jinja: correct stats for tojson and string filters (#19785)

Author: ngxson

common/jinja/runtime.cpp

@@ -85,7 +85,7 @@ value identifier::execute_impl(context & ctx) {
     auto builtins = global_builtins();
     if (!it->is_undefined()) {
         if (ctx.is_get_stats) {
-            it->stats.used = true;
+            value_t::stats_t::mark_used(it);
         }
         JJ_DEBUG("Identifier '%s' found, type = %s", val.c_str(), it->type().c_str());
         return it;
@@ -277,7 +277,7 @@ value binary_expression::execute_impl(context & ctx) {
 static value try_builtin_func(context & ctx, const std::string & name, value & input, bool undef_on_missing = false) {
     JJ_DEBUG("Trying built-in function '%s' for type %s", name.c_str(), input->type().c_str());
     if (ctx.is_get_stats) {
-        input->stats.used = true;
+        value_t::stats_t::mark_used(input);
         input->stats.ops.insert(name);
     }
     auto builtins = input->get_builtins();
@@ -448,7 +448,7 @@ value for_statement::execute_impl(context & ctx) {
 
     // mark the variable being iterated as used for stats
     if (ctx.is_get_stats) {
-        iterable_val->stats.used = true;
+        value_t::stats_t::mark_used(iterable_val);
         iterable_val->stats.ops.insert("array_access");
     }
 
@@ -470,7 +470,7 @@ value for_statement::execute_impl(context & ctx) {
             items.push_back(std::move(tuple));
         }
         if (ctx.is_get_stats) {
-            iterable_val->stats.used = true;
+            value_t::stats_t::mark_used(iterable_val);
             iterable_val->stats.ops.insert("object_access");
         }
     } else {
@@ -480,7 +480,7 @@ value for_statement::execute_impl(context & ctx) {
             items.push_back(item);
         }
         if (ctx.is_get_stats) {
-            iterable_val->stats.used = true;
+            value_t::stats_t::mark_used(iterable_val);
             iterable_val->stats.ops.insert("array_access");
         }
     }
@@ -817,8 +817,9 @@ value member_expression::execute_impl(context & ctx) {
     }
 
     if (ctx.is_get_stats && val && object && property) {
-        val->stats.used = true;
-        object->stats.used = true;
+        value_t::stats_t::mark_used(val);
+        value_t::stats_t::mark_used(object);
+        value_t::stats_t::mark_used(property);
         if (is_val<value_int>(property)) {
             object->stats.ops.insert("array_access");
         } else if (is_val<value_string>(property)) {

common/jinja/value.cpp

@@ -161,6 +161,11 @@ static value tojson(const func_args & args) {
     value val_separators = args.get_kwarg_or_pos("separators",   3);
     value val_sort       = args.get_kwarg_or_pos("sort_keys",    4);
     int indent = -1;
+    if (args.ctx.is_get_stats) {
+        // mark as used (recursively) for stats
+        auto val_input = args.get_pos(0);
+        value_t::stats_t::mark_used(const_cast<value&>(val_input), true);
+    }
     if (is_val<value_int>(val_indent)) {
         indent = static_cast<int>(val_indent->as_int());
     }
@@ -891,6 +896,11 @@ const func_builtins & value_array_t::get_builtins() const {
         }},
         {"string", [](const func_args & args) -> value {
             args.ensure_vals<value_array>();
+            if (args.ctx.is_get_stats) {
+                // mark as used (recursively) for stats
+                auto val_input = args.get_pos(0);
+                value_t::stats_t::mark_used(const_cast<value&>(val_input), true);
+            }
             return mk_val<value_string>(args.get_pos(0)->as_string());
         }},
         {"tojson", tojson},
@@ -1046,6 +1056,11 @@ const func_builtins & value_object_t::get_builtins() const {
         {"tojson", tojson},
         {"string", [](const func_args & args) -> value {
             args.ensure_vals<value_object>();
+            if (args.ctx.is_get_stats) {
+                // mark as used (recursively) for stats
+                auto val_input = args.get_pos(0);
+                value_t::stats_t::mark_used(const_cast<value&>(val_input), true);
+            }
             return mk_val<value_string>(args.get_pos(0)->as_string());
         }},
         {"length", [](const func_args & args) -> value {
@@ -1358,4 +1373,21 @@ std::string value_to_string_repr(const value & val) {
     }
 }
 
+// stats utility
+void value_t::stats_t::mark_used(value & val, bool deep) {
+    val->stats.used = true;
+    if (deep) {
+        if (is_val<value_array>(val)) {
+            for (auto & item : val->val_arr) {
+                mark_used(item, deep);
+            }
+        } else if (is_val<value_object>(val)) {
+            for (auto & pair : val->val_obj) {
+                mark_used(pair.first, deep);
+                mark_used(pair.second, deep);
+            }
+        }
+    }
+}
+
 } // namespace jinja

common/jinja/value.h

@@ -118,6 +118,8 @@ struct value_t {
         bool used = false;
         // ops can be builtin calls or operators: "array_access", "object_access"
         std::set<std::string> ops;
+        // utility to recursively mark value and its children as used
+        static void mark_used(value & val, bool deep = false);
     } stats;
 
     value_t() = default;

tests/test-jinja.cpp

@@ -32,6 +32,7 @@ static void test_string_methods(testing & t);
 static void test_array_methods(testing & t);
 static void test_object_methods(testing & t);
 static void test_hasher(testing & t);
+static void test_stats(testing & t);
 static void test_fuzzing(testing & t);
 
 static bool g_python_mode = false;
@@ -70,6 +71,7 @@ int main(int argc, char *argv[]) {
     t.test("object methods", test_object_methods);
     if (!g_python_mode) {
         t.test("hasher", test_hasher);
+        t.test("stats", test_stats);
         t.test("fuzzing", test_fuzzing);
     }
 
@@ -1795,6 +1797,63 @@ static void test_hasher(testing & t) {
     });
 }
 
+static void test_stats(testing & t) {
+    static auto get_stats = [](const std::string & tmpl, const json & vars) -> jinja::value {
+        jinja::lexer lexer;
+        auto lexer_res = lexer.tokenize(tmpl);
+
+        jinja::program prog = jinja::parse_from_tokens(lexer_res);
+
+        jinja::context ctx(tmpl);
+        jinja::global_from_json(ctx, json{{ "val", vars }}, true);
+        ctx.is_get_stats = true;
+
+        jinja::runtime runtime(ctx);
+        runtime.execute(prog);
+
+        return ctx.get_val("val");
+    };
+
+    t.test("stats", [](testing & t) {
+        jinja::value val = get_stats(
+            "{{val.num}} "
+            "{{val.str}} "
+            "{{val.arr[0]}} "
+            "{{val.obj.key1}} "
+            "{{val.nested | tojson}}",
+            // Note: the json below will be wrapped inside "val" in the context
+            json{
+                {"num", 1},
+                {"str", "abc"},
+                {"arr", json::array({1, 2, 3})},
+                {"obj", json::object({{"key1", 1}, {"key2", 2}, {"key3", 3}})},
+                {"nested", json::object({
+                    {"inner_key1", json::array({1, 2})},
+                    {"inner_key2", json::object({{"a", "x"}, {"b", "y"}})}
+                })},
+                {"mixed", json::object({
+                    {"used", 1},
+                    {"unused", 2},
+                })},
+            }
+        );
+
+        t.assert_true("num is used", val->at("num")->stats.used);
+        t.assert_true("str is used", val->at("str")->stats.used);
+
+        t.assert_true("arr is used", val->at("arr")->stats.used);
+        t.assert_true("arr[0] is used", val->at("arr")->at(0)->stats.used);
+        t.assert_true("arr[1] is not used", !val->at("arr")->at(1)->stats.used);
+
+        t.assert_true("obj is used", val->at("obj")->stats.used);
+        t.assert_true("obj.key1 is used", val->at("obj")->at("key1")->stats.used);
+        t.assert_true("obj.key2 is not used", !val->at("obj")->at("key2")->stats.used);
+
+        t.assert_true("inner_key1[0] is used", val->at("nested")->at("inner_key1")->at(0)->stats.used);
+        t.assert_true("inner_key2.a is used", val->at("nested")->at("inner_key2")->at("a")->stats.used);
+    });
+}
+
 static void test_template_cpp(testing & t, const std::string & name, const std::string & tmpl, const json & vars, const std::string & expect) {
     t.test(name, [&tmpl, &vars, &expect](testing & t) {
         jinja::lexer lexer;