azookey conversion accuracy improvement

src-tauri/Cargo.lock

@@ -1904,7 +1904,7 @@ version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e76a40b972b250590567018fd753f44adedc270994fb7be37e4d8c20764bab49"
 dependencies = [
- "itertools",
+ "itertools 0.10.5",
 ]
 
 [[package]]
@@ -1916,6 +1916,15 @@ dependencies = [
  "either",
 ]
 
+[[package]]
+name = "itertools"
+version = "0.14.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2b192c782037fadd9cfa75548310488aabdbf3d2da73885b31bd0abd03351285"
+dependencies = [
+ "either",
+]
+
 [[package]]
 name = "itoa"
 version = "0.4.8"
@@ -2050,7 +2059,7 @@ dependencies = [
  "diff",
  "ena",
  "is-terminal",
- "itertools",
+ "itertools 0.10.5",
  "lalrpop-util",
  "petgraph",
  "regex",
@@ -4960,6 +4969,7 @@ dependencies = [
  "clipboard",
  "clipboard-master",
  "ipc-channel",
+ "itertools 0.14.0",
  "once_cell",
  "platform-dirs",
  "regex",
@@ -5019,7 +5029,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a74666202acfcb4f9b995be2e3e9f7f530deb65e05a1407b8d0b30c9c451238a"
 dependencies = [
  "fnv",
- "itertools",
+ "itertools 0.10.5",
  "lazy_static",
 ]
 

src-tauri/Cargo.toml

@@ -35,6 +35,7 @@ tokio = { version = "1.44.2", features = ["full"] }
 ipc-channel = "0.19.0"
 zip = "2.6.1"
 wana_kana = "4.0.0"
+itertools = "0.14.0"
 
 [dependencies.tracing-subscriber]
 version = "0.3.16"

src-tauri/src/azookey/server.rs

@@ -2,6 +2,7 @@ use std::{collections::HashMap, sync::LazyLock};
 
 use azookey_binding::{Candidate, ComposingText, KanaKanjiConverter};
 use ipc_channel::ipc::IpcOneShotServer;
+use itertools::Itertools;
 use platform_dirs::AppDirs;
 
 use super::IpcMessage;
@@ -61,6 +62,56 @@ impl AzookeyConversionServer {
         instance
     }
 
+    fn pre_process_text(text: &str) -> String {
+        let mut result = String::new();
+
+        // replace all characters in the text with their corresponding replacements
+        for c in text.chars() {
+            if let Some(&replacement) = SIGNMAP.get(c.to_string().as_str()) {
+                result.push_str(replacement);
+            } else {
+                result.push(c);
+            }
+        }
+
+        // push 'n' if the last and second last characters are 'n'
+        if result.ends_with('n') {
+            let mut chars = result.chars().collect::<Vec<_>>();
+            if chars.len() > 1 && chars[chars.len() - 2] != 'n' {
+                chars.push('n');
+            }
+            result = chars.into_iter().collect::<String>();
+        }
+
+        // push '§' at the end of the string to avoid unnecessary prediction
+        result.push_str("§");
+
+        result
+    }
+
+    fn post_process_text(text: &str) -> String {
+        let mut result = text.to_string();
+
+        if result.ends_with('§') {
+            result.pop();
+        }
+
+        result
+    }
+
+    fn post_process_candidates(candidates: Vec<Candidate>) -> Vec<Candidate> {
+        candidates
+            .iter()
+            .take(8)
+            .map(|c| {
+                let mut candidate = c.clone();
+                candidate.text = Self::post_process_text(&candidate.text);
+                candidate
+            })
+            .unique_by(|c| c.text.clone())
+            .collect()
+    }
+
     pub fn server_loop(mut self) {
         let (a, _) = self.server.accept().unwrap();
         let mut sender = None;
@@ -74,16 +125,7 @@ impl AzookeyConversionServer {
                     self.composing_text = ComposingText::new();
                 }
                 Ok(IpcMessage::InsertAtCursorPosition(text)) => {
-                    let text: String = text
+                    let text = Self::pre_process_text(&text);
-                        .chars()
-                        .map(|c| {
-                            if let Some(&replacement) = SIGNMAP.get(c.to_string().as_str()) {
-                                return replacement.to_string();
-                            } else {
-                                c.to_string()
-                            }
-                        })
-                        .collect();
                     self.composing_text.insert_at_cursor_position(&text);
                 }
                 Ok(IpcMessage::RequestCandidates(context, weight_path)) => {
@@ -99,11 +141,7 @@ impl AzookeyConversionServer {
                         &weight_path,
                     );
                     if let Some(s) = sender.as_ref() {
-                        let candidates = candidates
+                        let candidates = Self::post_process_candidates(candidates);
-                            .iter()
-                            .take(8)
-                            .map(|c| c.clone())
-                            .collect::<Vec<Candidate>>();
                         s.send(IpcMessage::Candidates(candidates)).unwrap();
                     }
                 }