Improve API

backend/reconcile/src/diffs/myers.rs

@@ -30,7 +30,7 @@ use crate::{
     utils::{common_prefix_len::common_prefix_len, common_suffix_len::common_suffix_len},
 };
 
-/// Myers' diff algorithm with deadline.
+/// Myers' diff algorithm.
 ///
 /// Diff `old`, between indices `old_range` and `new` between indices
 /// `new_range`.

backend/reconcile/src/diffs/raw_operation.rs

@@ -30,12 +30,12 @@ where
 
     pub fn is_left_joinable(&self) -> bool {
         let first_token = self.tokens().first();
-        first_token.is_none_or(super::super::tokenizer::token::Token::get_is_left_joinable)
+        first_token.is_none_or(|token| token.is_left_joinable)
     }
 
     pub fn is_right_joinable(&self) -> bool {
         let last_token = self.tokens().last();
-        last_token.is_none_or(super::super::tokenizer::token::Token::get_is_right_joinable)
+        last_token.is_none_or(|token| token.is_right_joinable)
     }
 
     /// Extends the operation with another operation. Only operations of the
@@ -49,8 +49,8 @@ where
         );
 
         match (self, other) {
-            (RawOperation::Insert(tokens1), RawOperation::Insert(tokens2)) => {
+            (RawOperation::Insert(self_tokens), RawOperation::Insert(other_tokens)) => {
-                RawOperation::Insert(tokens1.into_iter().chain(tokens2).collect())
+                RawOperation::Insert(self_tokens.into_iter().chain(other_tokens).collect())
             }
             (RawOperation::Delete(tokens1), RawOperation::Delete(tokens2)) => {
                 RawOperation::Delete(tokens1.into_iter().chain(tokens2).collect())

backend/reconcile/src/tokenizer/token.rs

@@ -2,8 +2,12 @@
 use serde::{Deserialize, Serialize};
 
 /// A token is a string that has been normalised in some way.
-/// The normalised form is used for comparison, while the original form is used
+///
-/// for applying `Operation`-s.
+/// It's UTF-8 compatible.
+///
+/// A token consists of the normalised form is used for comparison, and the
+/// original form used for subsequently applying `Operation`-s to a text
+/// document.
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
 #[derive(Debug, Clone)]
 pub struct Token<T>
@@ -11,18 +15,20 @@ where
     T: PartialEq + Clone + std::fmt::Debug,
 {
     /// The normalised form of the token used deriving the diff.
-    pub normalised: T,
+    normalised: T,
 
     /// The original string, that should be inserted or deleted in the document.
     original: String,
 
-    /// Whether the token is joinable with the previous token.
+    /// Whether the token is semantically joinable with the previous token.
-    is_left_joinable: bool,
+    pub is_left_joinable: bool,
 
-    /// Whether the token is joinable with the next token.
+    /// Whether the token is semantically joinable with the next token.
-    is_right_joinable: bool,
+    pub is_right_joinable: bool,
 }
 
+/// Trivial implementation of Token when the normalised form is the same as the
+/// original string.
 impl From<&str> for Token<String> {
     fn from(text: &str) -> Self { Token::new(text.to_owned(), text.to_owned(), true, true) }
 }
@@ -47,13 +53,11 @@ where
 
     pub fn original(&self) -> &str { &self.original }
 
+    pub fn set_normalised(&mut self, normalised: T) { self.normalised = normalised; }
+
     pub fn normalised(&self) -> &T { &self.normalised }
 
     pub fn get_original_length(&self) -> usize { self.original.chars().count() }
-
-    pub fn get_is_left_joinable(&self) -> bool { self.is_left_joinable }
-
-    pub fn get_is_right_joinable(&self) -> bool { self.is_right_joinable }
 }
 
 impl<T> PartialEq for Token<T>

backend/reconcile/src/tokenizer/word_tokenizer.rs

@@ -1,7 +1,7 @@
 use super::token::Token;
 
 /// Splits on word boundaries creating alternating words and whitespaces with
-/// the whitesspaces getting unique IDs.
+/// the whitespaces getting unique IDs.
 ///
 /// ## Example
 ///
@@ -34,7 +34,8 @@ pub fn word_tokenizer(text: &str) -> Vec<Token<String>> {
 
     for i in 0..result.len() - 1 {
         if result[i].original().chars().all(char::is_whitespace) {
-            result[i].normalised = result[i].normalised().to_owned() + result[i + 1].original();
+            let normalised = result[i].normalised().to_owned() + result[i + 1].original();
+            result[i].set_normalised(normalised);
         }
     }