From 10ea99e95c4a8364114562e31b20f6929d9f9eca Mon Sep 17 00:00:00 2001
From: Greg Shuflin <greg.shuflin@protonmail.com>
Date: Fri, 12 Nov 2021 02:02:31 -0800
Subject: [PATCH] rewrite source reference to use raw offsets

---
 schala-lang/language/src/parsing/new.rs |  6 ++---
 schala-lang/language/src/schala.rs      | 31 ++++++++++++++++++-------
 schala-lang/language/src/tokenizing.rs  |  6 +++++
 3 files changed, 32 insertions(+), 11 deletions(-)

diff --git a/schala-lang/language/src/parsing/new.rs b/schala-lang/language/src/parsing/new.rs
index ad87cb3..e149b46 100644
--- a/schala-lang/language/src/parsing/new.rs
+++ b/schala-lang/language/src/parsing/new.rs
@@ -30,7 +30,7 @@ peg::parser! {
             stmt:statement() delimiter()+ { stmt }
 
         rule statement() -> Statement =
-            kind:statement_kind() { Statement { id: Default::default(), location: Default::default(), kind } }
+            pos:position!() kind:statement_kind() { Statement { id: Default::default(), location: pos.into(), kind } }
 
         rule statement_kind() -> StatementKind =
             __ import:import() { StatementKind::Import(import) } /
@@ -382,9 +382,9 @@ peg::parser! {
         rule condition_guard() -> Option<Expression> =
             ("if" _ expr:expression() { expr } )?
 
-        rule expr_or_block() -> Block = block() / ex:expression() {
+        rule expr_or_block() -> Block = block() / pos:position!() ex:expression() {
             Statement {
-                id: Default::default(), location: Default::default(),
+                id: Default::default(), location: pos.into(),
                 kind: StatementKind::Expression(ex)
             }.into()
         }
diff --git a/schala-lang/language/src/schala.rs b/schala-lang/language/src/schala.rs
index e588bb8..2a5d5c6 100644
--- a/schala-lang/language/src/schala.rs
+++ b/schala-lang/language/src/schala.rs
@@ -122,24 +122,39 @@ impl<'a> Schala<'a> {
 
 /// Represents lines of source code
 pub(crate) struct SourceReference {
-    lines: Option<Vec<String>>,
+    last_source: Option<String>,
+    /// Offsets in *bytes* (not chars) representing a newline character
+    newline_offsets: Vec<usize>,
 }
 
 impl SourceReference {
     fn new() -> SourceReference {
-        SourceReference { lines: None }
+        SourceReference { last_source: None, newline_offsets: vec![]}
     }
 
     fn load_new_source(&mut self, source: &str) {
-        //TODO this is a lot of heap allocations - maybe there's a way to make it more efficient?
-        self.lines = Some(source.lines().map(|s| s.to_string()).collect());
+
+        for (offset, ch) in source.as_bytes().iter().enumerate() {
+            if *ch == ('\n' as u8) {
+                self.newline_offsets.push(offset);
+            }
+        }
+        self.last_source = Some(source.to_string());
     }
 
     pub fn get_line(&self, line: usize) -> String {
-        self.lines
-            .as_ref()
-            .and_then(|x| x.get(line).map(|s| s.to_string()))
-            .unwrap_or_else(|| "NO LINE FOUND".to_string())
+        //TODO make sure this is utf8-safe
+        let start_idx = match self.newline_offsets.binary_search(&line) {
+            Ok(index) | Err(index) => index,
+        };
+
+        let last_source = self.last_source.as_ref().unwrap();
+
+        let start = self.newline_offsets[start_idx];
+        let end = self.newline_offsets.get(start_idx + 1).cloned().unwrap_or_else(|| last_source.len());
+
+        let slice = &last_source.as_bytes()[start..end];
+        std::str::from_utf8(slice).unwrap().to_string()
     }
 }
 
diff --git a/schala-lang/language/src/tokenizing.rs b/schala-lang/language/src/tokenizing.rs
index 08cdef5..9abc3ce 100644
--- a/schala-lang/language/src/tokenizing.rs
+++ b/schala-lang/language/src/tokenizing.rs
@@ -18,6 +18,12 @@ pub struct Location {
     pub(crate) offset: usize,
 }
 
+impl From<usize> for Location {
+    fn from(offset: usize) -> Self {
+        Self { offset }
+    }
+}
+
 impl fmt::Display for Location {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         write!(f, "{}", self.offset)