global_ids(): Include module names

Ad-hoc include module names in the global_ids() table. Table values will
have the type_name field set to "module" and their key in the table is
prefixed with "module " to avoid clashes with existing global identifiers
shadowing module names (Management::Node being an existing example).

Closes #3136

NEWS

@@ -35,6 +35,10 @@ New Functionality
   normalize JSON object key names. This can be useful if the keys in a JSON
   object are not valid Zeek identifiers or reserved keywords.
 
+- Module names are now included in ``global_ids()``. Their key in the returned
+  table is prefixed with "module " and their value will have the ``type_name``
+  field set to "module".
+
 Changed Functionality
 ---------------------
 

src/Var.cc

@@ -470,6 +470,18 @@ void add_type(ID* id, TypePtr t, std::unique_ptr<std::vector<AttrPtr>> attr)
 		id->SetAttrs(make_intrusive<Attributes>(std::move(*attr), tnew, false, false));
 	}
 
+static std::set<std::string> all_module_names;
+
+void add_module(const char* module_name)
+	{
+	all_module_names.emplace(module_name);
+	}
+
+const std::set<std::string>& module_names()
+	{
+	return all_module_names;
+	}
+
 static void transfer_arg_defaults(RecordType* args, RecordType* recv)
 	{
 	for ( int i = 0; i < args->NumFields(); ++i )

src/Var.h

@@ -42,6 +42,10 @@ extern ExprPtr add_and_assign_local(IDPtr id, ExprPtr init, ValPtr val);
 
 extern void add_type(ID* id, TypePtr t, std::unique_ptr<std::vector<AttrPtr>> attr);
 
+extern void add_module(const char* module_name);
+
+const std::set<std::string>& module_names();
+
 extern void begin_func(IDPtr id, const char* module_name, FunctionFlavor flavor, bool is_redef,
                        FuncTypePtr t, std::unique_ptr<std::vector<AttrPtr>> attrs = nullptr);
 

src/parse.y

@@ -157,8 +157,6 @@ static int func_hdr_cond_epoch = 0;
 EnumType* cur_enum_type = nullptr;
 static ID* cur_decl_type_id = nullptr;
 
-std::set<std::string> module_names;
-
 static void parse_new_enum(void)
 	{
 	// Starting a new enum definition.
@@ -1357,7 +1355,7 @@ decl:
 		TOK_MODULE TOK_ID ';'
 			{
 			current_module = $2;
-			module_names.insert($2);
+			zeek::detail::add_module($2);
 			zeekygen_mgr->ModuleUsage(::filename, current_module);
 			}
 

src/scan.l

@@ -90,8 +90,6 @@ const char* last_tok_filename = 0;
 const char* last_last_tok_filename = 0;
 char last_tok[128];
 
-extern std::set<std::string> module_names;
-
 #define YY_USER_ACTION	strncpy(last_tok, yytext, sizeof(last_tok) - 1); \
                         last_last_tok_filename = last_tok_filename; \
                         last_tok_filename = ::filename;
@@ -882,7 +880,7 @@ void do_atifdef(const char* id)
 	// If the ID didn't exist, look to see if this is a module name instead.
 	bool found = (i != nullptr);
 	if ( ! found )
-		found = (module_names.count(id) != 0);
+		found = (zeek::detail::module_names().count(id) != 0);
 
 	if ( ! found )
 		begin_ignoring();
@@ -897,7 +895,7 @@ void do_atifndef(const char *id)
 	// If the ID didn't exist, look to see if this is a module name instead.
 	bool found = (i != nullptr);
 	if ( ! found )
-		found = (module_names.count(id) != 0);
+		found = (zeek::detail::module_names().count(id) != 0);
 
 	if ( found )
 		begin_ignoring();

src/zeek.bif

@@ -2106,17 +2106,20 @@ function val_footprint%(v: any%): count
 ## exported, a constant, an enum constant, redefinable, and its value (if it
 ## has one).
 ##
+## Module names are included in the returned table as well. The ``type_name``
+## field is set to  "module" and their names are prefixed with "module " to avoid
+## clashing with global identifiers. Note that there is no module type in Zeek.
+##
 ## Returns: A table that maps identifier names to information about them.
 function global_ids%(%): id_table
 	%{
 	static auto id_table = zeek::id::find_type<zeek::TableType>("id_table");
-	auto ids = zeek::make_intrusive<zeek::TableVal>(id_table);
+	static auto script_id = zeek::id::find_type<zeek::RecordType>("script_id");
-	const auto& globals = zeek::detail::global_scope()->Vars();
 
-	for ( const auto& global : globals )
+	auto ids = zeek::make_intrusive<zeek::TableVal>(id_table);
+
+	for ( const auto& [_, id] : zeek::detail::global_scope()->Vars() )
 		{
-		const auto& id = global.second;
-		static auto script_id = zeek::id::find_type<zeek::RecordType>("script_id");
 		auto rec = zeek::make_intrusive<zeek::RecordVal>(script_id);
 		rec->Assign(0, type_name(id->GetType()->Tag()));
 		rec->Assign(1, id->IsExport());
@@ -2133,6 +2136,23 @@ function global_ids%(%): id_table
 		ids->Assign(std::move(id_name), std::move(rec));
 		}
 
+	auto module_str = zeek::make_intrusive<zeek::StringVal>("module");
+
+	for ( const auto& module_name : zeek::detail::module_names() )
+		{
+		auto rec = zeek::make_intrusive<zeek::RecordVal>(script_id);
+		rec->Assign(0, module_str); // type_name
+		rec->Assign(1, false); // exported
+		rec->Assign(2, false); // constant
+		rec->Assign(3, false); // enum_constant
+		rec->Assign(4, false); // option_value
+		rec->Assign(5, false); // redefinable
+		rec->Assign(6, false); // broker_backend
+
+		auto id_name = zeek::make_intrusive<zeek::StringVal>(zeek::util::fmt("module %s", module_name.c_str()));
+		ids->Assign(std::move(id_name), std::move(rec));
+		}
+
 	return ids;
 	%}
 

testing/btest/Baseline/bifs.global_ids-modules/out

@@ -0,0 +1,4 @@
+### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
+module MyModule, [type_name=module, exported=F, constant=F, enum_constant=F, option_value=F, redefinable=F, broker_backend=F, value=<uninitialized>]
+module GLOBAL, [type_name=module, exported=F, constant=F, enum_constant=F, option_value=F, redefinable=F, broker_backend=F, value=<uninitialized>]
+MyModule, [type_name=string, exported=F, constant=F, enum_constant=F, option_value=F, redefinable=F, broker_backend=F, value=shadows module]

testing/btest/bifs/global_ids-modules.zeek

@@ -0,0 +1,33 @@
+# @TEST-DOC: global_ids() also includes information about modules
+#
+# @TEST-EXEC: unset ZEEK_ALLOW_INIT_ERRORS; zeek -b %INPUT 2>&1 >out
+# @TEST-EXEC: btest-diff out
+
+module MyModule;
+
+module GLOBAL;
+
+# This shadows the module
+global MyModule = "shadows module";
+
+event zeek_init()
+	{
+	local a = global_ids();
+
+	# Assert based testing.
+	assert "module GLOBAL" in a;
+	assert a["module GLOBAL"]$type_name == "module";
+
+	assert "module MyModule" in a;
+	assert a["module MyModule"]$type_name == "module";
+
+	assert "MyModule" in a;
+	assert a["MyModule"]$type_name == "string";
+
+	# And a classic baseline test for now, too.
+	for ( [k], v in a )
+		{
+		if ( k in set("module GLOBAL", "module MyModule", "MyModule") )
+			print k, v;
+		}
+	}