Skip to content

Commit

Permalink
Tiny fix in field str formatting.
Browse files Browse the repository at this point in the history
  • Loading branch information
@scott-griffiths
scott-griffiths committed Jan 27, 2025
1 parent 26ce255 commit 7a4e388
Show file tree
Hide file tree
Showing 3 changed files with 29 additions and 76 deletions.
99 changes: 23 additions & 76 deletions bitformat/_dtypes.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -429,13 +429,9 @@ def __init__(
def allowed_size_checked_get_fn(bs):
if len(bs) not in self.allowed_sizes:
if self.allowed_sizes.only_one_value():
raise ValueError(
f"'{self.name}' dtypes must have a size of {self.allowed_sizes.values[0]}, but received a size of {len(bs)}."
)
raise ValueError(f"'{self.name}' dtypes must have a size of {self.allowed_sizes.values[0]}, but received a size of {len(bs)}.")
else:
raise ValueError(
f"'{self.name}' dtypes must have a size in {self.allowed_sizes}, but received a size of {len(bs)}."
)
raise ValueError(f"'{self.name}' dtypes must have a size in {self.allowed_sizes}, but received a size of {len(bs)}.")
return get_fn(bs)

self.get_fn = (
Expand All @@ -445,9 +441,7 @@ def allowed_size_checked_get_fn(bs):
self.get_fn = get_fn # Interpret everything
if bits_per_character is not None:
if bitlength2chars_fn is not None:
raise ValueError(
"You shouldn't specify both a bits_per_character and a bitlength2chars_fn."
)
raise ValueError("You shouldn't specify both a bits_per_character and a bitlength2chars_fn.")

def bitlength2chars_fn(x):
return x // bits_per_character
Expand All @@ -473,13 +467,9 @@ def sanitize(self, size: int, endianness: Endianness) -> tuple[int, Endianness]:
)
if endianness != Endianness.UNSPECIFIED:
if not self.endianness_variants:
raise ValueError(
f"The '{self.name}' dtype does not support endianness variants, but '{endianness.value}' was specified."
)
raise ValueError(f"The '{self.name}' dtype does not support endianness variants, but '{endianness.value}' was specified.")
if size % 8 != 0:
raise ValueError(
f"Endianness can only be specified for whole-byte dtypes, but '{self.name}' has a size of {size} bits."
)
raise ValueError(f"Endianness can only be specified for whole-byte dtypes, but '{self.name}' has a size of {size} bits.")
return size, endianness

def get_single_dtype(
Expand All @@ -495,7 +485,8 @@ def get_array_dtype(
size, endianness = self.sanitize(size, endianness)
d = Dtype._create(self, size, True, items, endianness)
if size == 0:
raise ValueError(f"Array dtypes must have a size specified. Got '{d}'. Note that the number of items in the array dtype can be unknown, but the dtype of each item must have a known size.")
raise ValueError(f"Array dtypes must have a size specified. Got '{d}'. "
f"Note that the number of items in the array dtype can be unknown, but the dtype of each item must have a known size.")
return d

def __repr__(self) -> str:
Expand Down Expand Up @@ -547,16 +538,12 @@ def add_dtype(cls, definition: DtypeDefinition):

def fget_be(b):
if len(b) % 8 != 0:
raise ValueError(
f"Cannot use endianness modifer for non whole-byte data. Got length of {len(b)} bits."
)
raise ValueError(f"Cannot use endianness modifer for non whole-byte data. Got length of {len(b)} bits.")
return definition.get_fn(b)

def fget_le(b):
if len(b) % 8 != 0:
raise ValueError(
f"Cannot use endianness modifer for non whole-byte data. Got length of {len(b)} bits."
)
raise ValueError(f"Cannot use endianness modifer for non whole-byte data. Got length of {len(b)} bits.")
return definition.get_fn(b.byte_swap())

fget_ne = fget_le if byteorder == "little" else fget_be
Expand Down Expand Up @@ -598,9 +585,7 @@ def get_single_dtype(
except KeyError:
aliases = {"int": "i", "uint": "u", "float": "f"}
extra = f"Did you mean '{aliases[name]}'? " if name in aliases else ""
raise ValueError(
f"Unknown Dtype name '{name}'. {extra}Names available: {list(cls.name_to_def.keys())}."
)
raise ValueError(f"Unknown Dtype name '{name}'. {extra}Names available: {list(cls.name_to_def.keys())}.")
else:
return definition.get_single_dtype(size, endianness)

Expand All @@ -616,9 +601,7 @@ def get_array_dtype(
try:
definition = cls.name_to_def[name]
except KeyError:
raise ValueError(
f"Unknown Dtype name '{name}'. Names available: {list(cls.name_to_def.keys())}."
)
raise ValueError(f"Unknown Dtype name '{name}'. Names available: {list(cls.name_to_def.keys())}.")
else:
d = definition.get_array_dtype(size, items, endianness)
return d
Expand Down Expand Up @@ -676,9 +659,7 @@ def from_string(cls, token: str, /) -> DtypeWithExpression:
token = "".join(token.split()) # Remove whitespace
if token.startswith("[") and token.endswith("]"):
if (p := token.find(";")) == -1:
raise ValueError(
f"Array Dtype strings should be of the form '[dtype; items]'. Got '{token}'."
)
raise ValueError(f"Array Dtype strings should be of the form '[dtype; items]'. Got '{token}'.")
t = token[p + 1 : -1]
try:
items = int(t) if t else 0
Expand Down Expand Up @@ -717,56 +698,26 @@ def evaluate(self, vars_: dict[str, Any]) -> Dtype:
return self.base_dtype
if self.base_dtype.is_array:
name = self.base_dtype.name
size = (
self.size_expression.evaluate(vars_)
if (self.size_expression and vars_)
else self.base_dtype.size
)
items = (
self.items_expression.evaluate(vars_)
if (self.items_expression and vars_)
else self.base_dtype.items
)
size = self.size_expression.evaluate(vars_) if (self.size_expression and vars_) else self.base_dtype.size
items = self.items_expression.evaluate(vars_) if (self.items_expression and vars_) else self.base_dtype.items
endianness = self.base_dtype.endianness
return Register().get_array_dtype(name, size, items, endianness)
else:
name = self.base_dtype.name
size = (
self.size_expression.evaluate(vars_)
if (self.size_expression and vars_)
else self.base_dtype.size
)
size = self.size_expression.evaluate(vars_) if (self.size_expression and vars_) else self.base_dtype.size
endianness = self.base_dtype.endianness
return Register().get_single_dtype(name, size, endianness)

def __str__(self) -> str:
hide_size = Register().name_to_def[
self.base_dtype.name
].allowed_sizes.only_one_value() or (
self.base_dtype.size == 0 and self.size_expression is None
)
size_str = (
""
if hide_size
else (
self.size_expression
if self.size_expression
else str(self.base_dtype.size)
)
)
only_one_value = Register().name_to_def[self.base_dtype.name].allowed_sizes.only_one_value()
no_value_given = self.base_dtype.size == 0 and self.size_expression is None
hide_size = only_one_value or no_value_given
size_str = "" if hide_size else (self.size_expression if self.size_expression else str(self.base_dtype.size))
if not self.base_dtype.is_array:
return f"{self.base_dtype.name}{self.base_dtype.endianness.value}{size_str}"
hide_items = self.base_dtype.items == 0 and self.items_expression is None
items_str = (
""
if hide_items
else (
self.items_expression
if self.items_expression
else str(self.base_dtype.items)
)
)
return f"[{self.base_dtype.name}{self.base_dtype.endianness.value}{size_str}; {items_str}]"
items_str = "" if hide_items else (" " + self.items_expression if self.items_expression else " " + str(self.base_dtype.items))
return f"[{self.base_dtype.name}{self.base_dtype.endianness.value}{size_str};{items_str}]"


class DtypeTuple:
Expand Down Expand Up @@ -807,9 +758,7 @@ def from_string(cls, s: str, /) -> DtypeTuple:
def pack(self, values: Sequence[Any]) -> bitformat.Bits:
if len(values) != len(self):
raise ValueError(f"Expected {len(self)} values, but got {len(values)}.")
return bitformat.Bits.from_joined(
dtype.pack(value) for dtype, value in zip(self._dtypes, values)
)
return bitformat.Bits.from_joined(dtype.pack(value) for dtype, value in zip(self._dtypes, values))

def unpack(
self,
Expand All @@ -823,9 +772,7 @@ def unpack(
"""
b = bitformat.Bits._from_any(b)
if self.bit_length > len(b):
raise ValueError(
f"{self!r} is {self.bit_length} bits long, but only got {len(b)} bits to unpack."
)
raise ValueError(f"{self!r} is {self.bit_length} bits long, but only got {len(b)} bits to unpack.")
vals = []
pos = 0
for dtype in self:
Expand Down
1 change: 1 addition & 0 deletions tests/test_dtypes.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -276,6 +276,7 @@ def test_str():

def test_unpacking_dtype_array_with_no_length():
d = Dtype('[bool;]')
assert str(d) == '[bool;]'
assert d.unpack('0b110') == (True, True, False)
assert Dtype('[u8;]').unpack('0x0001f') == (0, 1)

Expand Down
5 changes: 5 additions & 0 deletions tests/test_field.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -223,6 +223,11 @@ def test_unpack():
with pytest.raises(ValueError):
_ = f.unpack()

def test_unpack_with_unknown_items():
f = Field("[i9; ]")
assert str(f) == "[i9;]"
f.pack([5, -5, 0, 100])
assert f.unpack() == (5, -5, 0, 100)

def test_field_with_comment():
f = Field.from_params("u8", name="x", comment=" This is a comment ")
Expand Down

0 comments on commit 7a4e388

Please # to comment.