checker: stricter []&Type{len:x} check

vlib/os/os.c.v

@@ -949,23 +949,26 @@ pub fn open_append(path string) ?File {
 // Note: this function will NOT return when successfull, since
 // the child process will take control over execution.
 pub fn execvp(cmdpath string, cmdargs []string) ? {
-	mut cargs := []&char{}
-	cargs << &char(cmdpath.str)
-	for i in 0 .. cmdargs.len {
-		cargs << &char(cmdargs[i].str)
+	unsafe {
+		mut cargs := []&char{}
+		cargs << &char(cmdpath.str)
+		for i in 0 .. cmdargs.len {
+			cargs << &char(cmdargs[i].str)
+		}
+		cargs << &char(0)
+		mut res := int(0)
+		$if windows {
+			res = C._execvp(&char(cmdpath.str), cargs.data)
+		} $else {
+			res = C.execvp(&char(cmdpath.str), cargs.data)
+		}
+		if res == -1 {
+			return error_with_code(posix_get_error_msg(C.errno), C.errno)
+		}
+
+		// just in case C._execvp returned ... that happens on windows ...
+		exit(res)
 	}
-	cargs << &char(0)
-	mut res := int(0)
-	$if windows {
-		res = C._execvp(&char(cmdpath.str), cargs.data)
-	} $else {
-		res = C.execvp(&char(cmdpath.str), cargs.data)
-	}
-	if res == -1 {
-		return error_with_code(posix_get_error_msg(C.errno), C.errno)
-	}
-	// just in case C._execvp returned ... that happens on windows ...
-	exit(res)
 }
 
 // execve - loads and executes a new child process, *in place* of the current process.
@@ -975,27 +978,29 @@ pub fn execvp(cmdpath string, cmdargs []string) ? {
 // Note: this function will NOT return when successfull, since
 // the child process will take control over execution.
 pub fn execve(cmdpath string, cmdargs []string, envs []string) ? {
-	mut cargv := []&char{}
-	mut cenvs := []&char{}
-	cargv << &char(cmdpath.str)
-	for i in 0 .. cmdargs.len {
-		cargv << &char(cmdargs[i].str)
-	}
-	for i in 0 .. envs.len {
-		cenvs << &char(envs[i].str)
-	}
-	cargv << &char(0)
-	cenvs << &char(0)
-	mut res := int(0)
-	$if windows {
-		res = C._execve(&char(cmdpath.str), cargv.data, cenvs.data)
-	} $else {
-		res = C.execve(&char(cmdpath.str), cargv.data, cenvs.data)
-	}
-	// Note: normally execve does not return at all.
-	// If it returns, then something went wrong...
-	if res == -1 {
-		return error_with_code(posix_get_error_msg(C.errno), C.errno)
+	unsafe {
+		mut cargv := []&char{}
+		mut cenvs := []&char{}
+		cargv << &char(cmdpath.str)
+		for i in 0 .. cmdargs.len {
+			cargv << &char(cmdargs[i].str)
+		}
+		for i in 0 .. envs.len {
+			cenvs << &char(envs[i].str)
+		}
+		cargv << &char(0)
+		cenvs << &char(0)
+		mut res := int(0)
+		$if windows {
+			res = C._execve(&char(cmdpath.str), cargv.data, cenvs.data)
+		} $else {
+			res = C.execve(&char(cmdpath.str), cargv.data, cenvs.data)
+		}
+		// Note: normally execve does not return at all.
+		// If it returns, then something went wrong...
+		if res == -1 {
+			return error_with_code(posix_get_error_msg(C.errno), C.errno)
+		}
 	}
 }
 

vlib/sync/pool/pool.v

@@ -81,17 +81,19 @@ pub fn (mut pool PoolProcessor) work_on_pointers(items []voidptr) {
 	if pool.njobs > 0 {
 		njobs = pool.njobs
 	}
-	pool.thread_contexts = []voidptr{len: items.len}
-	pool.results = []voidptr{len: items.len}
-	pool.items = []voidptr{cap: items.len}
-	pool.items << items
-	pool.waitgroup.add(njobs)
-	for i := 0; i < njobs; i++ {
-		if njobs > 1 {
-			go process_in_thread(mut pool, i)
-		} else {
-			// do not run concurrently, just use the same thread:
-			process_in_thread(mut pool, i)
+	unsafe {
+		pool.thread_contexts = []voidptr{len: items.len}
+		pool.results = []voidptr{len: items.len}
+		pool.items = []voidptr{cap: items.len}
+		pool.items << items
+		pool.waitgroup.add(njobs)
+		for i := 0; i < njobs; i++ {
+			if njobs > 1 {
+				go process_in_thread(mut pool, i)
+			} else {
+				// do not run concurrently, just use the same thread:
+				process_in_thread(mut pool, i)
+			}
 		}
 	}
 	pool.waitgroup.wait()
@@ -136,11 +138,13 @@ pub fn (pool &PoolProcessor) get_results<T>() []T {
 
 // get_results_ref - get a list of type safe results in the main thread.
 pub fn (pool &PoolProcessor) get_results_ref<T>() []&T {
-	mut res := []&T{cap: pool.results.len}
-	for i in 0 .. pool.results.len {
-		res << &T(pool.results[i])
+	unsafe {
+		mut res := []&T{cap: pool.results.len}
+		for i in 0 .. pool.results.len {
+			res << &T(pool.results[i])
+		}
+		return res
 	}
-	return res
 }
 
 // set_shared_context - can be called during the setup so that you can

vlib/v/builder/builder.v

@@ -251,17 +251,19 @@ pub fn (mut b Builder) resolve_deps() {
 		eprintln(mods.str())
 		eprintln('-------------------------------')
 	}
-	mut reordered_parsed_files := []&ast.File{}
-	for m in mods {
-		for pf in b.parsed_files {
-			if m == pf.mod.name {
-				reordered_parsed_files << pf
-				// eprintln('pf.mod.name: $pf.mod.name | pf.path: $pf.path')
+	unsafe {
+		mut reordered_parsed_files := []&ast.File{}
+		for m in mods {
+			for pf in b.parsed_files {
+				if m == pf.mod.name {
+					reordered_parsed_files << pf
+					// eprintln('pf.mod.name: $pf.mod.name | pf.path: $pf.path')
+				}
 			}
 		}
+		b.table.modules = mods
+		b.parsed_files = reordered_parsed_files
 	}
-	b.table.modules = mods
-	b.parsed_files = reordered_parsed_files
 }
 
 // graph of all imported modules

vlib/v/checker/checker.v

@@ -271,9 +271,10 @@ pub fn (mut c Checker) check_files(ast_files []&ast.File) {
 	// c.files = ast_files
 	mut has_main_mod_file := false
 	mut has_main_fn := false
+	unsafe {
 	mut files_from_main_module := []&ast.File{}
 	for i in 0 .. ast_files.len {
-		mut file := unsafe { ast_files[i] }
+		mut file := ast_files[i]
 		c.timers.start('checker_check $file.path')
 		c.check(file)
 		if file.mod.name == 'main' {
@@ -302,6 +303,7 @@ pub fn (mut c Checker) check_files(ast_files []&ast.File) {
 			has_main_fn = true
 		}
 	}
+	}
 	c.timers.start('checker_post_process_generic_fns')
 	last_file := c.file
 	// post process generic functions. must be done after all files have been

vlib/v/checker/containers.v

@@ -56,6 +56,13 @@ pub fn (mut c Checker) array_init(mut node ast.ArrayInit) ast.Type {
 			&& c.table.cur_fn.generic_names.len == 0 {
 			c.error('generic struct cannot use in non-generic function', node.pos)
 		}
+
+		// &int{} check
+		if node.elem_type.is_any_kind_of_pointer() && !c.inside_unsafe
+			&& (node.has_len || node.has_cap) {
+			c.warn('arrays of references need to be initialized right away (unless used inside `unsafe`)',
+				node.pos)
+		}
 		return node.typ
 	}
 	if node.is_fixed {

vlib/v/checker/tests/ptr_array_init.out

@@ -0,0 +1,7 @@
+vlib/v/checker/tests/sum.vv:2:14: warning: arrays of references need to be initialized right away (unless used inside `unsafe`)
+    1 | fn main() {
+    2 |     println(*[]&int{len: 1}[0])
+      |              ~~~~~~~
+    3 | }
+    4 |
+

vlib/v/checker/tests/ptr_array_init.vv

@@ -0,0 +1,3 @@
+fn main() {
+    println(*[]&int{len: 1}[0])
+}

vlib/v/gen/c/cgen.v

@@ -4896,14 +4896,16 @@ fn (mut g Gen) write_sorted_types() {
 	defer {
 		g.type_definitions.writeln('// #end sorted_symbols')
 	}
-	mut symbols := []&ast.TypeSymbol{cap: g.table.type_symbols.len} // structs that need to be sorted
-	for sym in g.table.type_symbols {
-		if sym.name !in c.builtins {
-			symbols << sym
+	unsafe {
+		mut symbols := []&ast.TypeSymbol{cap: g.table.type_symbols.len} // structs that need to be sorted
+		for sym in g.table.type_symbols {
+			if sym.name !in c.builtins {
+				symbols << sym
+			}
 		}
+		sorted_symbols := g.sort_structs(symbols)
+		g.write_types(sorted_symbols)
 	}
-	sorted_symbols := g.sort_structs(symbols)
-	g.write_types(sorted_symbols)
 }
 
 fn (mut g Gen) write_types(symbols []&ast.TypeSymbol) {
@@ -5186,11 +5188,13 @@ fn (mut g Gen) sort_structs(typesa []&ast.TypeSymbol) []&ast.TypeSymbol {
 			'\nif you feel this is an error, please create a new issue here: https://github.com/vlang/v/issues and tag @joe-conigliaro')
 	}
 	// sort types
-	mut sorted_symbols := []&ast.TypeSymbol{cap: dep_graph_sorted.nodes.len}
-	for node in dep_graph_sorted.nodes {
-		sorted_symbols << g.table.sym_by_idx(g.table.type_idxs[node.name])
+	unsafe {
+		mut sorted_symbols := []&ast.TypeSymbol{cap: dep_graph_sorted.nodes.len}
+		for node in dep_graph_sorted.nodes {
+			sorted_symbols << g.table.sym_by_idx(g.table.type_idxs[node.name])
+		}
+		return sorted_symbols
 	}
-	return sorted_symbols
 }
 
 [inline]

vlib/v/parser/parser.v

@@ -99,7 +99,7 @@ pub mut:
 	vet_errors []vet.Error
 }
 
-__global codegen_files = []&ast.File{}
+__global codegen_files = unsafe { []&ast.File{} }
 
 // for tests
 pub fn parse_stmt(text string, table &ast.Table, scope &ast.Scope) ast.Stmt {
@@ -425,17 +425,19 @@ pub fn parse_files(paths []string, table &ast.Table, pref &pref.Preferences) []&
 		}
 		*/
 	}
-	mut files := []&ast.File{cap: paths.len}
-	for path in paths {
-		timers.start('parse_file $path')
-		files << parse_file(path, table, .skip_comments, pref)
-		timers.show('parse_file $path')
+	unsafe {
+		mut files := []&ast.File{cap: paths.len}
+		for path in paths {
+			timers.start('parse_file $path')
+			files << parse_file(path, table, .skip_comments, pref)
+			timers.show('parse_file $path')
+		}
+		if codegen_files.len > 0 {
+			files << codegen_files
+			codegen_files.clear()
+		}
+		return files
 	}
-	if codegen_files.len > 0 {
-		files << codegen_files
-		codegen_files.clear()
-	}
-	return files
 }
 
 // codegen allows you to generate V code, so that it can be parsed,

vlib/v/parser/struct.v

@@ -39,7 +39,7 @@ fn (mut p Parser) struct_decl(is_anon bool) ast.StructDecl {
 	if p.disallow_declarations_in_script_mode() {
 		return ast.StructDecl{}
 	}
-	mut name := if is_anon { '' } else { p.check_name()}
+	mut name := if is_anon { '' } else { p.check_name() }
 	if name.len == 1 && name[0].is_capital() {
 		p.error_with_pos('single letter capital names are reserved for generic template types.',
 			name_pos)