dissect.cstruct 4.6.dev2__tar.gz → 4.6.dev4__tar.gz

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/CHANGELOG.md

@@ -11,6 +11,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 - Optimize storage of field sizes.
 - Rename `_sizes` property of `Structure` to `__sizes__`.
 - Rename `_values` property of `Structure` to `__values__`.
+- Added `load` argument to `cstruct` class, allowing direct initialization with a definition (i.e. `cstruct(cdef)` instead of `cstruct().load(cdef)`. Other arguments to `cstruct` are now keyword only.
 
 ## [4.5] - 20-05-2025
 

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dissect.cstruct
-Version: 4.6.dev2
+Version: 4.6.dev4
 Summary: A Dissect module implementing a parser for C-like structures: structure parsing in Python made easy
 Author-email: Dissect Team <dissect@fox-it.com>
 License: Apache License 2.0

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect/cstruct/compiler.py

@@ -226,18 +226,18 @@ class _ReadSourceGenerator:
 
     def _generate_structure(self, field: Field) -> Iterator[str]:
         template = f"""
-        {'_s = stream.tell()' if field.type.dynamic else ''}
+        {"_s = stream.tell()" if field.type.dynamic else ""}
         r["{field._name}"] = {self._map_field(field)}._read(stream, context=r)
-        {f's["{field._name}"] = stream.tell() - _s' if field.type.dynamic else ''}
+        {f's["{field._name}"] = stream.tell() - _s' if field.type.dynamic else ""}
         """
 
         yield dedent(template)
 
     def _generate_array(self, field: Field) -> Iterator[str]:
         template = f"""
-        {'_s = stream.tell()' if field.type.dynamic else ''}
+        {"_s = stream.tell()" if field.type.dynamic else ""}
         r["{field._name}"] = {self._map_field(field)}._read(stream, context=r)
-        {f's["{field._name}"] = stream.tell() - _s' if field.type.dynamic else ''}
+        {f's["{field._name}"] = stream.tell() - _s' if field.type.dynamic else ""}
         """
 
         yield dedent(template)

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect/cstruct/cstruct.py

@@ -48,7 +48,7 @@ class cstruct:
     DEF_CSTYLE = 1
     DEF_LEGACY = 2
 
-    def __init__(self, endian: str = "<", pointer: str | None = None):
+    def __init__(self, load: str = "", *, endian: str = "<", pointer: str | None = None):
         self.endian = endian
 
         self.consts = {}
@@ -188,6 +188,9 @@ class cstruct:
         self.pointer: type[BaseType] = self.resolve(pointer)
         self._anonymous_count = 0
 
+        if load:
+            self.load(load)
+
     def __getattr__(self, attr: str) -> Any:
         try:
             return self.consts[attr]
@@ -369,14 +372,14 @@ class cstruct:
             "null_terminated": null_terminated,
         }
 
-        return cast(type[Array], self._make_type(name, bases, size, alignment=type_.alignment, attrs=attrs))
+        return cast("type[Array]", self._make_type(name, bases, size, alignment=type_.alignment, attrs=attrs))
 
     def _make_int_type(self, name: str, size: int, signed: bool, *, alignment: int | None = None) -> type[Int]:
-        return cast(type[Int], self._make_type(name, (Int,), size, alignment=alignment, attrs={"signed": signed}))
+        return cast("type[Int]", self._make_type(name, (Int,), size, alignment=alignment, attrs={"signed": signed}))
 
     def _make_packed_type(self, name: str, packchar: str, base: type, *, alignment: int | None = None) -> type[Packed]:
         return cast(
-            type[Packed],
+            "type[Packed]",
             self._make_type(
                 name,
                 (base, Packed),

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect/cstruct/expression.py

@@ -141,7 +141,7 @@ class ExpressionTokenizer:
                 self.tokens.append(">>")
             elif self.match(expected="<", append=False) and self.match(expected="<", append=False):
                 self.tokens.append("<<")
-            elif self.match(expected={" ", "\t"}, append=False):
+            elif self.match(expected={" ", "\n", "\t"}, append=False):
                 continue
             else:
                 raise ExpressionTokenizerError(

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect/cstruct/parser.py

@@ -63,7 +63,7 @@ class TokenParser(Parser):
             "ENUM",
         )
         TOK.add(r"(?<=})\s*(?P<defs>(?:[a-zA-Z0-9_]+\s*,\s*)+[a-zA-Z0-9_]+)\s*(?=;)", "DEFS")
-        TOK.add(r"(?P<name>\**?\s*[a-zA-Z0-9_]+)(?:\s*:\s*(?P<bits>\d+))?(?:\[(?P<count>[^;\n]*)\])?\s*(?=;)", "NAME")
+        TOK.add(r"(?P<name>\**?\s*[a-zA-Z0-9_]+)(?:\s*:\s*(?P<bits>\d+))?(?:\[(?P<count>[^;]*)\])?\s*(?=;)", "NAME")
         TOK.add(r"#include\s+(?P<name>[^\s]+)\s*", "INCLUDE")
         TOK.add(r"[a-zA-Z_][a-zA-Z0-9_]*", "IDENTIFIER")
         TOK.add(r"[{}]", "BLOCK")
@@ -194,7 +194,7 @@ class TokenParser(Parser):
         if tokens.next == self.TOK.NAME:
             # As part of a struct field
             # struct type_name field_name;
-            if not len(names):
+            if not names:
                 raise ParserError(f"line {self._lineno(tokens.next)}: unexpected anonymous struct")
             return self.cstruct.resolve(names[0])
 

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect/cstruct/types/structure.py

@@ -139,7 +139,7 @@ class StructureMetaType(MetaType):
 
         if cls.__compiled__:
             # If the previous class was compiled try to compile this too
-            from dissect.cstruct import compiler
+            from dissect.cstruct import compiler  # noqa: PLC0415
 
             try:
                 classdict["_read"] = compiler.Compiler(cls.cs).compile_read(fields, cls.__name__, align=cls.__align__)

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/dissect.cstruct.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dissect.cstruct
-Version: 4.6.dev2
+Version: 4.6.dev4
 Summary: A Dissect module implementing a parser for C-like structures: structure parsing in Python made easy
 Author-email: Dissect Team <dissect@fox-it.com>
 License: Apache License 2.0

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/pyproject.toml

@@ -42,7 +42,7 @@ cstruct-stubgen = "dissect.cstruct.tools.stubgen:main"
 
 [tool.ruff]
 line-length = 120
-required-version = ">=0.9.0"
+required-version = ">=0.12.0"
 
 [tool.ruff.format]
 docstring-code-format = true
@@ -85,7 +85,7 @@ select = [
 ignore = ["E203", "B904", "UP024", "ANN002", "ANN003", "ANN204", "ANN401", "SIM105", "TRY003"]
 
 [tool.ruff.lint.per-file-ignores]
-"tests/docs/**" = ["INP001"]
+"tests/_docs/**" = ["INP001"]
 
 [tool.ruff.lint.isort]
 known-first-party = ["dissect.cstruct"]

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/tests/test_annotations.py

@@ -21,7 +21,7 @@ def test_cstruct_type_annotation(name: str, monkeypatch: pytest.MonkeyPatch) ->
         for module in [module for module in sys.modules if module in ("dissect.cstruct.cstruct")]:
             monkeypatch.delitem(sys.modules, module)
 
-        from dissect.cstruct import cstruct
+        from dissect.cstruct import cstruct  # noqa: PLC0415
 
         if name.startswith("__"):
             name = f"_cstruct{name}"

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/tests/test_basic.py

@@ -6,6 +6,7 @@ from typing import TYPE_CHECKING, BinaryIO
 
 import pytest
 
+from dissect.cstruct.cstruct import cstruct
 from dissect.cstruct.exceptions import ArraySizeError, ParserError, ResolveError
 from dissect.cstruct.types import BaseType
 
@@ -14,8 +15,6 @@ from .utils import verify_compiled
 if TYPE_CHECKING:
     from pathlib import Path
 
-    from dissect.cstruct.cstruct import cstruct
-
 
 def test_duplicate_type(cs: cstruct, compiled: bool) -> None:
     cdef = """
@@ -41,6 +40,51 @@ def test_load_file(cs: cstruct, compiled: bool, tmp_path: Path) -> None:
     assert "test" in cs.typedefs
 
 
+def test_load_init() -> None:
+    cdef = """
+    struct test {
+        DWORD   a;
+        QWORD   b;
+    };
+    """
+    # load with first positional argument
+    cs = cstruct(cdef)
+    assert "test" in cs.typedefs
+    assert cs.endian == "<"
+
+    # load from keyword argument and big endian
+    cs = cstruct(load=cdef, endian=">")
+    assert "test" in cs.typedefs
+    a = cs.test(a=0xBADC0DE, b=0xACCE55ED)
+    assert len(bytes(a)) == 12
+    assert bytes(a) == a.dumps()
+    assert bytes(a) == b"\x0b\xad\xc0\xde\x00\x00\x00\x00\xac\xce\x55\xed"
+
+    # load using positional argument and little endian
+    cs = cstruct(cdef, endian="<")
+    assert "test" in cs.typedefs
+    a = cs.test(a=0xBADC0DE, b=0xACCE55ED)
+    assert len(bytes(a)) == 12
+    assert bytes(a) == a.dumps()
+    assert bytes(a) == b"\xde\xc0\xad\x0b\xed\x55\xce\xac\x00\x00\x00\x00"
+
+
+def test_load_init_kwargs_only() -> None:
+    cdef = """
+    struct test {
+        uint32  a;
+    };
+    """
+
+    # kwargs only check
+    with pytest.raises(TypeError, match="takes from .* positional arguments but .* were given"):
+        cs = cstruct(cdef, ">")
+
+    cs = cstruct(cdef, endian=">")
+    assert "test" in cs.typedefs
+    assert cs.endian == ">"
+
+
 def test_read_type_name(cs: cstruct) -> None:
     assert cs.read("uint32", b"\x01\x00\x00\x00") == 1
 

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/tests/test_parser.py

@@ -149,4 +149,3 @@ def test_includes(cs: cstruct) -> None:
     assert cs.myStruct.__name__ == "myStruct"
     assert len(cs.myStruct.fields) == 1
     assert cs.myStruct.fields.get("charVal")
-

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/tests/test_types_structure.py

@@ -265,8 +265,8 @@ def test_structure_definition_simple(cs: cstruct, compiled: bool) -> None:
     with pytest.raises(AttributeError):
         obj.nope  # noqa: B018
 
-    assert obj.__dynamic_sizes__ == { "string": 7, "wstring": 10 }
-    assert obj.__sizes__ == { "magic": 4, "wmagic": 8, "a": 1, "b": 2, "c": 4, "string": 7, "wstring": 10 }
+    assert obj.__dynamic_sizes__ == {"string": 7, "wstring": 10}
+    assert obj.__sizes__ == {"magic": 4, "wmagic": 8, "a": 1, "b": 2, "c": 4, "string": 7, "wstring": 10}
     assert len(obj) == len(buf)
     assert obj.dumps() == buf
 
@@ -276,6 +276,7 @@ def test_structure_definition_simple(cs: cstruct, compiled: bool) -> None:
     obj.write(fh)
     assert fh.getvalue() == buf
 
+
 def test_structure_values_dict(cs: cstruct, compiled: bool) -> None:
     cdef = """
     struct test {
@@ -810,3 +811,36 @@ def test_codegen_hashable(cs: cstruct) -> None:
 
     assert hash(structure._generate_structure__init__(hashable_fields).__code__)
     assert hash(structure._generate_structure__init__(unhashable_fields).__code__)
+
+
+def test_structure_definition_newline(cs: cstruct, compiled: bool) -> None:
+    cdef = """
+    struct test {
+        char    magic[4
+        ];
+
+        wchar   wmagic[4];
+        uint8   a;
+        uint16  b;
+        uint32  c;
+        char    string[];
+        wchar   wstring[];
+    };
+    """
+    cs.endian = ">"
+    cs.load(cdef, compiled=compiled)
+
+    assert verify_compiled(cs.test, compiled)
+
+    buf = b"test\x00t\x00e\x00s\x00t\x01\x02\x03\x04\x05\x06\x07lalala\x00\x00t\x00e\x00s\x00t\x00\x00"
+
+    obj = cs.test()
+    obj.magic = b"test"
+    obj.wmagic = "test"
+    obj.a = 0x01
+    obj.b = 0x0203
+    obj.c = 0x04050607
+    obj.string = b"lalala"
+    obj.wstring = "test"
+
+    assert obj.dumps() == buf

{dissect_cstruct-4.6.dev2 → dissect_cstruct-4.6.dev4}/tox.ini

@@ -32,16 +32,18 @@ commands =
 [testenv:fix]
 package = skip
 deps =
-    ruff==0.9.2
+    ruff==0.12.4
 commands =
     ruff format dissect tests
+    ruff check --fix dissect tests
 
 [testenv:lint]
 package = skip
 deps =
-    ruff==0.9.2
+    ruff==0.12.4
     vermin
 commands =
+    ruff format --check dissect tests
     ruff check dissect tests
     vermin -t=3.9- --no-tips --lint dissect tests