jinja : fix split and replace with empty first arg (ggml-org#24574)

* fix split and replace with empty first arg

* fix reserve size

Author: CISC

common/jinja/value.cpp

@@ -673,6 +673,9 @@ const func_builtins & value_string_t::get_builtins() const {
             std::string str = val_input->as_string().str();
             // FIXME: Support non-specified delimiter (split on consecutive (no leading or trailing) whitespace)
             std::string delim = (args.count() > 1) ? args.get_pos(1)->as_string().str() : " ";
+            if (delim.empty()) {
+                throw raised_exception("empty separator");
+            }
             int64_t maxsplit = (args.count() > 2) ? args.get_pos(2)->as_int() : -1;
             auto result = mk_val<value_array>();
             size_t pos = 0;
@@ -697,6 +700,9 @@ const func_builtins & value_string_t::get_builtins() const {
             std::string str = val_input->as_string().str();
             // FIXME: Support non-specified delimiter (split on consecutive (no leading or trailing) whitespace)
             std::string delim = (args.count() > 1) ? args.get_pos(1)->as_string().str() : " ";
+            if (delim.empty()) {
+                throw raised_exception("empty separator");
+            }
             int64_t maxsplit = (args.count() > 2) ? args.get_pos(2)->as_int() : -1;
             auto result = mk_val<value_array>();
             size_t pos = 0;
@@ -722,10 +728,23 @@ const func_builtins & value_string_t::get_builtins() const {
             if (count > 0) {
                 throw not_implemented_exception("String replace with count argument not implemented");
             }
-            size_t pos = 0;
-            while ((pos = str.find(old_str, pos)) != std::string::npos) {
-                str.replace(pos, old_str.length(), new_str);
-                pos += new_str.length();
+            if (old_str != new_str) {
+                size_t pos = 0;
+                if (old_str.empty()) {
+                    std::string new_res;
+                    new_res.reserve(str.length() + new_str.length() * (str.length() + 1));
+                    new_res += new_str;
+                    for (const char c : str) {
+                        new_res.push_back(c);
+                        new_res += new_str;
+                    }
+                    str = new_res;
+                } else {
+                    while ((pos = str.find(old_str, pos)) != std::string::npos) {
+                        str.replace(pos, old_str.length(), new_str);
+                        pos += new_str.length();
+                    }
+                }
             }
             auto res = mk_val<value_string>(str);
             res->val_str.mark_input_based_on(args.get_pos(0)->val_str);

tests/test-jinja.cpp

@@ -1320,6 +1320,12 @@ static void test_string_methods(testing & t) {
         "hello jinja"
     );
 
+    test_template(t, "string.replace() empty",
+        "{{ s.replace('', '.') }}",
+        {{"s", "hello world"}},
+        ".h.e.l.l.o. .w.o.r.l.d."
+    );
+
     test_template(t, "string.replace() with count",
         "{{ s.replace('a', 'X', 2) }}",
         {{"s", "banana"}},