def test_combine_ranges():
r1 = Range(1, 2)
r2 = Range(2, 4)
r3 = Range(10, 12)
l1 = [r1, r2, r3]
assert Range.combine_ranges([r1, r2]) == [Range(1, 4)]
assert Range.combine_ranges([r1, r2, r3]) == [Range(1, 4), Range(10, 12)]
def add_valid_data_description(self, desc):
"""Parse the variable descriptions typical of US Census Bureau data
files that are used to describe valid values."""
if "\n" in desc:
for line in desc.split("\n"):
self.add_valid_data_description(line)
return
if "," in desc:
for line in desc.split(","):
self.add_valid_data_description(line)
return
r = Range.extract_range_and_desc(desc, width=self.width)
self.ranges.add(r)
def python_validator(self):
ret = []
ret.append(" @classmethod")
ret.append(" def {}(self,x):".format(self.python_validator_name()))
ret.append(' """{}"""'.format(self.desc))
if self.allow_any:
ret.append(" return True")
return "\n".join(ret) + "\n"
if self.allow_null:
ret.append(" if x is None or x == \"None\" or len(x) == 0:")
ret.append(" return True")
else:
ret.append(" if x is None or x == \"None\":")
ret.append(" return False")
if self.allow_whitespace:
size = "".rjust(self.width, " ")
ret.append(" if x == '{}':".format(size))
ret.append(" return True")
if self.python_type == int or self.python_type == float:
ret.append(' x = str(x).strip()')
ret.append(' try:')
if self.python_type == int:
ret.append(' x = int(x)')
if self.python_type == float:
ret.append(' x = float(x)')
ret.append(' except ValueError:')
ret.append(' return False')
ranges = Range.combine_ranges(self.ranges)
try:
expr = " or ".join([
"({})".format(r.python_expr(self.python_type, self.width))
for r in ranges
])
except ValueError as e:
logging.error(
"Cannot create python range expression for variable " +
str(self))
raise RuntimeError(
"Cannot create python range expression for variable " +
str(self))
if expr == "":
expr = "True"
ret.append(" return " + expr)
return "\n".join(ret) + "\n"
def test_parse():
res = Range.extract_range_and_desc("1 hello", hardfail=True)
assert type(res) == Range
assert res.a == '1'
assert res.b == '1'
assert res.desc == "hello"
assert Range.extract_range_and_desc("1 hello",
python_type=int,
hardfail=True) == Range(1, 1, "hello")
assert Range.extract_range_and_desc("1-2 hello",
python_type=int,
hardfail=True) == Range(1, 2, "hello")
assert Range.extract_range_and_desc("1-2 = hello",
python_type=int,
hardfail=True) == Range(1, 2, "hello")
assert Range.extract_range_and_desc("1-2 = (hello)",
python_type=int,
hardfail=True) == Range(1, 2, "hello")
assert Range.extract_range_and_desc("1 hello 3-4",
python_type=int,
hardfail=True) == Range(
1, 1, "hello 3-4")
assert Range.extract_range_and_desc("1 (hello 3-4)",
python_type=int,
hardfail=True) == Range(
1, 1, "hello 3-4")
assert Range.extract_range_and_desc("1 = hello 3-4",
python_type=int,
hardfail=True) == Range(
1, 1, "hello 3-4")
assert Range.extract_range_and_desc("1 = (hello 3-4)",
python_type=int,
hardfail=True) == Range(
1, 1, "hello 3-4")
assert Range.extract_range_and_desc("1-2 hello 3-4",
python_type=int,
hardfail=True) == Range(
1, 2, "hello 3-4")
assert Range.extract_range_and_desc("1-2 (hello 3-4)",
python_type=int,
hardfail=True) == Range(
1, 2, "hello 3-4")
assert Range.extract_range_and_desc("1-2 = hello 3-4",
python_type=int,
hardfail=True) == Range(
1, 2, "hello 3-4")
assert Range.extract_range_and_desc("1-2 = (hello 3-4)",
python_type=int,
hardfail=True) == Range(
1, 2, "hello 3-4")
assert Range.extract_range_and_desc(" 1 = hello 3-4",
python_type=int,
hardfail=True) == Range(
1, 1, "hello 3-4")
# A few descriptions that we don't want to parse as Ranges
assert Range.extract_range_and_desc("Up to 22 values",
python_type=int,
hardfail=False) == None
assert Range.extract_range_and_desc(
"Fips State Code (The legal values for this file are 01-02, 04-06)",
python_type=int,
hardfail=False) == None
def test_range_funcs():
r1 = Range(1, 1)
r2 = Range(1, 1)
assert r1 == r2
r3 = Range(2, 2)
assert r1 < r3
def python_validation_text(self):
ranges = Range.combine_ranges(self.ranges)
return ", ".join(['{}-{}'.format(r.a, r.b) for r in ranges])