def process_entry(self, entry: str):
"""
Process a whole entry
:param entry:
:return:
"""
doc = self.nlp(entry)
analysis = Analysis()
for sent in doc.sents:
if self.do_general_analysis:
self.count_tokens(sent)
for noun_phrase in sent.noun_chunks:
covered = RangeSet()
for trigger in self.all_triggers:
if trigger in noun_phrase.text.lower():
start = noun_phrase.text.lower().index(trigger)
if not Range(start,
len(noun_phrase.text) +
start).intersection(covered):
covered.add(
Range(noun_phrase.start, noun_phrase.end))
self.__process_np(trigger, noun_phrase, doc,
analysis)
return analysis
def expand_star(self, max_line: int, other_split_files):
source_file_range = Range(1, max_line, include_end=True)
union_of_splits = RangeSet()
for split in other_split_files:
lines = split._lines # pylint: disable=protected-access
union_of_splits = union_of_splits.union(lines)
diff = source_file_range.symmetric_difference(union_of_splits)
self._lines.extend(diff)
def test_rangeset_docstring():
a = RangeSet()
b = RangeSet([Range(0, 1), Range(2, 3), Range(4, 5)])
c = RangeSet(Range(0, 1), Range(2, 3), Range(4, 5))
d = RangeSet("[0, 1)", ["[1.5, 2)", "[2.5, 3)"], "[4, 5]")
assert (str(a) == "{}")
assert (str(b) == "{[0, 1), [2, 3), [4, 5)}")
assert (str(c) == "{[0, 1), [2, 3), [4, 5)}")
assert (b == c)
assert (str(d) == "{[0, 1), [1.5, 2), [2.5, 3), [4, 5]}")
asserterror(
ValueError, RangeSet,
([[Range(0, 1), Range(2, 3)], [Range(4, 5), Range(6, 7)]], ))
f = RangeSet("[0, 3]", "[2, 4)", "[5, 6]")
assert (str(f) == "{[0, 4), [5, 6]}")
def _create_line_ranges(self, split_data: collections.abc.Mapping,
max_line: int):
lines = split_data.get("lines")
if not lines or not lines.strip():
raise ConfigError(
f'No lines specified for split file "{self._path}".')
range_set = RangeSet()
line_ranges = lines.split(",")
for line_range in line_ranges:
start, _, end = line_range.partition("-")
if start.strip() == "*":
self._has_star = True
continue
try:
start = int(start)
end = int(end) if end else start
if not 0 < start <= max_line or not 0 < end <= max_line:
raise ValueError(f"Out of range (1-{max_line})")
range_set.add(Range(start, end, include_end=True))
except ValueError as ex:
raise ConfigError(
f'Invalid lines for split file "{self._path}": {ex}')
return range_set
def test_rangeset_constructor_valid(args, ranges, strr, reprr, isempty):
"""
Tests that the constructor of rngset works as intended. Also, as a byproduct,
tests the .ranges(), .__str__(), .__repr__(), .clear(), and .isempty()
"""
rangeset = RangeSet(*args)
assert (ranges == rangeset.ranges())
assert (strr == str(rangeset))
assert (reprr == repr(rangeset))
assert (isempty == rangeset.isempty())
assert (isempty != bool(rangeset))
assert (hash(rangeset) == hash(rangeset.copy()))
rangeset.clear()
assert ("{}" == str(rangeset))
assert ("RangeSet{}" == repr(rangeset))
assert (rangeset.isempty())
assert (not bool(rangeset))
def test_issue8():
# issue: adding a Range to a RangeSet containing two non-overlapping ranges, such that the new range overlaps
# with one but not the other, leads to a TypeError being raised.
# cause: code was passing a Linked List Node instead of the node's value (a range)
try:
a = RangeSet()
a.add(Range(100, 300))
a.add(Range(400, 500))
a.add(Range(500, 600))
assert (str(a) == "{[100, 300), [400, 600)}")
b = RangeSet()
b.add(Range(400, 600))
b.add(Range(200, 300))
b.add(Range(100, 200))
assert (str(b) == "{[100, 300), [400, 600)}")
except TypeError:
fail(
"RangeSet should not have an issue concatenating to the second range of two in a RangeSet"
)
2018, 12, 1), False, "[2017-05-27, 2018-02-02]",
"Range[datetime.date(2017, 5, 27), datetime.date(2018, 2, 2)]"),
(Range(datetime.timedelta(0, 3600), datetime.timedelta(
0, 7200)), datetime.timedelta(0, 6000), True, "[1:00:00, 2:00:00)",
"Range[datetime.timedelta(seconds=3600), datetime.timedelta(seconds=7200))"
),
(Range(datetime.timedelta(1, 1804), datetime.timedelta(3)),
datetime.timedelta(1), False, "[1 day, 0:30:04, 3 days, 0:00:00)",
"Range[datetime.timedelta(days=1, seconds=1804), datetime.timedelta(days=3))"
),
(Range("begin", "end"), "middle", False, "[begin, end)",
"Range['begin', 'end')"),
(Range("begin",
"end"), "cows", True, "[begin, end)", "Range['begin', 'end')"),
# RangeSets
(Range(1, 10), RangeSet(Range(2, 9)), True, "[1, 10)", "Range[1, 10)"),
(Range(1,
10), RangeSet(Range(2, 3), Range(4, 5), Range(6, 7), Range(
8, 9)), True, "[1, 10)", "Range[1, 10)"),
(Range(1, 10), RangeSet(Range(0,
11)), False, "[1, 10)", "Range[1, 10)"),
(Range(1, 4), RangeSet(Range(2, 3), Range(
5, 6)), False, "[1, 4)", "Range[1, 4)"),
])
def test_range_contains(rng, item, contains, strr, reprr):
"""
Tests the __contains__, __str__, __repr__, and __hash__ methods of the range.
"""
assert (contains == (item in rng))
assert (strr == str(rng))
assert (reprr == repr(rng))
def test_rangeset_isdisjoint(rng1, rng2, isdisjoint, error_type):
if error_type is not None:
asserterror(error_type, rng1.isdisjoint, (rng2, ))
else:
assert (rng1.isdisjoint(rng2) == RangeSet(rng2).isdisjoint(rng1))
assert (isdisjoint == rng1.isdisjoint(rng2))
[Range(0, 1), Range(1.5, 2),
Range(2.5, 3),
Range(4, 5)], "{[0, 1), [1.5, 2), [2.5, 3), [4, 5)}",
"RangeSet{Range[0, 1), Range[1.5, 2), Range[2.5, 3), Range[4, 5)}",
False), # mix Rangelike, iterable args
(["[0, 3]", "[2, 4)", "[5, 6]"], [Range(0, 4),
Range("[5, 6]")], "{[0, 4), [5, 6]}",
"RangeSet{Range[0, 4), Range[5, 6]}", False), # overlapping
(["[0, 4)", "(1, 3)"], [Range(0, 4)], "{[0, 4)}",
"RangeSet{Range[0, 4)}", False), # overlapping 2
([Range(1, 3), Range(2, 4)], [Range(
1, 4)], "{[1, 4)}", "RangeSet{Range[1, 4)}", False),
([Range('apple', 'carrot'),
Range('banana', 'durian')], [Range('apple', 'durian')],
"{[apple, durian)}", "RangeSet{Range['apple', 'durian')}", False),
([RangeSet("(0, 1)", "(1, 2)", "(2, 3)")],
[Range("(0, 1)"), Range("(1, 2)"),
Range("(2, 3)")], "{(0, 1), (1, 2), (2, 3)}",
"RangeSet{Range(0, 1), Range(1, 2), Range(2, 3)}", False)
])
def test_rangeset_constructor_valid(args, ranges, strr, reprr, isempty):
"""
Tests that the constructor of rngset works as intended. Also, as a byproduct,
tests the .ranges(), .__str__(), .__repr__(), .clear(), and .isempty()
"""
rangeset = RangeSet(*args)
assert (ranges == rangeset.ranges())
assert (strr == str(rangeset))
assert (reprr == repr(rangeset))
assert (isempty == rangeset.isempty())
assert (isempty != bool(rangeset))