Upload 2 files

ChatIPC.cpp

@@ -359,13 +359,107 @@ static std::optional<std::string> get_valid_base_form(const std::string &word) {
         }
     }
 
-    // 4. Plural / 3rd-person singular form (-ies -> -y)
+    // 4. Superlative form (-est)
+    if (word.length() > 3 && word.substr(word.length() - 3) == "est") {
+        // e.g., simplest -> simple (strip 'st')
+        std::string base1 = word.substr(0, word.length() - 2);
+        if (check_dictionary(base1)) return base1;
+
+        // e.g., fastest -> fast
+        std::string base2 = word.substr(0, word.length() - 3);
+        if (check_dictionary(base2)) return base2;
+
+        // e.g., happiest -> happy (-iest -> -y)
+        if (word.length() > 4 && word.substr(word.length() - 4) == "iest") {
+            std::string base_y = word.substr(0, word.length() - 4) + "y";
+            if (check_dictionary(base_y)) return base_y;
+        }
+
+        // e.g., biggest -> big (doubled consonant check)
+        if (base2.length() >= 2 && base2.back() == base2[base2.length() - 2]) {
+            std::string base3 = base2.substr(0, base2.length() - 1);
+            if (check_dictionary(base3)) return base3;
+        }
+    }
+
+    // 5. Comparative form (-er)
+    if (word.length() > 2 && word.substr(word.length() - 2) == "er") {
+        // e.g., simpler -> simple (strip 'r')
+        std::string base1 = word.substr(0, word.length() - 1);
+        if (check_dictionary(base1)) return base1;
+
+        // e.g., faster -> fast
+        std::string base2 = word.substr(0, word.length() - 2);
+        if (check_dictionary(base2)) return base2;
+
+        // e.g., happier -> happy (-ier -> -y)
+        if (word.length() > 3 && word.substr(word.length() - 3) == "ier") {
+            std::string base_y = word.substr(0, word.length() - 3) + "y";
+            if (check_dictionary(base_y)) return base_y;
+        }
+
+        // e.g., bigger -> big (doubled consonant check)
+        if (base2.length() >= 2 && base2.back() == base2[base2.length() - 2]) {
+            std::string base3 = base2.substr(0, base2.length() - 1);
+            if (check_dictionary(base3)) return base3;
+        }
+    }
+
+    // 6. Adverbial form (-ly)
+    if (word.length() > 2 && word.substr(word.length() - 2) == "ly") {
+        // e.g., quickly -> quick
+        std::string base1 = word.substr(0, word.length() - 2);
+        if (check_dictionary(base1)) return base1;
+
+        // e.g., happily -> happy (-ily -> -y)
+        if (word.length() > 3 && word.substr(word.length() - 3) == "ily") {
+            std::string base2 = word.substr(0, word.length() - 3) + "y";
+            if (check_dictionary(base2)) return base2;
+        }
+
+        // e.g., basically -> basic (-ally -> -ic)
+        if (word.length() > 4 && word.substr(word.length() - 4) == "ally") {
+            std::string base3 = word.substr(0, word.length() - 4);
+            if (check_dictionary(base3)) return base3;
+
+            std::string base4 = base3 + "al";
+            if (check_dictionary(base4)) return base4;
+        }
+
+        // e.g., gently -> gentle (-ly -> -le)
+        if (word.back() == 'y' && word.length() > 2 && word[word.length() - 2] == 'l') {
+            std::string base5 = word.substr(0, word.length() - 1) + "e";
+            if (check_dictionary(base5)) return base5;
+        }
+    }
+
+    // 7. Noun suffix (-ness)
+    if (word.length() > 4 && word.substr(word.length() - 4) == "ness") {
+        // e.g., sadness -> sad
+        std::string base1 = word.substr(0, word.length() - 4);
+        if (check_dictionary(base1)) return base1;
+
+        // e.g., happiness -> happy (-iness -> -y)
+        if (base1.length() > 1 && base1.back() == 'i') {
+            std::string base2 = base1.substr(0, base1.length() - 1) + "y";
+            if (check_dictionary(base2)) return base2;
+        }
+    }
+
+    // 8. Noun suffix (-ment)
+    if (word.length() > 4 && word.substr(word.length() - 4) == "ment") {
+        // e.g., development -> develop
+        std::string base1 = word.substr(0, word.length() - 4);
+        if (check_dictionary(base1)) return base1;
+    }
+
+    // 9. Plural / 3rd-person singular form (-ies -> -y)
     if (word.length() > 3 && word.substr(word.length() - 3) == "ies") {
         std::string base = word.substr(0, word.length() - 3) + "y";
         if (check_dictionary(base)) return base;
     }
 
-    // 5. Plural / 3rd-person singular form (-ves -> -f / -fe)
+    // 10. Plural / 3rd-person singular form (-ves -> -f / -fe)
     if (word.length() > 3 && word.substr(word.length() - 3) == "ves") {
         std::string base1 = word.substr(0, word.length() - 3) + "f";
         if (check_dictionary(base1)) return base1;
@@ -374,13 +468,13 @@ static std::optional<std::string> get_valid_base_form(const std::string &word) {
         if (check_dictionary(base2)) return base2;
     }
 
-    // 6. Plural form (-es)
+    // 11. Plural form (-es)
     if (word.length() > 2 && word.substr(word.length() - 2) == "es") {
         std::string base = word.substr(0, word.length() - 2);
         if (check_dictionary(base)) return base;
     }
 
-    // 7. Plural form (-s)
+    // 12. Plural form (-s)
     if (word.length() > 1 && word.back() == 's' && word[word.length() - 2] != 's') {
         std::string base = word.substr(0, word.length() - 1);
         if (check_dictionary(base)) return base;
@@ -545,19 +639,52 @@ static void build_def_tokens_cache(){
     global_pos_cache.reserve(global_dictionary_entries.size());
 
     for (const auto &entry : global_dictionary_entries){
-        const std::string key = normalize_dictionary_key(entry.word);
-        if (key.empty()) continue;
+        // 1. Tokenize definitions once per entry to avoid redundant processing
+        std::vector<std::string> all_def_toks;
+        for (const auto &def : entry.definitions){
+            auto toks = tokenize_others(def);
+            all_def_toks.insert(all_def_toks.end(), toks.begin(), toks.end());
+        }
 
         std::string pos = normalize_pos_tag(entry.pos);
-        if (!pos.empty()) global_pos_cache[key].push_back(std::move(pos));
 
-        auto &defs = global_def_tokens_cache[key];
-        for (const auto &def : entry.definitions){
-            auto toks = tokenize_others(def);
-            defs.insert(defs.end(), toks.begin(), toks.end());
+        // 2. Split entry.word by semicolons or commas
+        std::vector<std::string> sub_words;
+        std::string current_sub;
+        for (char c : entry.word) {
+            if (c == ';' || c == ',') {
+                if (!current_sub.empty()) {
+                    sub_words.push_back(current_sub);
+                    current_sub.clear();
+                }
+            } else {
+                current_sub.push_back(c);
+            }
+        }
+        if (!current_sub.empty()) {
+            sub_words.push_back(current_sub);
+        }
+
+        // 3. Register each sub-word to the cache
+        for (size_t i = 0; i < sub_words.size(); ++i) {
+            const std::string key = normalize_dictionary_key(sub_words[i]);
+            if (key.empty()) continue;
+
+            if (!pos.empty()) {
+                // Move on the final sub-word, copy otherwise
+                if (i == sub_words.size() - 1) {
+                    global_pos_cache[key].push_back(std::move(pos));
+                } else {
+                    global_pos_cache[key].push_back(pos);
+                }
+            }
+
+            auto &defs = global_def_tokens_cache[key];
+            defs.insert(defs.end(), all_def_toks.begin(), all_def_toks.end());
         }
     }
 
+    // 4. Keep existing sorting and deduplication logic intact
     for (auto &pr : global_def_tokens_cache){
         auto &v = pr.second;
         std::sort(v.begin(), v.end());