cpython/Lib/test/test_bdb.py

1215 lines
44 KiB
Python

""" Test the bdb module.
A test defines a list of tuples that may be seen as paired tuples, each
pair being defined by 'expect_tuple, set_tuple' as follows:
([event, [lineno[, co_name[, eargs]]]]), (set_type, [sargs])
* 'expect_tuple' describes the expected current state of the Bdb instance.
It may be the empty tuple and no check is done in that case.
* 'set_tuple' defines the set_*() method to be invoked when the Bdb
instance reaches this state.
Example of an 'expect_tuple, set_tuple' pair:
('line', 2, 'tfunc_main'), ('step', )
Definitions of the members of the 'expect_tuple':
event:
Name of the trace event. The set methods that do not give back
control to the tracer [1] do not trigger a tracer event and in
that case the next 'event' may be 'None' by convention, its value
is not checked.
[1] Methods that trigger a trace event are set_step(), set_next(),
set_return(), set_until() and set_continue().
lineno:
Line number. Line numbers are relative to the start of the
function when tracing a function in the test_bdb module (i.e. this
module).
co_name:
Name of the function being currently traced.
eargs:
A tuple:
* On an 'exception' event the tuple holds a class object, the
current exception must be an instance of this class.
* On a 'line' event, the tuple holds a dictionary and a list. The
dictionary maps each breakpoint number that has been hit on this
line to its hits count. The list holds the list of breakpoint
number temporaries that are being deleted.
Definitions of the members of the 'set_tuple':
set_type:
The type of the set method to be invoked. This may
be the type of one of the Bdb set methods: 'step', 'next',
'until', 'return', 'continue', 'break', 'quit', or the type of one
of the set methods added by test_bdb.Bdb: 'ignore', 'enable',
'disable', 'clear', 'up', 'down'.
sargs:
The arguments of the set method if any, packed in a tuple.
"""
import bdb as _bdb
import sys
import os
import unittest
import textwrap
import importlib
import linecache
from contextlib import contextmanager
from itertools import islice, repeat
from test.support import import_helper
from test.support import os_helper
from test.support import patch_list
class BdbException(Exception): pass
class BdbError(BdbException): """Error raised by the Bdb instance."""
class BdbSyntaxError(BdbException): """Syntax error in the test case."""
class BdbNotExpectedError(BdbException): """Unexpected result."""
# When 'dry_run' is set to true, expect tuples are ignored and the actual
# state of the tracer is printed after running each set_*() method of the test
# case. The full list of breakpoints and their attributes is also printed
# after each 'line' event where a breakpoint has been hit.
dry_run = 0
def reset_Breakpoint():
_bdb.Breakpoint.clearBreakpoints()
def info_breakpoints():
bp_list = [bp for bp in _bdb.Breakpoint.bpbynumber if bp]
if not bp_list:
return ''
header_added = False
for bp in bp_list:
if not header_added:
info = 'BpNum Temp Enb Hits Ignore Where\n'
header_added = True
disp = 'yes ' if bp.temporary else 'no '
enab = 'yes' if bp.enabled else 'no '
info += ('%-5d %s %s %-4d %-6d at %s:%d' %
(bp.number, disp, enab, bp.hits, bp.ignore,
os.path.basename(bp.file), bp.line))
if bp.cond:
info += '\n\tstop only if %s' % (bp.cond,)
info += '\n'
return info
class Bdb(_bdb.Bdb):
"""Extend Bdb to enhance test coverage."""
def trace_dispatch(self, frame, event, arg):
self.currentbp = None
return super().trace_dispatch(frame, event, arg)
def set_break(self, filename, lineno, temporary=False, cond=None,
funcname=None):
if isinstance(funcname, str):
if filename == __file__:
globals_ = globals()
else:
module = importlib.import_module(filename[:-3])
globals_ = module.__dict__
func = eval(funcname, globals_)
code = func.__code__
filename = code.co_filename
lineno = code.co_firstlineno
funcname = code.co_name
res = super().set_break(filename, lineno, temporary=temporary,
cond=cond, funcname=funcname)
if isinstance(res, str):
raise BdbError(res)
return res
def get_stack(self, f, t):
self.stack, self.index = super().get_stack(f, t)
self.frame = self.stack[self.index][0]
return self.stack, self.index
def set_ignore(self, bpnum):
"""Increment the ignore count of Breakpoint number 'bpnum'."""
bp = self.get_bpbynumber(bpnum)
bp.ignore += 1
def set_enable(self, bpnum):
bp = self.get_bpbynumber(bpnum)
bp.enabled = True
def set_disable(self, bpnum):
bp = self.get_bpbynumber(bpnum)
bp.enabled = False
def set_clear(self, fname, lineno):
err = self.clear_break(fname, lineno)
if err:
raise BdbError(err)
def set_up(self):
"""Move up in the frame stack."""
if not self.index:
raise BdbError('Oldest frame')
self.index -= 1
self.frame = self.stack[self.index][0]
def set_down(self):
"""Move down in the frame stack."""
if self.index + 1 == len(self.stack):
raise BdbError('Newest frame')
self.index += 1
self.frame = self.stack[self.index][0]
class Tracer(Bdb):
"""A tracer for testing the bdb module."""
def __init__(self, expect_set, skip=None, dry_run=False, test_case=None):
super().__init__(skip=skip)
self.expect_set = expect_set
self.dry_run = dry_run
self.header = ('Dry-run results for %s:' % test_case if
test_case is not None else None)
self.init_test()
def init_test(self):
self.cur_except = None
self.expect_set_no = 0
self.breakpoint_hits = None
self.expected_list = list(islice(self.expect_set, 0, None, 2))
self.set_list = list(islice(self.expect_set, 1, None, 2))
def trace_dispatch(self, frame, event, arg):
# On an 'exception' event, call_exc_trace() in Python/ceval.c discards
# a BdbException raised by the Tracer instance, so we raise it on the
# next trace_dispatch() call that occurs unless the set_quit() or
# set_continue() method has been invoked on the 'exception' event.
if self.cur_except is not None:
raise self.cur_except
if event == 'exception':
try:
res = super().trace_dispatch(frame, event, arg)
return res
except BdbException as e:
self.cur_except = e
return self.trace_dispatch
else:
return super().trace_dispatch(frame, event, arg)
def user_call(self, frame, argument_list):
# Adopt the same behavior as pdb and, as a side effect, skip also the
# first 'call' event when the Tracer is started with Tracer.runcall()
# which may be possibly considered as a bug.
if not self.stop_here(frame):
return
self.process_event('call', frame, argument_list)
self.next_set_method()
def user_line(self, frame):
self.process_event('line', frame)
if self.dry_run and self.breakpoint_hits:
info = info_breakpoints().strip('\n')
# Indent each line.
for line in info.split('\n'):
print(' ' + line)
self.delete_temporaries()
self.breakpoint_hits = None
self.next_set_method()
def user_return(self, frame, return_value):
self.process_event('return', frame, return_value)
self.next_set_method()
def user_exception(self, frame, exc_info):
self.exc_info = exc_info
self.process_event('exception', frame)
self.next_set_method()
def do_clear(self, arg):
# The temporary breakpoints are deleted in user_line().
bp_list = [self.currentbp]
self.breakpoint_hits = (bp_list, bp_list)
def delete_temporaries(self):
if self.breakpoint_hits:
for n in self.breakpoint_hits[1]:
self.clear_bpbynumber(n)
def pop_next(self):
self.expect_set_no += 1
try:
self.expect = self.expected_list.pop(0)
except IndexError:
raise BdbNotExpectedError(
'expect_set list exhausted, cannot pop item %d' %
self.expect_set_no)
self.set_tuple = self.set_list.pop(0)
def process_event(self, event, frame, *args):
# Call get_stack() to enable walking the stack with set_up() and
# set_down().
tb = None
if event == 'exception':
tb = self.exc_info[2]
self.get_stack(frame, tb)
# A breakpoint has been hit and it is not a temporary.
if self.currentbp is not None and not self.breakpoint_hits:
bp_list = [self.currentbp]
self.breakpoint_hits = (bp_list, [])
# Pop next event.
self.event= event
self.pop_next()
if self.dry_run:
self.print_state(self.header)
return
# Validate the expected results.
if self.expect:
self.check_equal(self.expect[0], event, 'Wrong event type')
self.check_lno_name()
if event in ('call', 'return'):
self.check_expect_max_size(3)
elif len(self.expect) > 3:
if event == 'line':
bps, temporaries = self.expect[3]
bpnums = sorted(bps.keys())
if not self.breakpoint_hits:
self.raise_not_expected(
'No breakpoints hit at expect_set item %d' %
self.expect_set_no)
self.check_equal(bpnums, self.breakpoint_hits[0],
'Breakpoint numbers do not match')
self.check_equal([bps[n] for n in bpnums],
[self.get_bpbynumber(n).hits for
n in self.breakpoint_hits[0]],
'Wrong breakpoint hit count')
self.check_equal(sorted(temporaries), self.breakpoint_hits[1],
'Wrong temporary breakpoints')
elif event == 'exception':
if not isinstance(self.exc_info[1], self.expect[3]):
self.raise_not_expected(
"Wrong exception at expect_set item %d, got '%s'" %
(self.expect_set_no, self.exc_info))
def check_equal(self, expected, result, msg):
if expected == result:
return
self.raise_not_expected("%s at expect_set item %d, got '%s'" %
(msg, self.expect_set_no, result))
def check_lno_name(self):
"""Check the line number and function co_name."""
s = len(self.expect)
if s > 1:
lineno = self.lno_abs2rel()
self.check_equal(self.expect[1], lineno, 'Wrong line number')
if s > 2:
self.check_equal(self.expect[2], self.frame.f_code.co_name,
'Wrong function name')
def check_expect_max_size(self, size):
if len(self.expect) > size:
raise BdbSyntaxError('Invalid size of the %s expect tuple: %s' %
(self.event, self.expect))
def lno_abs2rel(self):
fname = self.canonic(self.frame.f_code.co_filename)
lineno = self.frame.f_lineno
return ((lineno - self.frame.f_code.co_firstlineno + 1)
if fname == self.canonic(__file__) else lineno)
def lno_rel2abs(self, fname, lineno):
return (self.frame.f_code.co_firstlineno + lineno - 1
if (lineno and self.canonic(fname) == self.canonic(__file__))
else lineno)
def get_state(self):
lineno = self.lno_abs2rel()
co_name = self.frame.f_code.co_name
state = "('%s', %d, '%s'" % (self.event, lineno, co_name)
if self.breakpoint_hits:
bps = '{'
for n in self.breakpoint_hits[0]:
if bps != '{':
bps += ', '
bps += '%s: %s' % (n, self.get_bpbynumber(n).hits)
bps += '}'
bps = '(' + bps + ', ' + str(self.breakpoint_hits[1]) + ')'
state += ', ' + bps
elif self.event == 'exception':
state += ', ' + self.exc_info[0].__name__
state += '), '
return state.ljust(32) + str(self.set_tuple) + ','
def print_state(self, header=None):
if header is not None and self.expect_set_no == 1:
print()
print(header)
print('%d: %s' % (self.expect_set_no, self.get_state()))
def raise_not_expected(self, msg):
msg += '\n'
msg += ' Expected: %s\n' % str(self.expect)
msg += ' Got: ' + self.get_state()
raise BdbNotExpectedError(msg)
def next_set_method(self):
set_type = self.set_tuple[0]
args = self.set_tuple[1] if len(self.set_tuple) == 2 else None
set_method = getattr(self, 'set_' + set_type)
# The following set methods give back control to the tracer.
if set_type in ('step', 'continue', 'quit'):
set_method()
return
elif set_type in ('next', 'return'):
set_method(self.frame)
return
elif set_type == 'until':
lineno = None
if args:
lineno = self.lno_rel2abs(self.frame.f_code.co_filename,
args[0])
set_method(self.frame, lineno)
return
# The following set methods do not give back control to the tracer and
# next_set_method() is called recursively.
if (args and set_type in ('break', 'clear', 'ignore', 'enable',
'disable')) or set_type in ('up', 'down'):
if set_type in ('break', 'clear'):
fname, lineno, *remain = args
lineno = self.lno_rel2abs(fname, lineno)
args = [fname, lineno]
args.extend(remain)
set_method(*args)
elif set_type in ('ignore', 'enable', 'disable'):
set_method(*args)
elif set_type in ('up', 'down'):
set_method()
# Process the next expect_set item.
# It is not expected that a test may reach the recursion limit.
self.event= None
self.pop_next()
if self.dry_run:
self.print_state()
else:
if self.expect:
self.check_lno_name()
self.check_expect_max_size(3)
self.next_set_method()
else:
raise BdbSyntaxError('"%s" is an invalid set_tuple' %
self.set_tuple)
class TracerRun():
"""Provide a context for running a Tracer instance with a test case."""
def __init__(self, test_case, skip=None):
self.test_case = test_case
self.dry_run = test_case.dry_run
self.tracer = Tracer(test_case.expect_set, skip=skip,
dry_run=self.dry_run, test_case=test_case.id())
self._original_tracer = None
def __enter__(self):
# test_pdb does not reset Breakpoint class attributes on exit :-(
reset_Breakpoint()
self._original_tracer = sys.gettrace()
return self.tracer
def __exit__(self, type_=None, value=None, traceback=None):
reset_Breakpoint()
sys.settrace(self._original_tracer)
not_empty = ''
if self.tracer.set_list:
not_empty += 'All paired tuples have not been processed, '
not_empty += ('the last one was number %d\n' %
self.tracer.expect_set_no)
not_empty += repr(self.tracer.set_list)
# Make a BdbNotExpectedError a unittest failure.
if type_ is not None and issubclass(BdbNotExpectedError, type_):
if isinstance(value, BaseException) and value.args:
err_msg = value.args[0]
if not_empty:
err_msg += '\n' + not_empty
if self.dry_run:
print(err_msg)
return True
else:
self.test_case.fail(err_msg)
else:
assert False, 'BdbNotExpectedError with empty args'
if not_empty:
if self.dry_run:
print(not_empty)
else:
self.test_case.fail(not_empty)
def run_test(modules, set_list, skip=None):
"""Run a test and print the dry-run results.
'modules': A dictionary mapping module names to their source code as a
string. The dictionary MUST include one module named
'test_module' with a main() function.
'set_list': A list of set_type tuples to be run on the module.
For example, running the following script outputs the following results:
***************************** SCRIPT ********************************
from test.test_bdb import run_test, break_in_func
code = '''
def func():
lno = 3
def main():
func()
lno = 7
'''
set_list = [
break_in_func('func', 'test_module.py'),
('continue', ),
('step', ),
('step', ),
('step', ),
('quit', ),
]
modules = { 'test_module': code }
run_test(modules, set_list)
**************************** results ********************************
1: ('line', 2, 'tfunc_import'), ('next',),
2: ('line', 3, 'tfunc_import'), ('step',),
3: ('call', 5, 'main'), ('break', ('test_module.py', None, False, None, 'func')),
4: ('None', 5, 'main'), ('continue',),
5: ('line', 3, 'func', ({1: 1}, [])), ('step',),
BpNum Temp Enb Hits Ignore Where
1 no yes 1 0 at test_module.py:2
6: ('return', 3, 'func'), ('step',),
7: ('line', 7, 'main'), ('step',),
8: ('return', 7, 'main'), ('quit',),
*************************************************************************
"""
def gen(a, b):
try:
while 1:
x = next(a)
y = next(b)
yield x
yield y
except StopIteration:
return
# Step over the import statement in tfunc_import using 'next' and step
# into main() in test_module.
sl = [('next', ), ('step', )]
sl.extend(set_list)
test = BaseTestCase()
test.dry_run = True
test.id = lambda : None
test.expect_set = list(gen(repeat(()), iter(sl)))
with create_modules(modules):
with TracerRun(test, skip=skip) as tracer:
tracer.runcall(tfunc_import)
@contextmanager
def create_modules(modules):
with os_helper.temp_cwd():
sys.path.append(os.getcwd())
try:
for m in modules:
fname = m + '.py'
with open(fname, 'w', encoding="utf-8") as f:
f.write(textwrap.dedent(modules[m]))
linecache.checkcache(fname)
importlib.invalidate_caches()
yield
finally:
for m in modules:
import_helper.forget(m)
sys.path.pop()
def break_in_func(funcname, fname=__file__, temporary=False, cond=None):
return 'break', (fname, None, temporary, cond, funcname)
TEST_MODULE = 'test_module_for_bdb'
TEST_MODULE_FNAME = TEST_MODULE + '.py'
def tfunc_import():
import test_module_for_bdb
test_module_for_bdb.main()
def tfunc_main():
lno = 2
tfunc_first()
tfunc_second()
lno = 5
lno = 6
lno = 7
def tfunc_first():
lno = 2
lno = 3
lno = 4
def tfunc_second():
lno = 2
class BaseTestCase(unittest.TestCase):
"""Base class for all tests."""
dry_run = dry_run
def fail(self, msg=None):
# Override fail() to use 'raise from None' to avoid repetition of the
# error message and traceback.
raise self.failureException(msg) from None
class StateTestCase(BaseTestCase):
"""Test the step, next, return, until and quit 'set_' methods."""
def test_step(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('step', ),
('line', 2, 'tfunc_first'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_step_next_on_last_statement(self):
for set_type in ('step', 'next'):
with self.subTest(set_type=set_type):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('break', (__file__, 3)),
('None', 1, 'tfunc_first'), ('continue', ),
('line', 3, 'tfunc_first', ({1:1}, [])), (set_type, ),
('line', 4, 'tfunc_first'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_next(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('next', ),
('line', 4, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_second'), ('step', ),
('line', 2, 'tfunc_second'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_next_over_import(self):
code = """
def main():
lno = 3
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'), ('next', ),
('line', 3, 'tfunc_import'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_next_on_plain_statement(self):
# Check that set_next() is equivalent to set_step() on a plain
# statement.
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('next', ),
('line', 2, 'tfunc_first'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_next_in_caller_frame(self):
# Check that set_next() in the caller frame causes the tracer
# to stop next in the caller frame.
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('up', ),
('None', 3, 'tfunc_main'), ('next', ),
('line', 4, 'tfunc_main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_return(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('step', ),
('line', 2, 'tfunc_first'), ('return', ),
('return', 4, 'tfunc_first'), ('step', ),
('line', 4, 'tfunc_main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_return_in_caller_frame(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('up', ),
('None', 3, 'tfunc_main'), ('return', ),
('return', 7, 'tfunc_main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_until(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('step', ),
('line', 2, 'tfunc_first'), ('until', (4, )),
('line', 4, 'tfunc_first'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_until_with_too_large_count(self):
self.expect_set = [
('line', 2, 'tfunc_main'), break_in_func('tfunc_first'),
('None', 2, 'tfunc_main'), ('continue', ),
('line', 2, 'tfunc_first', ({1:1}, [])), ('until', (9999, )),
('return', 4, 'tfunc_first'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
def test_until_in_caller_frame(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('up', ),
('None', 3, 'tfunc_main'), ('until', (6, )),
('line', 6, 'tfunc_main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
@patch_list(sys.meta_path)
def test_skip(self):
# Check that tracing is skipped over the import statement in
# 'tfunc_import()'.
# Remove all but the standard importers.
sys.meta_path[:] = (
item
for item in sys.meta_path
if item.__module__.startswith('_frozen_importlib')
)
code = """
def main():
lno = 3
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'), ('step', ),
('line', 3, 'tfunc_import'), ('quit', ),
]
skip = ('importlib*', 'zipimport', 'encodings.*', TEST_MODULE)
with TracerRun(self, skip=skip) as tracer:
tracer.runcall(tfunc_import)
def test_skip_with_no_name_module(self):
# some frames have `globals` with no `__name__`
# for instance the second frame in this traceback
# exec(compile('raise ValueError()', '', 'exec'), {})
bdb = Bdb(skip=['anything*'])
self.assertIs(bdb.is_skipped_module(None), False)
def test_down(self):
# Check that set_down() raises BdbError at the newest frame.
self.expect_set = [
('line', 2, 'tfunc_main'), ('down', ),
]
with TracerRun(self) as tracer:
self.assertRaises(BdbError, tracer.runcall, tfunc_main)
def test_up(self):
self.expect_set = [
('line', 2, 'tfunc_main'), ('step', ),
('line', 3, 'tfunc_main'), ('step', ),
('call', 1, 'tfunc_first'), ('up', ),
('None', 3, 'tfunc_main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_main)
class BreakpointTestCase(BaseTestCase):
"""Test the breakpoint set method."""
def test_bp_on_non_existent_module(self):
self.expect_set = [
('line', 2, 'tfunc_import'), ('break', ('/non/existent/module.py', 1))
]
with TracerRun(self) as tracer:
self.assertRaises(BdbError, tracer.runcall, tfunc_import)
def test_bp_after_last_statement(self):
code = """
def main():
lno = 3
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'), ('break', (TEST_MODULE_FNAME, 4))
]
with TracerRun(self) as tracer:
self.assertRaises(BdbError, tracer.runcall, tfunc_import)
def test_temporary_bp(self):
code = """
def func():
lno = 3
def main():
for i in range(2):
func()
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME, True),
('None', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME, True),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:1}, [1])), ('continue', ),
('line', 3, 'func', ({2:1}, [2])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_disabled_temporary_bp(self):
code = """
def func():
lno = 3
def main():
for i in range(3):
func()
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME, True),
('None', 2, 'tfunc_import'), ('disable', (2, )),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:1}, [])), ('enable', (2, )),
('None', 3, 'func'), ('disable', (1, )),
('None', 3, 'func'), ('continue', ),
('line', 3, 'func', ({2:1}, [2])), ('enable', (1, )),
('None', 3, 'func'), ('continue', ),
('line', 3, 'func', ({1:2}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_bp_condition(self):
code = """
def func(a):
lno = 3
def main():
for i in range(3):
func(i)
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME, False, 'a == 2'),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:3}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_bp_exception_on_condition_evaluation(self):
code = """
def func(a):
lno = 3
def main():
func(0)
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME, False, '1 / 0'),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:1}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_bp_ignore_count(self):
code = """
def func():
lno = 3
def main():
for i in range(2):
func()
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('ignore', (1, )),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:2}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_ignore_count_on_disabled_bp(self):
code = """
def func():
lno = 3
def main():
for i in range(3):
func()
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'),
break_in_func('func', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('ignore', (1, )),
('None', 2, 'tfunc_import'), ('disable', (1, )),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({2:1}, [])), ('enable', (1, )),
('None', 3, 'func'), ('continue', ),
('line', 3, 'func', ({2:2}, [])), ('continue', ),
('line', 3, 'func', ({1:2}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_clear_two_bp_on_same_line(self):
code = """
def func():
lno = 3
lno = 4
def main():
for i in range(3):
func()
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'), ('break', (TEST_MODULE_FNAME, 3)),
('None', 2, 'tfunc_import'), ('break', (TEST_MODULE_FNAME, 3)),
('None', 2, 'tfunc_import'), ('break', (TEST_MODULE_FNAME, 4)),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:1}, [])), ('continue', ),
('line', 4, 'func', ({3:1}, [])), ('clear', (TEST_MODULE_FNAME, 3)),
('None', 4, 'func'), ('continue', ),
('line', 4, 'func', ({3:2}, [])), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_clear_at_no_bp(self):
self.expect_set = [
('line', 2, 'tfunc_import'), ('clear', (__file__, 1))
]
with TracerRun(self) as tracer:
self.assertRaises(BdbError, tracer.runcall, tfunc_import)
def test_load_bps_from_previous_Bdb_instance(self):
reset_Breakpoint()
db1 = Bdb()
fname = db1.canonic(__file__)
db1.set_break(__file__, 1)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
db2 = Bdb()
db2.set_break(__file__, 2)
db2.set_break(__file__, 3)
db2.set_break(__file__, 4)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [1, 2, 3, 4]})
db2.clear_break(__file__, 1)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [2, 3, 4]})
db3 = Bdb()
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [2, 3, 4]})
self.assertEqual(db3.get_all_breaks(), {fname: [2, 3, 4]})
db2.clear_break(__file__, 2)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [3, 4]})
self.assertEqual(db3.get_all_breaks(), {fname: [2, 3, 4]})
db4 = Bdb()
db4.set_break(__file__, 5)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [3, 4]})
self.assertEqual(db3.get_all_breaks(), {fname: [2, 3, 4]})
self.assertEqual(db4.get_all_breaks(), {fname: [3, 4, 5]})
reset_Breakpoint()
db5 = Bdb()
db5.set_break(__file__, 6)
self.assertEqual(db1.get_all_breaks(), {fname: [1]})
self.assertEqual(db2.get_all_breaks(), {fname: [3, 4]})
self.assertEqual(db3.get_all_breaks(), {fname: [2, 3, 4]})
self.assertEqual(db4.get_all_breaks(), {fname: [3, 4, 5]})
self.assertEqual(db5.get_all_breaks(), {fname: [6]})
class RunTestCase(BaseTestCase):
"""Test run, runeval and set_trace."""
def test_run_step(self):
# Check that the bdb 'run' method stops at the first line event.
code = """
lno = 2
"""
self.expect_set = [
('line', 2, '<module>'), ('step', ),
('return', 2, '<module>'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.run(compile(textwrap.dedent(code), '<string>', 'exec'))
def test_runeval_step(self):
# Test bdb 'runeval'.
code = """
def main():
lno = 3
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 1, '<module>'), ('step', ),
('call', 2, 'main'), ('step', ),
('line', 3, 'main'), ('step', ),
('return', 3, 'main'), ('step', ),
('return', 1, '<module>'), ('quit', ),
]
import test_module_for_bdb
with TracerRun(self) as tracer:
tracer.runeval('test_module_for_bdb.main()', globals(), locals())
class IssuesTestCase(BaseTestCase):
"""Test fixed bdb issues."""
def test_step_at_return_with_no_trace_in_caller(self):
# Issue #13183.
# Check that the tracer does step into the caller frame when the
# trace function is not set in that frame.
code_1 = """
from test_module_for_bdb_2 import func
def main():
func()
lno = 5
"""
code_2 = """
def func():
lno = 3
"""
modules = {
TEST_MODULE: code_1,
'test_module_for_bdb_2': code_2,
}
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('func', 'test_module_for_bdb_2.py'),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'func', ({1:1}, [])), ('step', ),
('return', 3, 'func'), ('step', ),
('line', 5, 'main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_next_until_return_in_generator(self):
# Issue #16596.
# Check that set_next(), set_until() and set_return() do not treat the
# `yield` and `yield from` statements as if they were returns and stop
# instead in the current frame.
code = """
def test_gen():
yield 0
lno = 4
return 123
def main():
it = test_gen()
next(it)
next(it)
lno = 11
"""
modules = { TEST_MODULE: code }
for set_type in ('next', 'until', 'return'):
with self.subTest(set_type=set_type):
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('test_gen', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'test_gen', ({1:1}, [])), (set_type, ),
]
if set_type == 'return':
self.expect_set.extend(
[('exception', 10, 'main', StopIteration), ('step',),
('return', 10, 'main'), ('quit', ),
]
)
else:
self.expect_set.extend(
[('line', 4, 'test_gen'), ('quit', ),]
)
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_next_command_in_generator_for_loop(self):
# Issue #16596.
code = """
def test_gen():
yield 0
lno = 4
yield 1
return 123
def main():
for i in test_gen():
lno = 10
lno = 11
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('test_gen', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'test_gen', ({1:1}, [])), ('next', ),
('line', 4, 'test_gen'), ('next', ),
('line', 5, 'test_gen'), ('next', ),
('line', 6, 'test_gen'), ('next', ),
('exception', 9, 'main', StopIteration), ('step', ),
('line', 11, 'main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_next_command_in_generator_with_subiterator(self):
# Issue #16596.
code = """
def test_subgen():
yield 0
return 123
def test_gen():
x = yield from test_subgen()
return 456
def main():
for i in test_gen():
lno = 12
lno = 13
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('test_gen', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 7, 'test_gen', ({1:1}, [])), ('next', ),
('line', 8, 'test_gen'), ('next', ),
('exception', 11, 'main', StopIteration), ('step', ),
('line', 13, 'main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
def test_return_command_in_generator_with_subiterator(self):
# Issue #16596.
code = """
def test_subgen():
yield 0
return 123
def test_gen():
x = yield from test_subgen()
return 456
def main():
for i in test_gen():
lno = 12
lno = 13
"""
modules = { TEST_MODULE: code }
with create_modules(modules):
self.expect_set = [
('line', 2, 'tfunc_import'),
break_in_func('test_subgen', TEST_MODULE_FNAME),
('None', 2, 'tfunc_import'), ('continue', ),
('line', 3, 'test_subgen', ({1:1}, [])), ('return', ),
('exception', 7, 'test_gen', StopIteration), ('return', ),
('exception', 11, 'main', StopIteration), ('step', ),
('line', 13, 'main'), ('quit', ),
]
with TracerRun(self) as tracer:
tracer.runcall(tfunc_import)
class TestRegressions(unittest.TestCase):
def test_format_stack_entry_no_lineno(self):
# See gh-101517
Bdb().format_stack_entry((sys._getframe(), None))
if __name__ == "__main__":
unittest.main()