def raw_seg_list():
return [
RawSegment('bar', FilePositionMarker.from_fresh()),
RawSegment('foo',
FilePositionMarker.from_fresh().advance_by('bar')),
RawSegment('bar',
FilePositionMarker.from_fresh().advance_by('barfoo'))
]
def test__parser_2__base_segments_base_compare():
fp1 = FilePositionMarker.from_fresh()
fp2 = FilePositionMarker.from_fresh()
rs1 = RawSegment('foobar', fp1)
rs2 = RawSegment('foobar', fp2)
ds1 = DummySegment([rs1])
ds2 = DummySegment([rs2])
dsa2 = DummyAuxSegment([rs2])
# Check for equality
assert ds1 == ds2
# Check a different match on the same details are not the same
assert ds1 != dsa2
def test__parser_2__common_marker():
# test making one from fresh
fp1 = FilePositionMarker.from_fresh()
fp2 = fp1.advance_by('abc')
fp3 = fp2.advance_by('def\nghi\njlk')
fp4 = fp3.advance_by('mno', idx=1)
# check comparisons
assert fp1 == FilePositionMarker(1, 1, 1, 0)
assert fp4 > fp3 > fp2 > fp1
assert fp1 < fp2 < fp3 < fp4
# Check advance works without newline
assert fp2 == FilePositionMarker(1, 1, 4, 3)
assert fp3 == FilePositionMarker(1, 3, 4, 14)
assert fp4 == FilePositionMarker(2, 3, 7, 17)
def test__parser_2__lexer_multimatcher(caplog):
matcher = RepeatedMultiMatcher(
SingletonMatcher("dot", ".",
RawSegment.make('.', name='dot', is_code=True)),
RegexMatcher("test", r"#[^#]*#", RawSegment.make('test', name='test')))
start_pos = FilePositionMarker.from_fresh()
with caplog.at_level(logging.DEBUG):
res = matcher.match('..#..#..#', start_pos)
assert res.new_string == '#' # Should match right up to the final element
assert res.new_pos == start_pos.advance_by('..#..#..')
assert len(res.segments) == 5
assert res.segments[2].raw == '#..#'
def assert_matches(instring, matcher, matchstring):
start_pos = FilePositionMarker.from_fresh()
res = matcher.match(instring, start_pos)
# Check we've got the right type
assert isinstance(res, LexMatch)
if matchstring is None:
assert res.new_string == instring
assert res.new_pos == start_pos
assert res.segments == tuple()
else:
new_pos = start_pos.advance_by(matchstring)
assert res.new_string == instring[len(matchstring):]
assert res.new_pos == new_pos
assert len(res.segments) == 1
assert res.segments[0].raw == matchstring
def generate_test_segments(elems):
""" This function isn't totally robust, but good enough
for testing. Use with caution. """
buff = []
raw_buff = ''
for elem in elems:
if set(elem) <= set([' ', '\t']):
cls = RawSegment.make(' ', name='whitespace')
elif set(elem) <= set(['\n']):
cls = RawSegment.make('\n', name='newline')
elif elem.startswith('--'):
cls = RawSegment.make('--', name='inline_comment')
elif elem.startswith('"'):
cls = RawSegment.make('"', name='double_quote', _is_code=True)
elif elem.startswith("'"):
cls = RawSegment.make("'", name='single_quote', _is_code=True)
else:
cls = RawSegment.make('', _is_code=True)
buff.append(
cls(elem,
FilePositionMarker.from_fresh().advance_by(raw_buff)))
raw_buff += elem
return tuple(buff) # Make sure we return a tuple
def raw_seg():
fp = FilePositionMarker.from_fresh().advance_by('abc')
return RawSegment('foobar', fp)
def test__parser_2__base_segments_raw_compare():
fp1 = FilePositionMarker.from_fresh()
fp2 = FilePositionMarker.from_fresh()
rs1 = RawSegment('foobar', fp1)
rs2 = RawSegment('foobar', fp2)
assert rs1 == rs2
def test__parser_2__base_segments_raw_init():
""" Test initialisation. Other tests just use the fixture """
fp = FilePositionMarker.from_fresh()
RawSegment('foobar', fp)
def test__parser_2__common_marker_format():
fp1 = FilePositionMarker(1, 2, 3, 0)
# Check Formatting Style
assert str(fp1) == "[0](1, 2, 3)"