vieter-v
/
v
2
0
Fork 0
Code Issues Pull Requests Releases Activity

tests: improved test output formatting

pull/3270/head
Delyan Angelov 2019-12-30 06:23:54 +02:00 committed by Alexander Medvednikov
parent 4f173c8900
commit a0f32f5c29
11 changed files with 144 additions and 95 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
tools/modules/testing/common.v
View File

@ -14,8 +14,6 @@ pub mut:
vargs string
failed bool
benchmark benchmark.Benchmark
ok string
fail string
}
pub fn new_test_session(vargs string) TestSession {
@ -34,8 +32,6 @@ pub fn vexe_path() string {
}
pub fn (ts mut TestSession) init() {
ts.ok = term.ok_message('OK')
ts.fail = term.fail_message('FAIL')
ts.benchmark = benchmark.new_benchmark()
}
@ -43,6 +39,7 @@ pub fn (ts mut TestSession) test() {
tmpd := os.tmpdir()
ts.init()
show_stats := '-stats' in ts.vargs.split(' ')
mut remaining_files := []string
for dot_relative_file in ts.files {
relative_file := dot_relative_file.replace('./', '')
file := os.realpath(relative_file)
@ -66,6 +63,12 @@ pub fn (ts mut TestSession) test() {
continue
}
}
remaining_files << dot_relative_file
}
ts.benchmark.set_total_expected_steps(remaining_files.len)
for dot_relative_file in remaining_files {
relative_file := dot_relative_file.replace('./', '')
file := os.realpath(relative_file)
// Ensure that the generated binaries will be stored in the temporary folder.
// Remove them after a test passes/fails.
fname := filepath.filename(file)
@ -97,17 +100,17 @@ pub fn (ts mut TestSession) test() {
r := os.exec(cmd) or {
ts.benchmark.fail()
ts.failed = true
eprintln(ts.benchmark.step_message('$relative_file ${ts.fail}'))
eprintln(ts.benchmark.step_message_fail(relative_file))
continue
}
if r.exit_code != 0 {
ts.benchmark.fail()
ts.failed = true
eprintln(ts.benchmark.step_message('$relative_file ${ts.fail}\n`$file`\n (\n$r.output\n)'))
eprintln(ts.benchmark.step_message_fail('${relative_file}\n`$file`\n (\n$r.output\n)'))
}
else {
ts.benchmark.ok()
eprintln(ts.benchmark.step_message('$relative_file ${ts.ok}'))
eprintln(ts.benchmark.step_message_ok(relative_file))
}
}
if os.exists(generated_binary_fpath) {
@ -171,19 +174,17 @@ pub fn building_any_v_binaries_failed() bool {
'$vexe -o v_prod -prod v.v',
]
mut bmark := benchmark.new_benchmark()
bok := term.ok_message('OK')
bfail := term.fail_message('FAIL')
for cmd in v_build_commands {
bmark.step()
if build_v_cmd_failed(cmd) {
bmark.fail()
failed = true
eprintln(bmark.step_message('$cmd => ${bfail} . See details above ^^^^^^^'))
eprintln(bmark.step_message_fail('command: ${cmd} . See details above ^^^^^^^'))
eprintln('')
continue
}
bmark.ok()
eprintln(bmark.step_message('$cmd => ${bok}'))
eprintln(bmark.step_message_ok('command: ${cmd}'))
}
bmark.stop()
eprintln(term.h_divider())

52
vlib/benchmark/benchmark.v
View File

@ -26,6 +26,11 @@ println( bmark.total_message('remarks about the benchmark') )
*/
const (
BOK = term.ok_message('OK')
BFAIL = term.fail_message('FAIL')
)
pub struct Benchmark {
pub mut:
bench_start_time i64
@ -36,6 +41,10 @@ pub mut:
nok int
nfail int
verbose bool
nexpected_steps int
cstep int
bok string
bfail string
}
pub fn new_benchmark() Benchmark {
@ -45,12 +54,17 @@ pub fn new_benchmark() Benchmark {
}
}
pub fn (b mut Benchmark) set_total_expected_steps(n int) {
b.nexpected_steps = n
}
pub fn (b mut Benchmark) stop() {
b.bench_end_time = benchmark.now()
}
pub fn (b mut Benchmark) step() {
b.step_start_time = benchmark.now()
b.cstep++
}
pub fn (b mut Benchmark) fail() {
@ -81,16 +95,45 @@ pub fn (b mut Benchmark) neither_fail_nor_ok() {
b.step_end_time = benchmark.now()
}
pub fn (b &Benchmark) step_message_with_label(label string, msg string) string {
mut timed_line := ''
if b.nexpected_steps > 0 {
mut sprogress := ''
if b.nexpected_steps < 10 {
sprogress = '${b.cstep:1d}/${b.nexpected_steps:1d}'
}
if b.nexpected_steps >= 10 && b.nexpected_steps < 100 {
sprogress = '${b.cstep:2d}/${b.nexpected_steps:2d}'
}
if b.nexpected_steps >= 100 && b.nexpected_steps < 1000 {
sprogress = '${b.cstep:3d}/${b.nexpected_steps:3d}'
}
timed_line = b.tdiff_in_ms('[${sprogress}] $msg', b.step_start_time, b.step_end_time)
}
else {
timed_line = b.tdiff_in_ms(msg, b.step_start_time, b.step_end_time)
}
return '${label:-5s}${timed_line}'
}
pub fn (b &Benchmark) step_message(msg string) string {
return b.tdiff_in_ms(msg, b.step_start_time, b.step_end_time)
return b.step_message_with_label('', msg)
}
pub fn (b &Benchmark) step_message_ok(msg string) string {
return b.step_message_with_label(BOK, msg)
}
pub fn (b &Benchmark) step_message_fail(msg string) string {
return b.step_message_with_label(BFAIL, msg)
}
pub fn (b &Benchmark) total_message(msg string) string {
mut tmsg := '$msg \n ok, fail, total = ' + term.ok_message('${b.nok:5d}') + ', ' + if b.nfail > 0 { term.fail_message('${b.nfail:5d}') } else { '${b.nfail:5d}' } + ', ' + '${b.ntotal:5d}'
mut tmsg := '${msg}\n ok, fail, total = ' + term.ok_message('${b.nok:5d}') + ', ' + if b.nfail > 0 { term.fail_message('${b.nfail:5d}') } else { '${b.nfail:5d}' } + ', ' + '${b.ntotal:5d}'
if b.verbose {
tmsg = '<=== total time spent $tmsg'
}
return b.tdiff_in_ms(tmsg, b.bench_start_time, b.bench_end_time)
return ' ' + b.tdiff_in_ms(tmsg, b.bench_start_time, b.bench_end_time)
}
pub fn (b &Benchmark) total_duration() i64 {
@ -101,7 +144,7 @@ pub fn (b &Benchmark) total_duration() i64 {
fn (b &Benchmark) tdiff_in_ms(s string, sticks i64, eticks i64) string {
if b.verbose {
tdiff := (eticks - sticks)
return '${tdiff:6lld} ms | $s'
return '${tdiff:6lld} ms $s'
}
return s
}
@ -109,4 +152,3 @@ fn (b &Benchmark) tdiff_in_ms(s string, sticks i64, eticks i64) string {
fn now() i64 {
return time.ticks()
}

6
vlib/compiler/main.v
View File

@ -510,7 +510,7 @@ pub fn (v mut V) generate_main() {
cgen.genln('$cgen.fn_main;')
v.gen_main_end('return 0')
}
else if !v.pref.is_repl {
else if v.v_fmt_file=='' && !v.pref.is_repl {
verror('function `main` is not declared in the main module')
}
}
@ -525,13 +525,13 @@ pub fn (v mut V) generate_main() {
// Generate a C `main`, which calls every single test function
v.gen_main_start(false)
if v.pref.is_stats {
cgen.genln('BenchedTests bt = main__start_testing();')
cgen.genln('BenchedTests bt = main__start_testing(${test_fn_names.len},tos3("$v.dir"));')
}
for tfname in test_fn_names {
if v.pref.is_stats {
cgen.genln('BenchedTests_testing_step_start(&bt, tos3("$tfname"));')
}
cgen.genln('$tfname ();')
cgen.genln('${tfname}();')
if v.pref.is_stats {
cgen.genln('BenchedTests_testing_step_end(&bt);')
}

6
vlib/compiler/preludes/tests_assertions.v
View File

@ -17,7 +17,8 @@ fn cb_assertion_failed(filename string, line int, sourceline string, funcname st
false
}
else {
true}}
true}
}
final_filename := if use_relative_paths { filename } else { os.realpath(filename) }
final_funcname := funcname.replace('main__', '').replace('__', '.')
mut fail_message := 'FAILED assertion'
@ -32,5 +33,4 @@ fn cb_assertion_failed(filename string, line int, sourceline string, funcname st
fn cb_assertion_ok(filename string, line int, sourceline string, funcname string) {
// do nothing for now on an OK assertion
// println('OK ${line:5d}|$sourceline ')
}
}

80
vlib/compiler/preludes/tests_with_stats.v
View File

@ -1,21 +1,22 @@
module main
///////////////////////////////////////////////////////////////////////
/// This file will get compiled as a part of the same module,
/// in which a given _test.v file is, when v is given -stats argument
/// The methods defined here are called back by the test program's
/// main function, generated by compiler/main.v so that customizing the
/// look & feel of the results is easy, since it is done in normal V
/// code, instead of in embedded C ...
///////////////////////////////////////////////////////////////////////
// /////////////////////////////////////////////////////////////////////
// / This file will get compiled as a part of the same module,
// / in which a given _test.v file is, when v is given -stats argument
// / The methods defined here are called back by the test program's
// / main function, generated by compiler/main.v so that customizing the
// / look & feel of the results is easy, since it is done in normal V
// / code, instead of in embedded C ...
// /////////////////////////////////////////////////////////////////////
import (
os
term
filepath
benchmark
)
const (
INNER_INDENT = ' '
)
struct BenchedTests {
mut:
oks int
@ -25,19 +26,21 @@ mut:
bench benchmark.Benchmark
}
/////////////////////////////////////////////////////////////////////
// ///////////////////////////////////////////////////////////////////
// Called at the start of the test program produced by `v -stats file_test.v`
fn start_testing() BenchedTests {
mut b := BenchedTests{ bench: benchmark.new_benchmark() }
b.test_suit_file = os.executable() + '.v'
fn start_testing(total_number_of_tests int, vfilename string) BenchedTests {
mut b := BenchedTests{
bench: benchmark.new_benchmark()
}
b.bench.set_total_expected_steps(total_number_of_tests)
b.test_suit_file = vfilename
println('running tests in: $b.test_suit_file')
return b
}
// Called before each test_ function, defined in file_test.v
fn (b mut BenchedTests) testing_step_start(stepfunc string) {
b.step_func_name = stepfunc.replace('main__','').replace('__','.')
b.step_func_name = stepfunc.replace('main__', '').replace('__', '.')
b.oks = C.g_test_oks
b.fails = C.g_test_fails
b.bench.step()
@ -47,26 +50,26 @@ fn (b mut BenchedTests) testing_step_start(stepfunc string) {
fn (b mut BenchedTests) testing_step_end() {
ok_diff := C.g_test_oks - b.oks
fail_diff := C.g_test_fails - b.fails
//////////////////////////////////////////////////////////////////
// ////////////////////////////////////////////////////////////////
if ok_diff == 0 && fail_diff == 0 {
b.bench.neither_fail_nor_ok()
println(' ' + b.bench.step_message('NO asserts | ') + b.fn_name() )
println(INNER_INDENT + b.bench.step_message_ok('NO asserts | ') + b.fn_name())
return
}
//////////////////////////////////////////////////////////////////
// ////////////////////////////////////////////////////////////////
if ok_diff > 0 {
b.bench.ok_many(ok_diff)
}
if fail_diff > 0 {
b.bench.fail_many(fail_diff)
}
//////////////////////////////////////////////////////////////////
// ////////////////////////////////////////////////////////////////
if ok_diff > 0 && fail_diff == 0 {
println(ok_text('OK') + b.bench.step_message(nasserts(ok_diff)) + b.fn_name() )
println(INNER_INDENT + b.bench.step_message_ok(nasserts(ok_diff)) + b.fn_name())
return
}
if fail_diff > 0 {
println(fail_text('FAIL') + b.bench.step_message(nasserts(fail_diff)) + b.fn_name() )
println(INNER_INDENT + b.bench.step_message_fail(nasserts(fail_diff)) + b.fn_name())
return
}
}
@ -78,22 +81,25 @@ fn (b &BenchedTests) fn_name() string {
// Called at the end of the test program produced by `v -stats file_test.v`
fn (b mut BenchedTests) end_testing() {
b.bench.stop()
println( ' ' + b.bench.total_message('running V tests in "' + filepath.filename(b.test_suit_file) + '"' ) )
println(INNER_INDENT + b.bench.total_message('running V tests in "' + filepath.filename(b.test_suit_file) + '"'))
}
/////////////////////////////////////////////////////////////////////
// ///////////////////////////////////////////////////////////////////
fn nasserts(n int) string {
if n==0 { return '${n:2d} asserts | ' }
if n==1 { return '${n:2d} assert | ' }
if n == 0 {
return '${n:2d} asserts | '
}
if n == 1 {
return '${n:2d} assert | '
}
if n < 10 {
return '${n:2d} asserts | '
}
if n < 100 {
return '${n:3d} asserts | '
}
if n < 1000 {
return '${n:4d} asserts | '
}
return '${n:5d} asserts | '
}
fn ok_text(s string) string {
return term.ok_message('${s:5s}')
}
fn fail_text(s string) string {
return term.fail_message('${s:5s}')
}

7
vlib/compiler/tests/prod_test.v
View File

@ -1,6 +1,5 @@
// Build and run files in ./prod/ folder, comparing their output to *.expected.txt files.
// (Similar to REPL tests, but in -prod mode.)
// import os
import compiler.tests.repl.runner
import benchmark
@ -15,15 +14,15 @@ fn test_all_v_prod_files() {
bmark.step()
fres := runner.run_prod_file(options.wd, options.vexec, file) or {
bmark.fail()
eprintln( bmark.step_message(err) )
eprintln(bmark.step_message_fail(err))
assert false
continue
}
bmark.ok()
println( bmark.step_message(fres) )
println(bmark.step_message_ok(fres))
assert true
}
bmark.stop()
println( bmark.total_message('total time spent running PROD files') )
println(bmark.total_message('total time spent running PROD files'))
}
}

21
vlib/compiler/tests/repl/repl_test.v
View File

@ -8,15 +8,17 @@ fn test_the_v_compiler_can_be_invoked() {
vexec := runner.full_path_to_v(5)
println('vexecutable: $vexec')
assert vexec != ''
vcmd := '"$vexec" --version'
r := os.exec(vcmd) or { panic(err) }
//println('"$vcmd" exit_code: $r.exit_code | output: $r.output')
r := os.exec(vcmd) or {
panic(err)
}
// println('"$vcmd" exit_code: $r.exit_code | output: $r.output')
assert r.exit_code == 0
vcmd_error := '"$vexec" nonexisting.v'
r_error := os.exec(vcmd_error) or { panic(err) }
//println('"$vcmd_error" exit_code: $r_error.exit_code | output: $r_error.output')
r_error := os.exec(vcmd_error) or {
panic(err)
}
// println('"$vcmd_error" exit_code: $r_error.exit_code | output: $r_error.output')
assert r_error.exit_code == 1
assert r_error.output == '`nonexisting.v` does not exist'
}
@ -28,15 +30,14 @@ fn test_all_v_repl_files() {
bmark.step()
fres := runner.run_repl_file(options.wd, options.vexec, file) or {
bmark.fail()
eprintln( bmark.step_message(err) )
eprintln(bmark.step_message_fail(err))
assert false
continue
}
bmark.ok()
println( bmark.step_message(fres) )
println(bmark.step_message_ok(fres))
assert true
}
bmark.stop()
println( bmark.total_message('total time spent running REPL files') )
println(bmark.total_message('total time spent running REPL files'))
}

9
vlib/compiler/tests/repl/run.v
View File

@ -4,22 +4,21 @@ import compiler.tests.repl.runner
import log
import benchmark
fn main(){
fn main() {
mut logger := log.Log{}
logger.set_level(log.DEBUG)
options := runner.new_options()
mut bmark := benchmark.new_benchmark()
for file in options.files {
bmark.step()
fres := runner.run_repl_file(options.wd, options.vexec, file) or {
bmark.fail()
logger.error( bmark.step_message( err ) )
logger.error(bmark.step_message_fail(err))
continue
}
bmark.ok()
logger.info( bmark.step_message( fres ) )
logger.info(bmark.step_message_ok(fres))
}
bmark.stop()
logger.info( bmark.total_message('total time spent running REPL files') )
logger.info(bmark.total_message('total time spent running REPL files'))
}

5
vlib/os/os.v
View File

@ -23,6 +23,7 @@ struct C.dirent {
fn C.readdir(voidptr) C.dirent
pub const (
args = []string
MAX_PATH = 4096
@ -972,7 +973,7 @@ pub fn walk_ext(path, ext string) []string {
continue
}
p := path + separator + file
if os.is_dir(p) {
if os.is_dir(p) && !os.is_link(p) {
res << walk_ext(p, ext)
}
else if file.ends_with(ext) {
@ -993,7 +994,7 @@ pub fn walk(path string, fnc fn(path string)) {
}
for file in files {
p := path + os.path_separator + file
if os.is_dir(p) {
if os.is_dir(p) && !os.is_link(p) {
walk(p, fnc)
}
else if os.exists(p) {