected the end of the parameter list",
             |lexer, heap| lexer.consume_parameter(heap)
         )? {
             Some(params) => Ok(params),
             None => {
                 Err(ParseError2::new_error(
                     &self.source, self.source.pos(),
@@ @@ -1442,286 +1428,282 @@ impl Lexer<'_> { @@
             return Ok(self.consume_array_expression(h)?.upcast());
+        }
         if self.has_builtin_literal() {
             return Ok(self.consume_builtin_literal_expression(h)?.upcast());
+        }
         if self.has_struct_literal() {
             return Ok(self.consume_struct_literal_expression(h)?.upcast());
+        }
         if self.has_call_expression() {
             return Ok(self.consume_call_expression(h)?.upcast());
+        }
         Ok(self.consume_variable_expression(h)?.upcast())
+    }
     fn consume_array_expression(&mut self, h: &mut Heap) -> Result<ArrayExpressionId, ParseError2> {
         let position = self.source.pos();
         let mut elements = Vec::new();
         self.consume_string(b"{")?;
         self.consume_whitespace(false)?;
         if !self.has_string(b"}") {
             while self.source.next().is_some() {
                 elements.push(self.consume_expression(h)?);
                 self.consume_whitespace(false)?;
                 if self.has_string(b"}") {
                     break;
+                }
                 self.consume_string(b",")?;
                 self.consume_whitespace(false)?;
+            }
+        }
         self.consume_string(b"}")?;
         Ok(h.alloc_array_expression(|this| ArrayExpression {
             this,
             position,
             elements,
             parent: ExpressionParent::None,
             concrete_type: ConcreteType::default(),
         }))
+    }
     fn has_builtin_literal(&self) -> bool {
         is_constant(self.source.next())
             || self.has_keyword(b"null")
             || self.has_keyword(b"true")
             || self.has_keyword(b"false")
+    }
     fn consume_builtin_literal_expression(
         &mut self,
         h: &mut Heap,
     ) -> Result<LiteralExpressionId, ParseError2> {
         let position = self.source.pos();
         let value;
         if self.has_keyword(b"null") {
             self.consume_keyword(b"null")?;
             value = Literal::Null;
         } else if self.has_keyword(b"true") {
             self.consume_keyword(b"true")?;
             value = Literal::True;
         } else if self.has_keyword(b"false") {
             self.consume_keyword(b"false")?;
             value = Literal::False;
         } else if self.source.next() == Some(b'\'') {
             self.source.consume();
             let mut data = Vec::new();
             let mut next = self.source.next();
             while next != Some(b'\'') && (is_vchar(next) || next == Some(b' ')) {
                 data.push(next.unwrap());
                 self.source.consume();
                 next = self.source.next();
+            }
             if next != Some(b'\'') || data.is_empty() {
                 return Err(self.error_at_pos("Expected character constant"));
+            }
             self.source.consume();
             value = Literal::Character(data);
         } else {
             if !self.has_integer() {
                 return Err(self.error_at_pos("Expected integer constant"));
+            }
             value = Literal::Integer(self.consume_integer()?);
+        }
         Ok(h.alloc_literal_expression(|this| LiteralExpression {
             this,
             position,
             value,
             parent: ExpressionParent::None,
             concrete_type: ConcreteType::default(),
         }))
+    }
     fn has_struct_literal(&mut self) -> bool {
         // A struct literal is written as:
         //      namespace::StructName<maybe_one_of_these, auto>{ field: expr }
         // We will parse up until the opening brace to see if we're dealing with
         // a struct literal.
         let backup_pos = self.source.pos();
         let result = self.consume_namespaced_identifier_spilled().is_ok() &&
             self.consume_whitespace(false).is_ok() &&
             self.maybe_consume_poly_args_spilled_without_pos_recovery().is_ok() &&
             self.consume_whitespace(false).is_ok() &&
             self.source.next() == Some(b'{');
         self.source.seek(backup_pos);
         return result;
+    }
     fn consume_struct_literal_expression(&mut self, h: &mut Heap) -> Result<LiteralExpressionId, ParseError2> {
         // Consume identifier and polymorphic arguments
         debug_log!("consume_struct_literal_expression: {}", debug_line!(self.source));
         let position = self.source.pos();
-        let identifier = self.consume_namespaced_identifier2(h)?;
+        let identifier = self.consume_namespaced_identifier(h)?;
         self.consume_whitespace(false)?;
         // Consume fields
         let fields = match self.consume_comma_separated(
             h, b'{', b'}', "Expected the end of the list of struct fields",
             |lexer, heap| {
                 let identifier = lexer.consume_identifier()?;
                 lexer.consume_whitespace(false)?;
                 lexer.consume_string(b":")?;
                 lexer.consume_whitespace(false)?;
                 let value = lexer.consume_expression(heap)?;
                 Ok(LiteralStructField{ identifier, value, field_idx: 0 })
+            }
         )? {
             Some(fields) => fields,
             None => return Err(ParseError2::new_error(
                 self.source, self.source.pos(),
                 "A struct literal must be followed by its field values"
             ))
         };
         Ok(h.alloc_literal_expression(|this| LiteralExpression{
             this,
             position,
             value: Literal::Struct(LiteralStruct{
                 identifier,
                 fields,
                 poly_args2: Vec::new(),
                 definition: None,
             }),
             parent: ExpressionParent::None,
             concrete_type: Default::default()
         }))
+    }
     fn has_call_expression(&mut self) -> bool {
         // We need to prevent ambiguity with various operators (because we may
         // be specifying polymorphic variables) and variables.
         if self.has_builtin_keyword() {
             return true;
+        }
         let backup_pos = self.source.pos();
         let mut result = false;
         if self.consume_namespaced_identifier_spilled().is_ok() &&
             self.consume_whitespace(false).is_ok() &&
             self.maybe_consume_poly_args_spilled_without_pos_recovery().is_ok() &&
             self.consume_whitespace(false).is_ok() &&
             self.source.next() == Some(b'(') {
             // Seems like we have a function call or an enum literal
             result = true;
+        }
         self.source.seek(backup_pos);
         return result;
+    }
     fn consume_call_expression(&mut self, h: &mut Heap) -> Result<CallExpressionId, ParseError2> {
         let position = self.source.pos();
         // Consume method identifier
         // TODO: @token Replace this conditional polymorphic arg parsing once we have a tokenizer.
         debug_log!("consume_call_expression: {}", debug_line!(self.source));
         let method;
         let mut consume_poly_args_explicitly = true;
         if self.has_keyword(b"get") {
             self.consume_keyword(b"get")?;
             method = Method::Get;
         } else if self.has_keyword(b"put") {
             self.consume_keyword(b"put")?;
             method = Method::Put;
         } else if self.has_keyword(b"fires") {
             self.consume_keyword(b"fires")?;
             method = Method::Fires;
         } else if self.has_keyword(b"create") {
             self.consume_keyword(b"create")?;
             method = Method::Create;
         } else {
-            let identifier = self.consume_namespaced_identifier2(h)?;
+            let identifier = self.consume_namespaced_identifier(h)?;
             method = Method::Symbolic(MethodSymbolic{
                 identifier,
                 definition: None
             });
             consume_poly_args_explicitly = false;
         };
         // Consume polymorphic arguments
         let poly_args = if consume_poly_args_explicitly {
             self.consume_whitespace(false)?;
             self.consume_polymorphic_args(h, true)?.unwrap_or_default()
         } else {
             Vec::new()
         };
         // Consume arguments to call
         self.consume_whitespace(false)?;
         let mut arguments = Vec::new();
         self.consume_string(b"(")?;
         self.consume_whitespace(false)?;
         if !self.has_string(b")") {
             // TODO: allow trailing comma
             while self.source.next().is_some() {
                 arguments.push(self.consume_expression(h)?);
                 self.consume_whitespace(false)?;
                 if self.has_string(b")") {
                     break;
+                }
                 self.consume_string(b",")?;
                 self.consume_whitespace(false)?
+            }
+        }
         self.consume_string(b")")?;
         Ok(h.alloc_call_expression(|this| CallExpression {
             this,
             position,
             method,
             arguments,
             poly_args,
             parent: ExpressionParent::None,
             concrete_type: ConcreteType::default(),
         }))
+    }
     fn consume_variable_expression(
         &mut self,
         h: &mut Heap,
     ) -> Result<VariableExpressionId, ParseError2> {
         let position = self.source.pos();
         debug_log!("consume_variable_expression: {}", debug_line!(self.source));
         // TODO: @token Reimplement when tokenizer is implemented, prevent ambiguities
         let identifier = identifier_as_namespaced(self.consume_identifier()?);
         Ok(h.alloc_variable_expression(|this| VariableExpression {
             this,
             position,
             identifier,
             declaration: None,
             parent: ExpressionParent::None,
             concrete_type: ConcreteType::default(),
         }))
+    }
     // ====================
     // Statements
     // ====================
     /// Consumes any kind of statement from the source and will error if it
     /// did not encounter a statement. Will also return an error if the
     /// statement is nested too deeply.
     ///
     /// `wrap_in_block` may be set to true to ensure that the parsed statement
     /// will be wrapped in a block statement if it is not already a block
     /// statement. This is used to ensure that all `if`, `while` and `sync`
     /// statements have a block statement as body.
     fn consume_statement(&mut self, h: &mut Heap, wrap_in_block: bool) -> Result<StatementId, ParseError2> {
         if self.level >= MAX_LEVEL {
             return Err(self.error_at_pos("Too deeply nested statement"));
+        }
         self.level += 1;
         let result = self.consume_statement_impl(h, wrap_in_block);
         self.level -= 1;
         result
+    }
     fn has_label(&mut self) -> bool {
         // To prevent ambiguity with expression statements consisting only of an
         // identifier or a namespaced identifier, we look ahead and match on the
         // *single* colon that signals a labeled statement.
         let backup_pos = self.source.pos();
         let mut result = false;
         if self.consume_identifier_spilled().is_ok() {
             // next character is ':', second character is NOT ':'
             result = Some(b':') == self.source.next() && Some(b':') != self.source.lookahead(1)
+        }
         self.source.seek(backup_pos);
         return result;

src/protocol/parser/symbol_table.rs

➞

Show inline comments

 // TODO: Maybe allow namespaced-aliased imports. It is currently not possible
 //  to express the following:
 //      import Module.Submodule as SubMod
 //      import SubMod::{Symbol}
 //  And it is especially not possible to express the following:
 //      import SubMod::{Symbol}
 //      import Module.Submodule as SubMod
 use crate::protocol::ast::*;
 use crate::protocol::inputsource::*;
 use std::collections::{HashMap, hash_map::Entry};
 use crate::protocol::parser::LexedModule;
 #[derive(PartialEq, Eq, Hash)]
 struct SymbolKey {
     module_id: RootId,
     symbol_name: Vec<u8>,
+}
 impl SymbolKey {
     fn from_identifier(module_id: RootId, symbol: &Identifier) -> Self {
         Self{ module_id, symbol_name: symbol.value.clone() }
+    }
-    fn from_namespaced_identifier(module_id: RootId, symbol: &NamespacedIdentifier2) -> Self {
+    fn from_namespaced_identifier(module_id: RootId, symbol: &NamespacedIdentifier) -> Self {
         Self{ module_id, symbol_name: symbol.strip_poly_args() }
+    }
+}
 pub(crate) enum Symbol {
     Namespace(RootId),
     Definition((RootId, DefinitionId)),
+}
 pub(crate) struct SymbolValue {
     // Position is the place where the symbol is introduced to a module (this
     // position always corresponds to the module whose RootId is stored in the
     // `SymbolKey` associated with this `SymbolValue`). For a definition this
     // is the position where the symbol is defined, for an import this is the
     // position of the import statement.
     pub(crate) position: InputPosition,
     pub(crate) symbol: Symbol,
+}
 impl SymbolValue {
     pub(crate) fn is_namespace(&self) -> bool {
         match &self.symbol {
             Symbol::Namespace(_) => true,
             _ => false
+        }
+    }
     pub(crate) fn as_namespace(&self) -> Option<RootId> {
         match &self.symbol {
             Symbol::Namespace(root_id) => Some(*root_id),
             _ => None,
+        }
+    }
     pub(crate) fn as_definition(&self) -> Option<(RootId, DefinitionId)> {
         match &self.symbol {
             Symbol::Definition((root_id, definition_id)) => Some((*root_id, *definition_id)),
             _ => None,
+        }
+    }
+}
 /// `SymbolTable` is responsible for two parts of the parsing process: firstly
 /// it ensures that there are no clashing symbol definitions within each file,
 /// and secondly it will resolve all symbols within a module to their
 /// appropriate definitions (in case of enums, functions, etc.) and namespaces
 /// (currently only external modules can act as namespaces). If a symbol clashes
 /// or if a symbol cannot be resolved this will be an error.
 ///
 /// Within the compilation process the symbol table is responsible for resolving
 /// namespaced identifiers (e.g. Module::Enum::EnumVariant) to the appropriate
 /// definition (i.e. not namespaces; as the language has no way to use
 /// namespaces except for using them in namespaced identifiers).
 pub(crate) struct SymbolTable {
     // Lookup from module name (not any aliases) to the root id
     module_lookup: HashMap<Vec<u8>, RootId>,
     // Lookup from within a module, to a particular imported (potentially
     // aliased) or defined symbol. Basically speaking: if the source code of a
     // module contains correctly imported/defined symbols, then this lookup
     // will always return the corresponding definition
     symbol_lookup: HashMap<SymbolKey, SymbolValue>,
+}
 impl SymbolTable {
     pub(crate) fn new() -> Self {
         Self{ module_lookup: HashMap::new(), symbol_lookup: HashMap::new() }
+    }
     pub(crate) fn build(&mut self, heap: &Heap, modules: &[LexedModule]) -> Result<(), ParseError2> {
         // Sanity checks
         debug_assert!(self.module_lookup.is_empty());
         debug_assert!(self.symbol_lookup.is_empty());
         if cfg!(debug_assertions) {
             for (index, module) in modules.iter().enumerate() {
                 debug_assert_eq!(
                     index, module.root_id.index as usize,
                     "module RootId does not correspond to LexedModule index"
+                )
+            }
+        }
         // Preparation: create a lookup from module name to root id. This does
         // not take aliasing into account.
         self.module_lookup.reserve(modules.len());
         for module in modules {
             // TODO: Maybe put duplicate module name checking here?
             // TODO: @string
             self.module_lookup.insert(module.module_name.clone(), module.root_id);
+        }
         // Preparation: determine total number of imports we will be inserting
         // into the lookup table. We could just iterate over the arena, but then
         // we don't know the source file the import belongs to.
         let mut lookup_reserve_size = 0;
         for module in modules {
             let module_root = &heap[module.root_id];
             for import_id in &module_root.imports {
                 match &heap[*import_id] {
@@ @@ -242,194 +242,194 @@ impl SymbolTable { @@
                                 // to "import a module's imported symbol". And so if we do find
                                 // a symbol match, we need to make sure it is a definition from
                                 // within that module by checking `source_root_id == target_root_id`
                                 let key = SymbolKey::from_identifier(target_root_id, &symbol.name);
                                 let target_symbol = self.symbol_lookup.get(&key);
                                 let symbol_definition_id = match target_symbol {
                                     Some(target_symbol) => {
                                         match target_symbol.symbol {
                                             Symbol::Definition((symbol_root_id, symbol_definition_id)) => {
                                                 if symbol_root_id == target_root_id {
                                                     Some(symbol_definition_id)
                                                 } else {
                                                     // This is imported within the target module, and not
                                                     // defined within the target module
                                                     None
+                                                }
                                             },
                                             Symbol::Namespace(_) => {
                                                 // We don't import a module's "module import"
                                                 None
+                                            }
+                                        }
                                     },
                                     None => None
                                 };
                                 if symbol_definition_id.is_none() {
                                     return Err(
                                         ParseError2::new_error(&module.source, symbol.position, "Could not resolve symbol")
+                                    )
+                                }
                                 let symbol_definition_id = symbol_definition_id.unwrap();
                                 if let Err(previous_position) = self.add_definition_symbol(
                                     symbol.position, SymbolKey::from_identifier(module.root_id, &symbol.alias),
                                     target_root_id, symbol_definition_id
                                 ) {
                                     return Err(
                                         ParseError2::new_error(&module.source, symbol.position, "Symbol is multiply defined")
                                             .with_postfixed_info(&module.source, previous_position, "Previous definition was here")
+                                    )
+                                }
+                            }
+                        }
+                    }
+                }
+            }
+        }
         fn find_name(heap: &Heap, root_id: RootId) -> String {
             let root = &heap[root_id];
             for pragma_id in &root.pragmas {
                 match &heap[*pragma_id] {
                     Pragma::Module(module) => {
                         return String::from_utf8_lossy(&module.value).to_string()
                     },
                     _ => {},
+                }
+            }
             return String::from("Unknown")
+        }
         debug_assert_eq!(
             self.symbol_lookup.len(), lookup_reserve_size,
             "miscalculated reserved size for symbol lookup table"
         );
         Ok(())
+    }
     /// Resolves a module by its defined name
     pub(crate) fn resolve_module(&self, identifier: &Vec<u8>) -> Option<RootId> {
         self.module_lookup.get(identifier).map(|v| *v)
+    }
     pub(crate) fn resolve_symbol<'t>(
         &'t self, root_module_id: RootId, identifier: &[u8]
     ) -> Option<&'t SymbolValue> {
         let lookup_key = SymbolKey{ module_id: root_module_id, symbol_name: Vec::from(identifier) };
         self.symbol_lookup.get(&lookup_key)
+    }
     pub(crate) fn resolve_identifier<'t>(
         &'t self, root_module_id: RootId, identifier: &Identifier
     ) -> Option<&'t SymbolValue> {
         let lookup_key = SymbolKey::from_identifier(root_module_id, identifier);
         self.symbol_lookup.get(&lookup_key)
+    }
     /// Resolves a namespaced symbol. This method will go as far as possible in
     /// going to the right symbol. It will halt the search when:
     /// 1. Polymorphic arguments are encountered on the identifier.
     /// 2. A non-namespace symbol is encountered.
     /// 3. A part of the identifier couldn't be resolved to anything
     /// The returned iterator will always point to the next symbol (even if
     /// nothing was found)
     pub(crate) fn resolve_namespaced_identifier<'t, 'i>(
         &'t self, root_module_id: RootId, identifier: &'i NamespacedIdentifier2
     ) -> (Option<&'t SymbolValue>, NamespacedIdentifier2Iter<'i>) {
         &'t self, root_module_id: RootId, identifier: &'i NamespacedIdentifier
     ) -> (Option<&'t SymbolValue>, NamespacedIdentifierIter<'i>) {
         let mut iter = identifier.iter();
         let mut symbol: Option<&SymbolValue> = None;
         let mut within_module_id = root_module_id;
         while let Some((partial, poly_args)) = iter.next() {
             // Lookup the symbol within the currently iterated upon module
             let lookup_key = SymbolKey{ module_id: within_module_id, symbol_name: Vec::from(partial) };
             let new_symbol = self.symbol_lookup.get(&lookup_key);
             match new_symbol {
                 None => {
                     // Can't find anything
                     symbol = None;
                     break;
                 },
                 Some(new_symbol) => {
                     // Found something, but if we already moved to another
                     // module then we don't want to keep jumping across modules,
                     // we're only interested in symbols defined within that
                     // module.
                     match &new_symbol.symbol {
                         Symbol::Namespace(new_root_id) => {
                             if root_module_id != within_module_id {
                                 // This new symbol is imported by a foreign
                                 // module, so this is an error
                                 debug_assert!(symbol.is_some());
                                 debug_assert!(symbol.unwrap().is_namespace());
                                 debug_assert!(iter.num_returned() > 1);
                                 symbol = None;
                                 break;
+                            }
                             within_module_id = *new_root_id;
                             symbol = Some(new_symbol);
                         },
                         Symbol::Definition((definition_root_id, _)) => {
                             // Found a definition, but if we already jumped
                             // modules, then this must be defined within that
                             // module.
                             if root_module_id != within_module_id && within_module_id != *definition_root_id {
                                 // This is an imported definition within the module
                                 // So keep the old
                                 debug_assert!(symbol.is_some());
                                 debug_assert!(symbol.unwrap().is_namespace());
                                 debug_assert!(iter.num_returned() > 1);
                                 symbol = None;
                                 break;
+                            }
                             symbol = Some(new_symbol);
                             break;
+                        }
+                    }
+                }
+            }
             if poly_args.is_some() {
                 // Polymorphic argument specification should also be a fully
                 // resolved result.
                 break;
+            }
+        }
         match symbol {
             None => (None, iter),
             Some(symbol) => (Some(symbol), iter)
+        }
+    }
     /// Attempts to add a namespace symbol. Returns `Ok` if the symbol was
     /// inserted. If the symbol already exists then `Err` will be returned
     /// together with the previous definition's source position (in the origin
     /// module's source file).
     // Note: I would love to return a reference to the value, but Rust is
     // preventing me from doing so... That, or I'm not smart enough...
     fn add_namespace_symbol(
         &mut self, origin_position: InputPosition, key: SymbolKey, target_module_id: RootId
     ) -> Result<(), InputPosition> {
         match self.symbol_lookup.entry(key) {
             Entry::Occupied(o) => Err(o.get().position),
             Entry::Vacant(v) => {
                 v.insert(SymbolValue{
                     position: origin_position,
                     symbol: Symbol::Namespace(target_module_id)
                 });
                 Ok(())
+            }
+        }
+    }
     /// Attempts to add a definition symbol. Returns `Ok` if the symbol was
     /// inserted. If the symbol already exists then `Err` will be returned
     /// together with the previous definition's source position (in the origin
     /// module's source file).
     fn add_definition_symbol(
         &mut self, origin_position: InputPosition, key: SymbolKey,
         target_module_id: RootId, target_definition_id: DefinitionId,
     ) -> Result<(), InputPosition> {

src/protocol/parser/utils.rs

➞

Show inline comments

     use crate::protocol::ast::*;
 use crate::protocol::inputsource::*;
 use super::symbol_table::*;
 use super::type_table::*;
 /// Utility result type.
 pub(crate) enum FindTypeResult<'t, 'i> {
     // Found the type exactly
-    Found((&'t DefinedType, NamespacedIdentifier2Iter<'i>)),
+    Found((&'t DefinedType, NamespacedIdentifierIter<'i>)),
     // Could not match symbol
     SymbolNotFound{ident_pos: InputPosition},
     // Matched part of the namespaced identifier, but not completely
-    SymbolPartial{ident_pos: InputPosition, ident_iter: NamespacedIdentifier2Iter<'i>},
+    SymbolPartial{ident_pos: InputPosition, ident_iter: NamespacedIdentifierIter<'i>},
     // Symbol matched, but points to a namespace/module instead of a type
     SymbolNamespace{ident_pos: InputPosition, symbol_pos: InputPosition},
+}
 // TODO: @cleanup Find other uses of this pattern
 impl<'t, 'i> FindTypeResult<'t, 'i> {
     /// Utility function to transform the `FindTypeResult` into a `Result` where
     /// `Ok` contains the resolved type, and `Err` contains a `ParseError` which
     /// can be readily returned. This is the most common use.
-    pub(crate) fn as_parse_error(self, module_source: &InputSource) -> Result<(&'t DefinedType, NamespacedIdentifier2Iter<'i>), ParseError2> {
+    pub(crate) fn as_parse_error(self, module_source: &InputSource) -> Result<(&'t DefinedType, NamespacedIdentifierIter<'i>), ParseError2> {
         match self {
             FindTypeResult::Found(defined_type) => Ok(defined_type),
             FindTypeResult::SymbolNotFound{ident_pos} => {
                 Err(ParseError2::new_error(
                     module_source, ident_pos,
                     "Could not resolve this identifier to a symbol"
                 ))
             },
             FindTypeResult::SymbolPartial{ident_pos, ident_iter} => {
                 Err(ParseError2::new_error(
                     module_source, ident_pos,
                     &format!(
                         "Could not fully resolve this identifier to a symbol, was only able to match '{}'",
                         &String::from_utf8_lossy(ident_iter.returned_section())
+                    )
                 ))
             },
             FindTypeResult::SymbolNamespace{ident_pos, symbol_pos} => {
                 Err(ParseError2::new_error(
                     module_source, ident_pos,
                     "This identifier was resolved to a namespace instead of a type"
                 ).with_postfixed_info(
                     module_source, symbol_pos,
                     "This is the referenced namespace"
                 ))
+            }
+        }
+    }
+}
 /// Attempt to find the type pointer to by a (root, identifier) combination. The
 /// type must match exactly (no parts in the namespace iterator remaining) and
 /// must be a type, not a namespace.
 pub(crate) fn find_type_definition<'t, 'i>(
     symbols: &SymbolTable, types: &'t TypeTable,
-    root_id: RootId, identifier: &'i NamespacedIdentifier2
+    root_id: RootId, identifier: &'i NamespacedIdentifier
 ) -> FindTypeResult<'t, 'i> {
     // Lookup symbol
     let (symbol, ident_iter) = symbols.resolve_namespaced_identifier(root_id, identifier);
     if symbol.is_none() {
         return FindTypeResult::SymbolNotFound{ident_pos: identifier.position};
+    }
     // Make sure we resolved it exactly
     let symbol = symbol.unwrap();
     if ident_iter.num_remaining() != 0 {
         return FindTypeResult::SymbolPartial{
             ident_pos: identifier.position,
             ident_iter
         };
+    }
     match symbol.symbol {
         Symbol::Namespace(_) => {
             FindTypeResult::SymbolNamespace{
                 ident_pos: identifier.position,
                 symbol_pos: symbol.position
+            }
         },
         Symbol::Definition((_, definition_id)) => {
             // If this function is called correctly, then we should always be
             // able to match the definition's ID to an entry in the type table.
             let definition = types.get_base_definition(&definition_id);
             debug_assert!(definition.is_some());
             FindTypeResult::Found((definition.unwrap(), ident_iter))
+        }
+    }
+}
 pub(crate) enum MatchPolymorphResult<'t> {
     Matching,
     InferAll(usize),
     Mismatch{defined_type: &'t DefinedType, ident_position: InputPosition, num_specified: usize},
-    NoneExpected{defined_type: &'t DefinedType, ident_position: InputPosition, num_specified: usize},
     NoneExpected{defined_type: &'t DefinedType, ident_position: InputPosition},
+}
 impl<'t> MatchPolymorphResult<'t> {
     pub(crate) fn as_parse_error(self, heap: &Heap, module_source: &InputSource) -> Result<usize, ParseError2> {
         match self {
             MatchPolymorphResult::Matching => Ok(0),
             MatchPolymorphResult::InferAll(count) => {
                 debug_assert!(count > 0);
                 Ok(count)
             },
             MatchPolymorphResult::Mismatch{defined_type, ident_position, num_specified} => {
                 let type_identifier = heap[defined_type.ast_definition].identifier();
                 let args_name = if defined_type.poly_vars.len() == 1 {
                     "argument"
                 } else {
                     "arguments"
                 };
                 return Err(ParseError2::new_error(
                     module_source, ident_position,
                     &format!(
                         "expected {} polymorphic {} (or none, to infer them) for the type {}, but {} were specified",
                         defined_type.poly_vars.len(), args_name,
                         &String::from_utf8_lossy(&type_identifier.value),
                         num_specified
+                    )
                 ))
             },
             MatchPolymorphResult::NoneExpected{defined_type, ident_position, ..} => {
                 let type_identifier = heap[defined_type.ast_definition].identifier();
                 return Err(ParseError2::new_error(
                     module_source, ident_position,
                     &format!(
                         "the type {} is not polymorphic",
                         &String::from_utf8_lossy(&type_identifier.value)
+                    )
                 ))
+            }
+        }
+    }
+}
 /// Attempt to match the polymorphic arguments to the number of polymorphic
 /// variables in the definition.
 pub(crate) fn match_polymorphic_args_to_vars<'t>(
     defined_type: &'t DefinedType, poly_args: Option<&[ParserTypeId]>, ident_position: InputPosition
 ) -> MatchPolymorphResult<'t> {
     if defined_type.poly_vars.is_empty() {
         // No polymorphic variables on type
         if poly_args.is_some() {
             return MatchPolymorphResult::NoneExpected{
                 defined_type,
                 ident_position,
-                num_specified: poly_args.unwrap().len()};
             };
+        }
     } else {
         // Polymorphic variables on type
         let has_specified = poly_args.map_or(false, |a| a.len() != 0);
         if !has_specified {
             // Implicitly infer all of the polymorphic arguments
             return MatchPolymorphResult::InferAll(defined_type.poly_vars.len());
+        }
         let num_specified = poly_args.unwrap().len();
         if num_specified != defined_type.poly_vars.len() {
             return MatchPolymorphResult::Mismatch{
                 defined_type,
                 ident_position,
                 num_specified,
             };
+        }
+    }
     MatchPolymorphResult::Matching
+}
@@ \ No newline at end of file @@

src/protocol/parser/visitor_linker.rs

➞

Show inline comments

@@ @@ -1146,327 +1146,327 @@ impl ValidityAndLinkerVisitor { @@
                 if num_inferred_to_allocate != 0 {
                     symbolic.poly_args2.reserve(num_inferred_to_allocate);
                     for _ in 0..num_inferred_to_allocate {
                         symbolic.poly_args2.push(self.parser_type_buffer.pop().unwrap());
+                    }
                 } else if !symbolic.identifier.poly_args.is_empty() {
                     symbolic.poly_args2.extend(&symbolic.identifier.poly_args)
+                }
                 symbolic.variant = Some(symbolic_variant);
             } else {
                 unreachable!();
+            }
+        }
         Ok(())
+    }
     //--------------------------------------------------------------------------
     // Utilities
     //--------------------------------------------------------------------------
     /// Adds a local variable to the current scope. It will also annotate the
     /// `Local` in the AST with its relative position in the block.
     fn checked_local_add(&mut self, ctx: &mut Ctx, relative_pos: u32, id: LocalId) -> Result<(), ParseError2> {
         debug_assert!(self.cur_scope.is_some());
         // Make sure we do not conflict with any global symbols
+        {
             let ident = &ctx.heap[id].identifier;
             if let Some(symbol) = ctx.symbols.resolve_symbol(ctx.module.root_id, &ident.value) {
                 return Err(
                     ParseError2::new_error(&ctx.module.source, ident.position, "Local variable declaration conflicts with symbol")
                         .with_postfixed_info(&ctx.module.source, symbol.position, "Conflicting symbol is found here")
                 );
+            }
+        }
         let local = &mut ctx.heap[id];
         local.relative_pos_in_block = relative_pos;
         // Make sure we do not shadow any variables in any of the scopes. Note
         // that variables in parent scopes may be declared later
         let local = &ctx.heap[id];
         let mut scope = self.cur_scope.as_ref().unwrap();
         let mut local_relative_pos = self.relative_pos_in_block;
         loop {
             debug_assert!(scope.is_block(), "scope is not a block");
             let block = &ctx.heap[scope.to_block()];
             for other_local_id in &block.locals {
                 let other_local = &ctx.heap[*other_local_id];
                 // Position check in case another variable with the same name
                 // is defined in a higher-level scope, but later than the scope
                 // in which the current variable resides.
                 if local.this != *other_local_id &&
                     local_relative_pos >= other_local.relative_pos_in_block &&
                     local.identifier == other_local.identifier {
                     // Collision within this scope
                     return Err(
                         ParseError2::new_error(&ctx.module.source, local.position, "Local variable name conflicts with another variable")
                             .with_postfixed_info(&ctx.module.source, other_local.position, "Previous variable is found here")
                     );
+                }
+            }
             // Current scope is fine, move to parent scope if any
             debug_assert!(block.parent_scope.is_some(), "block scope does not have a parent");
             scope = block.parent_scope.as_ref().unwrap();
             if let Scope::Definition(definition_id) = scope {
                 // At outer scope, check parameters of function/component
                 for parameter_id in ctx.heap[*definition_id].parameters() {
                     let parameter = &ctx.heap[*parameter_id];
                     if local.identifier == parameter.identifier {
                         return Err(
                             ParseError2::new_error(&ctx.module.source, local.position, "Local variable name conflicts with parameter")
                                 .with_postfixed_info(&ctx.module.source, parameter.position, "Parameter definition is found here")
                         );
+                    }
+                }
                 break;
+            }
             // If here, then we are dealing with a block-like parent block
             local_relative_pos = ctx.heap[scope.to_block()].relative_pos_in_parent;
+        }
         // No collisions at all
         let block = &mut ctx.heap[self.cur_scope.as_ref().unwrap().to_block()];
         block.locals.push(id);
         Ok(())
+    }
     /// Finds a variable in the visitor's scope that must appear before the
     /// specified relative position within that block.
-    fn find_variable(&self, ctx: &Ctx, mut relative_pos: u32, identifier: &NamespacedIdentifier2) -> Result<VariableId, ParseError2> {
+    fn find_variable(&self, ctx: &Ctx, mut relative_pos: u32, identifier: &NamespacedIdentifier) -> Result<VariableId, ParseError2> {
         debug_assert!(self.cur_scope.is_some());
         debug_assert!(identifier.parts.len() == 1, "implement namespaced seeking of target associated with identifier");
         // TODO: May still refer to an alias of a global symbol using a single
         //  identifier in the namespace.
         // No need to use iterator over namespaces if here
         let mut scope = self.cur_scope.as_ref().unwrap();
         loop {
             debug_assert!(scope.is_block());
             let block = &ctx.heap[scope.to_block()];
             for local_id in &block.locals {
                 let local = &ctx.heap[*local_id];
                 if local.relative_pos_in_block < relative_pos && identifier.matches_identifier(&local.identifier) {
                     return Ok(local_id.upcast());
+                }
+            }
             debug_assert!(block.parent_scope.is_some());
             scope = block.parent_scope.as_ref().unwrap();
             if !scope.is_block() {
                 // Definition scope, need to check arguments to definition
                 match scope {
                     Scope::Definition(definition_id) => {
                         let definition = &ctx.heap[*definition_id];
                         for parameter_id in definition.parameters() {
                             let parameter = &ctx.heap[*parameter_id];
                             if identifier.matches_identifier(&parameter.identifier) {
                                 return Ok(parameter_id.upcast());
+                            }
+                        }
                     },
                     _ => unreachable!(),
+                }
                 // Variable could not be found
                 return Err(ParseError2::new_error(
                     &ctx.module.source, identifier.position, "This variable is not declared"
                 ));
             } else {
                 relative_pos = block.relative_pos_in_parent;
+            }
+        }
+    }
     /// Adds a particular label to the current scope. Will return an error if
     /// there is another label with the same name visible in the current scope.
     fn checked_label_add(&mut self, ctx: &mut Ctx, id: LabeledStatementId) -> Result<(), ParseError2> {
         debug_assert!(self.cur_scope.is_some());
         // Make sure label is not defined within the current scope or any of the
         // parent scope.
         let label = &ctx.heap[id];
         let mut scope = self.cur_scope.as_ref().unwrap();
         loop {
             debug_assert!(scope.is_block(), "scope is not a block");
             let block = &ctx.heap[scope.to_block()];
             for other_label_id in &block.labels {
                 let other_label = &ctx.heap[*other_label_id];
                 if other_label.label == label.label {
                     // Collision
                     return Err(
                         ParseError2::new_error(&ctx.module.source, label.position, "Label name conflicts with another label")
                             .with_postfixed_info(&ctx.module.source, other_label.position, "Other label is found here")
                     );
+                }
+            }
             debug_assert!(block.parent_scope.is_some(), "block scope does not have a parent");
             scope = block.parent_scope.as_ref().unwrap();
             if !scope.is_block() {
                 break;
+            }
+        }
         // No collisions
         let block = &mut ctx.heap[self.cur_scope.as_ref().unwrap().to_block()];
         block.labels.push(id);
         Ok(())
+    }
     /// Finds a particular labeled statement by its identifier. Once found it
     /// will make sure that the target label does not skip over any variable
     /// declarations within the scope in which the label was found.
     fn find_label(&self, ctx: &Ctx, identifier: &Identifier) -> Result<LabeledStatementId, ParseError2> {
         debug_assert!(self.cur_scope.is_some());
         let mut scope = self.cur_scope.as_ref().unwrap();
         loop {
             debug_assert!(scope.is_block(), "scope is not a block");
             let relative_scope_pos = ctx.heap[scope.to_block()].relative_pos_in_parent;
             let block = &ctx.heap[scope.to_block()];
             for label_id in &block.labels {
                 let label = &ctx.heap[*label_id];
                 if label.label == *identifier {
                     for local_id in &block.locals {
                         // TODO: Better to do this in control flow analysis, it
                         //  is legal to skip over a variable declaration if it
                         //  is not actually being used. I might be missing
                         //  something here when laying out the bytecode...
                         let local = &ctx.heap[*local_id];
                         if local.relative_pos_in_block > relative_scope_pos && local.relative_pos_in_block < label.relative_pos_in_block {
                             return Err(
                                 ParseError2::new_error(&ctx.module.source, identifier.position, "This target label skips over a variable declaration")
                                     .with_postfixed_info(&ctx.module.source, label.position, "Because it jumps to this label")
                                     .with_postfixed_info(&ctx.module.source, local.position, "Which skips over this variable")
                             );
+                        }
+                    }
                     return Ok(*label_id);
+                }
+            }
             debug_assert!(block.parent_scope.is_some(), "block scope does not have a parent");
             scope = block.parent_scope.as_ref().unwrap();
             if !scope.is_block() {
                 return Err(ParseError2::new_error(&ctx.module.source, identifier.position, "Could not find this label"));
+            }
+        }
+    }
     /// Finds a particular symbol in the symbol table which must correspond to
     /// a definition of a particular type.
     // Note: root_id, symbols and types passed in explicitly to prevent
     //  borrowing errors
     fn find_symbol_of_type<'a>(
         &self, source: &InputSource, root_id: RootId, symbols: &SymbolTable, types: &'a TypeTable,
-        identifier: &NamespacedIdentifier2, expected_type_class: TypeClass
+        identifier: &NamespacedIdentifier, expected_type_class: TypeClass
     ) -> Result<&'a DefinedType, ParseError2> {
         // Find symbol associated with identifier
         let (find_result, _) = find_type_definition(symbols, types, root_id, identifier)
             .as_parse_error(source)?;
         let definition_type_class = find_result.definition.type_class();
         if expected_type_class != definition_type_class {
             return Err(ParseError2::new_error(
                 source, identifier.position,
                 &format!(
                     "Expected to find a {}, this symbol points to a {}",
                     expected_type_class, definition_type_class
+                )
             ))
+        }
         Ok(find_result)
+    }
     /// This function will check if the provided while statement ID has a block
     /// statement that is one of our current parents.
     fn has_parent_while_scope(&self, ctx: &Ctx, id: WhileStatementId) -> bool {
         debug_assert!(self.cur_scope.is_some());
         let mut scope = self.cur_scope.as_ref().unwrap();
         let while_stmt = &ctx.heap[id];
         loop {
             debug_assert!(scope.is_block());
             let block = scope.to_block();
             if while_stmt.body == block.upcast() {
                 return true;
+            }
             let block = &ctx.heap[block];
             debug_assert!(block.parent_scope.is_some(), "block scope does not have a parent");
             scope = block.parent_scope.as_ref().unwrap();
             if !scope.is_block() {
                 return false;
+            }
+        }
+    }
     /// This function should be called while dealing with break/continue
     /// statements. It will try to find the targeted while statement, using the
     /// target label if provided. If a valid target is found then the loop's
     /// ID will be returned, otherwise a parsing error is constructed.
     /// The provided input position should be the position of the break/continue
     /// statement.
     fn resolve_break_or_continue_target(&self, ctx: &Ctx, position: InputPosition, label: &Option<Identifier>) -> Result<WhileStatementId, ParseError2> {
         let target = match label {
             Some(label) => {
                 let target_id = self.find_label(ctx, label)?;
                 // Make sure break target is a while statement
                 let target = &ctx.heap[target_id];
                 if let Statement::While(target_stmt) = &ctx.heap[target.body] {
                     // Even though we have a target while statement, the break might not be
                     // present underneath this particular labeled while statement
                     if !self.has_parent_while_scope(ctx, target_stmt.this) {
                         ParseError2::new_error(&ctx.module.source, label.position, "Break statement is not nested under the target label's while statement")
                             .with_postfixed_info(&ctx.module.source, target.position, "The targeted label is found here");
+                    }
                     target_stmt.this
                 } else {
                     return Err(
                         ParseError2::new_error(&ctx.module.source, label.position, "Incorrect break target label, it must target a while loop")
                             .with_postfixed_info(&ctx.module.source, target.position, "The targeted label is found here")
                     );
+                }
             },
             None => {
                 // Use the enclosing while statement, the break must be
                 // nested within that while statement
                 if self.in_while.is_none() {
                     return Err(
                         ParseError2::new_error(&ctx.module.source, position, "Break statement is not nested under a while loop")
                     );
+                }
                 self.in_while.unwrap()
+            }
         };
         // We have a valid target for the break statement. But we need to
         // make sure we will not break out of a synchronous block
+        {
             let target_while = &ctx.heap[target];
             if target_while.in_sync != self.in_sync {
                 // Break is nested under while statement, so can only escape a
                 // sync block if the sync is nested inside the while statement.
                 debug_assert!(self.in_sync.is_some());
                 let sync_stmt = &ctx.heap[self.in_sync.unwrap()];
                 return Err(
                     ParseError2::new_error(&ctx.module.source, position, "Break may not escape the surrounding synchronous block")
                         .with_postfixed_info(&ctx.module.source, target_while.position, "The break escapes out of this loop")
                         .with_postfixed_info(&ctx.module.source, sync_stmt.position, "And would therefore escape this synchronous block")

0 comments (0 inline, 0 general)