sharedbox 0.1.0__cp313-cp313-win_amd64.whl

sharedbox/__init__.py

@@ -0,0 +1,3 @@
+from ._shareddict import SharedDict
+
+__all__ = ["SharedDict"]

sharedbox/_shareddict.pyi

@@ -0,0 +1,32 @@
+from typing import Any, Generic, TypeVar
+
+K = TypeVar("K")
+V = TypeVar("V")
+
+class SharedDict(Generic[K, V]):
+    def __init__(
+        self,
+        name: str,
+        data: dict[str, Any] | None = None,
+        /,
+        *,
+        size: int = 128 * 1024 * 1024,
+        create: bool = False,
+        max_keys: int = 128,
+    ) -> None: ...
+    def close(self) -> None: ...
+    def unlink(self) -> None: ...
+    def is_closed(self) -> bool: ...
+    def __len__(self) -> int: ...
+    def __contains__(self, key: K) -> bool: ...
+    def __getitem__(self, key: K) -> V: ...
+    def __setitem__(self, key: K, value: V) -> None: ...
+    def __delitem__(self, key: K) -> None: ...
+    def __iter__(self) -> Any: ...
+    def get(self, key: K, default: V = None) -> V: ...
+    def keys(self) -> list[K]: ...
+    def keys_atomic(self) -> list[K]: ...
+    def items(self) -> list[tuple[K, V]]: ...
+    def recommend_sizing(
+        self, target_entries: int | None = None
+    ) -> dict[str, object]: ...

sharedbox/_shareddict.pyx

@@ -0,0 +1,350 @@
+# distutils: language = c++
+from libcpp.string cimport string
+from libcpp.vector cimport vector
+from libcpp cimport bool as cbool
+import pickle
+import struct
+from cpython.bytes cimport PyBytes_FromStringAndSize
+from cpython.unicode cimport PyUnicode_AsUTF8AndSize, PyUnicode_DecodeUTF8
+from .utils import SegmentSizer, LockTuner
+import numpy as np
+
+cdef inline int calculate_target_entries(int current_entries, int min_entries) nogil:
+    """Pure C function to calculate target entries with 10x growth."""
+    cdef int growth_target = current_entries * 10
+    return growth_target if growth_target > min_entries else min_entries
+
+cdef inline str _decode_to_str(const string& key_string):
+    """Encode a C++ string object to a Python string."""
+    return PyUnicode_DecodeUTF8(key_string.c_str(), key_string.size(), NULL)
+
+cdef inline string _encode_to_string(str key):
+    """Encode a Python string to a C++ string using UTF-8 encoding."""
+    cdef const char* key_ptr
+    cdef Py_ssize_t key_len
+    key_ptr = PyUnicode_AsUTF8AndSize(key, &key_len)
+    return string(key_ptr, key_len)
+
+
+cdef extern from "shared_dict.hpp" namespace "shared_memory":
+    cdef cppclass SharedMemoryDict:
+        SharedMemoryDict(const string& name, size_t size, cbool create, size_t max_keys) except +
+        void set(const string& k, const string& v) except +
+        cbool get(const string& k, string& out) const
+        cbool erase(const string& k)
+        cbool contains(const string& k) const
+        size_t size() const
+        vector[string] keys() const
+        void close() except +
+        void unlink() except +
+        cbool is_closed() const
+
+cdef class SharedDict:
+    cdef SharedMemoryDict* c_map
+    cdef str name
+
+    def __cinit__(self, str name, dict data = None, /, *, int size = 128 * 1024 * 1024, cbool create = True, int max_keys = 128) -> None:
+        self.name = name
+        cdef string nm = _encode_to_string(name)
+        self.c_map = new SharedMemoryDict(nm, <size_t>size, <cbool>create, <size_t>max_keys)
+        
+        # Initialize with provided data if specified
+        if data is not None:
+            self._initialize_data(data)
+
+    def __dealloc__(self) -> None:
+        """Destructor that releases the connection but does NOT remove shared memory.
+        
+        This follows the multiprocessing.SharedMemory API pattern:
+        - Destructor only closes the connection (like close())
+        - Shared memory removal must be done explicitly via unlink()
+        - This prevents race conditions when multiple processes use the same segment
+        """
+        if self.c_map is not NULL:
+            # The C++ destructor will call close() if not already closed
+            del self.c_map
+            self.c_map = NULL
+
+    cpdef void close(self):
+        """Close access to shared memory without removing it.
+        
+        After calling close(), this SharedDict instance becomes unusable
+        but other processes can still access the shared memory.
+        Similar to multiprocessing.SharedMemory.close().
+        """
+        if self.c_map is not NULL:
+            self.c_map.close()
+
+    cpdef void unlink(self):
+        """Remove the shared memory segment entirely.
+        
+        This removes the shared memory from the system, making it
+        inaccessible to all processes. Similar to multiprocessing.SharedMemory.unlink().
+        Only call this from the process that created the segment.
+        """
+        if self.c_map is not NULL:
+            if not self.c_map.is_closed():
+                raise RuntimeError("Cannot unlink a SharedDict that is still open. Call close() first.")
+            self.c_map.unlink()
+
+    cpdef cbool is_closed(self):
+        """Check if this SharedDict connection has been closed.
+        
+        Returns True if close() has been called or the object has been destructed.
+        A closed SharedDict cannot perform any operations.
+        """
+        if self.c_map is NULL:
+            return True
+        return self.c_map.is_closed()
+
+    cdef bytes _dumps_value(self, object obj):
+        """Value serialization with numpy array support."""
+        if isinstance(obj, np.ndarray):
+            return self._serialize_numpy_array(obj)
+        else:
+            # Fallback to pickle for other types
+            header = b'\x00'  # Non-numpy marker
+            data = pickle.dumps(obj, protocol=pickle.HIGHEST_PROTOCOL)
+            return header + data
+
+    cdef object _loads_value(self, bytes b):
+        """Value deserialization with numpy array support."""
+        if len(b) == 0:
+            raise ValueError("Empty data cannot be deserialized")
+        
+        cdef unsigned char marker = b[0]
+        if marker == 1:  # Numpy array marker
+            return self._deserialize_numpy_array(b[1:])
+        else:  # Pickle marker (0) or legacy data
+            if marker == 0:
+                return pickle.loads(b[1:])  # Skip marker
+            else:
+                return pickle.loads(b)  # Legacy data without marker
+    
+    cdef bytes _serialize_numpy_array(self, object arr):
+        """Efficiently serialize numpy array without pickle."""
+        cdef bytes dtype_str = str(arr.dtype).encode('utf-8')
+        
+        # Get raw array data - ensure it's contiguous
+        cdef bytes array_data
+        if arr.flags.c_contiguous:
+            array_data = arr.tobytes()
+        else:
+            # Make contiguous copy
+            array_data = np.ascontiguousarray(arr).tobytes()
+        
+        # Pack header: marker(1) + dtype_len(4) + ndim(4) + data_len(4)
+        cdef bytes header = struct.pack('<BIII', 
+                                       1,  # Numpy marker
+                                       len(dtype_str),
+                                       arr.ndim, 
+                                       len(array_data))
+        
+        # Pack shape data
+        cdef bytes shape_data = b''
+        for dim in arr.shape:
+            shape_data += struct.pack('<Q', dim)
+        
+        # Combine all parts
+        return header + dtype_str + shape_data + array_data
+    
+    cdef object _deserialize_numpy_array(self, bytes data):
+        """Efficiently deserialize numpy array."""
+        cdef int offset = 0
+        
+        # Unpack header: dtype_len(4) + ndim(4) + data_len(4)  
+        dtype_len, ndim, data_len = struct.unpack('<III', data[offset:offset+12])
+        offset += 12
+        
+        # Extract dtype string
+        dtype_str = data[offset:offset+dtype_len].decode('utf-8')
+        offset += dtype_len
+        
+        # Extract shape
+        cdef list shape = []
+        for i in range(ndim):
+            dim = struct.unpack('<Q', data[offset:offset+8])[0]
+            shape.append(dim)
+            offset += 8
+        
+        # Extract array data
+        array_data = data[offset:offset+data_len]
+        
+        # Reconstruct numpy array
+        arr = np.frombuffer(array_data, dtype=dtype_str).reshape(tuple(shape))
+        
+        # Make a copy to ensure proper memory ownership
+        return np.array(arr)
+
+    cdef void _initialize_data(self, dict data):
+        """Initialize SharedDict with provided data.
+        
+        Args:
+            data: Dictionary with string keys and values to populate SharedDict.
+                  Values must be serializable (built-in types, numpy arrays, etc.)
+                  
+        Raises:
+            TypeError: If data is not a dictionary or contains invalid keys/values
+            ValueError: If serialization fails for any value
+        """
+        if not isinstance(data, dict):
+            raise TypeError("Initialization data must be a dictionary")
+        
+        cdef int initialized_count = 0
+        cdef object key_obj
+        cdef str key = ""
+        cdef object value
+        
+        try:
+            for key_obj, value in data.items():
+                if not isinstance(key_obj, str):
+                    raise TypeError(f"All keys must be strings, got {type(key_obj)} for key: {key_obj}")
+                
+                key = <str>key_obj  # Now we know it's safe to cast
+                # Use the existing __setitem__ method for consistency
+                self[key] = value
+                initialized_count += 1
+                
+        except TypeError as e:
+            # Re-raise TypeError with original message for key type errors
+            if "All keys must be strings" in str(e):
+                raise e
+            else:
+                raise ValueError(
+                    f"Failed to initialize SharedDict during key iteration: {e}"
+                ) from e
+        except Exception as e:
+            # If initialization fails partway through, provide helpful context
+            if key:
+                raise ValueError(
+                    f"Failed to initialize SharedDict after {initialized_count} items. "
+                    f"Error on key '{key}': {e}"
+                ) from e
+            else:
+                raise ValueError(
+                    f"Failed to initialize SharedDict during iteration: {e}"
+                ) from e
+
+    def __len__(self) -> int:
+        return <int> self.c_map.size()
+
+    def __contains__(self, str key) -> bool:
+        cdef string ks = _encode_to_string(key)
+        return bool(self.c_map.contains(ks))
+
+    def __getitem__(self, str key) -> object:
+        cdef string ks = _encode_to_string(key)
+        cdef string out
+        if not self.c_map.get(ks, out):
+            raise KeyError(key)
+        # Convert std::string (raw bytes) to Python bytes then unpickle
+        cdef bytes b = <bytes>PyBytes_FromStringAndSize(out.c_str(), out.size())
+        return self._loads_value(b)
+
+    def get(self, str key, object default = None) -> object:
+        try:
+            return self[key]
+        except KeyError:
+            return default
+
+    def __setitem__(self, str key, object value) -> None:
+        cdef string ks = _encode_to_string(key)
+        cdef bytes vb = self._dumps_value(value)
+        cdef string vs = vb
+        self.c_map.set(ks, vs)
+
+    def __delitem__(self, str key) -> None:
+        cdef string ks = _encode_to_string(key)
+        if not self.c_map.erase(ks):
+            raise KeyError(key)
+
+    def __iter__(self):
+        cdef vector[string] ks = self.c_map.keys()  # Use atomic full-lock version instead of striped
+        cdef string s
+        for i in range(ks.size()):
+            s = ks[i]
+            yield _decode_to_str(s)
+
+    def keys(self) -> list[str]:
+        return list(iter(self))
+    
+    def keys_atomic(self) -> list[str]:
+        """Get keys with full atomic snapshot (locks all stripes at once)."""
+        cdef vector[string] ks = self.c_map.keys()  # Full lock version
+        cdef string s
+        result: list[str] = []
+        for i in range(ks.size()):
+            s = ks[i]
+            result.append(_decode_to_str(s))
+        return result
+
+    def items(self) -> list[tuple[str, object]]:
+        return [(k, self[k]) for k in self]
+
+    def values(self) -> list[object]:
+        return [self[k] for k in self]
+    
+    def get_stats(self) -> dict[str, object]:
+        """Get runtime statistics and diagnostic information."""
+        # Sample some keys to estimate sizes
+        cdef vector[string] sample_keys = self.c_map.keys()
+        cdef string key_bytes, value_bytes
+        sample_size = min(<int>sample_keys.size(), 100)
+        
+        total_key_bytes = 0
+        total_value_bytes = 0
+        
+        if sample_size > 0:
+            for i in range(sample_size):
+                key_bytes = sample_keys[i] 
+                total_key_bytes += key_bytes.size()
+                if self.c_map.get(key_bytes, value_bytes):
+                    total_value_bytes += value_bytes.size()
+        
+        avg_key_bytes = total_key_bytes / sample_size if sample_size > 0 else 0
+        avg_value_bytes = total_value_bytes / sample_size if sample_size > 0 else 0
+        
+        return {
+            'total_entries': <int>self.c_map.size(),
+            'sample_size': sample_size,
+            'avg_key_utf8_bytes': avg_key_bytes,
+            'avg_value_pickle_bytes': avg_value_bytes,
+            'estimated_data_bytes': <int>self.c_map.size() * (avg_key_bytes + avg_value_bytes),
+            'segment_name': self.name,
+        }
+    
+    def recommend_sizing(self, target_entries: int | None = None) -> dict[str, object]:
+        """Get sizing recommendations based on current usage."""
+        stats = self.get_stats()
+        current_entries = stats['total_entries']
+        
+        if target_entries is None:
+            target_entries = calculate_target_entries(current_entries, 10000)
+        
+        if current_entries == 0:
+            return {
+                'current_stats': stats,
+                'target_entries': target_entries,
+                'sizing_recommendation': None,
+                'lock_recommendation': None,
+                'message': 'No data in SharedMemoryDict yet - cannot provide recommendations'
+            }
+        
+        # Size analysis
+        sizing = SegmentSizer.calculate_segment_size(
+            target_entries,
+            int(stats['avg_key_utf8_bytes']),
+            int(stats['avg_value_pickle_bytes'])
+        )
+        
+        # Lock analysis  
+        lock_rec = LockTuner.recommend_lock_count(target_entries)
+        
+        # Return data instead of printing
+        
+        return {
+            'current_stats': stats,
+            'target_entries': target_entries,
+            'sizing_recommendation': sizing,
+            'lock_recommendation': lock_rec
+        }

sharedbox/_version.py

@@ -0,0 +1,34 @@
+# file generated by setuptools-scm
+# don't change, don't track in version control
+
+__all__ = [
+    "__version__",
+    "__version_tuple__",
+    "version",
+    "version_tuple",
+    "__commit_id__",
+    "commit_id",
+]
+
+TYPE_CHECKING = False
+if TYPE_CHECKING:
+    from typing import Tuple
+    from typing import Union
+
+    VERSION_TUPLE = Tuple[Union[int, str], ...]
+    COMMIT_ID = Union[str, None]
+else:
+    VERSION_TUPLE = object
+    COMMIT_ID = object
+
+version: str
+__version__: str
+__version_tuple__: VERSION_TUPLE
+version_tuple: VERSION_TUPLE
+commit_id: COMMIT_ID
+__commit_id__: COMMIT_ID
+
+__version__ = version = '0.1.0'
+__version_tuple__ = version_tuple = (0, 1, 0)
+
+__commit_id__ = commit_id = 'gec7f23c35'

sharedbox/utils.py

@@ -0,0 +1,293 @@
+import math
+import pickle
+from typing import Any
+
+
+class SegmentSizer:
+    """Helper class for calculating optimal shared memory segment sizes."""
+
+    # Conservative estimates for overhead per entry
+    MAP_NODE_OVERHEAD = 128  # bytes per map entry for pointers, alignment, etc.
+    ALLOCATOR_OVERHEAD_RATIO = 0.05  # 5% of total segment for allocator metadata
+    MUTEX_SIZE = 96  # bytes per interprocess_mutex (conservative estimate)
+
+    @classmethod
+    def estimate_pickle_size(cls, obj: Any) -> int:
+        """Estimate the pickle size of an object."""
+        try:
+            return len(pickle.dumps(obj, protocol=pickle.HIGHEST_PROTOCOL))
+        except Exception:
+            # Fallback for unpicklable objects
+            return len(str(obj).encode("utf-8"))
+
+    @classmethod
+    def calculate_segment_size(
+        cls,
+        num_entries: int,
+        avg_key_size: int,
+        avg_value_size: int,
+        num_locks: int = 2048,
+        safety_margin: float = 1.5,
+    ) -> dict[str, int]:
+        """
+        Calculate recommended segment size based on workload characteristics.
+
+        Args:
+            num_entries: Expected number of dictionary entries
+            avg_key_size: Average size of pickled keys in bytes
+            avg_value_size: Average size of pickled values in bytes
+            num_locks: Number of lock stripes (default 2048)
+            safety_margin: Multiplier for safety/fragmentation (default 1.5x)
+
+        Returns:
+            dict with size breakdown and recommendation
+        """
+        # Calculate component sizes
+        data_size: int = num_entries * (avg_key_size + avg_value_size)
+        entry_overhead: int = num_entries * cls.MAP_NODE_OVERHEAD
+        locks_overhead: int = num_locks * cls.MUTEX_SIZE
+
+        # Base size before safety margin
+        base_size: int = data_size + entry_overhead + locks_overhead
+
+        # Add allocator overhead
+        allocator_overhead: int = int(base_size * cls.ALLOCATOR_OVERHEAD_RATIO)
+        base_size += allocator_overhead
+
+        # Apply safety margin for fragmentation and growth
+        recommended_size: int = int(base_size * safety_margin)
+
+        # Round up to next power of 2 for better allocation
+        recommended_size_pow2: int = 2 ** math.ceil(
+            math.log2(max(recommended_size, 1024 * 1024))
+        )
+
+        return {
+            "data_bytes": data_size,
+            "entry_overhead_bytes": entry_overhead,
+            "locks_overhead_bytes": locks_overhead,
+            "allocator_overhead_bytes": allocator_overhead,
+            "base_size_bytes": base_size,
+            "recommended_size_bytes": recommended_size,
+            "recommended_size_pow2_bytes": recommended_size_pow2,
+            "safety_margin": safety_margin,
+        }
+
+    @classmethod
+    def size_for_workload(cls, workload_type: str) -> dict[str, Any]:
+        """
+        Get pre-calculated size recommendations for common workload patterns.
+
+        Args:
+            workload_type: One of 'small', 'medium', 'large', 'xlarge'
+
+        Returns:
+            dict with workload parameters and size recommendations
+        """
+        workloads = {
+            "small": {
+                "description": "1K entries, ~100 bytes avg per key+value",
+                "num_entries": 1_000,
+                "avg_key_size": 50,
+                "avg_value_size": 50,
+                "num_locks": 1024,
+            },
+            "medium": {
+                "description": "100K entries, ~500 bytes avg per key+value",
+                "num_entries": 100_000,
+                "avg_key_size": 100,
+                "avg_value_size": 400,
+                "num_locks": 2048,
+            },
+            "large": {
+                "description": "1M entries, ~1KB avg per key+value",
+                "num_entries": 1_000_000,
+                "avg_key_size": 200,
+                "avg_value_size": 800,
+                "num_locks": 8192,
+            },
+            "xlarge": {
+                "description": "10M entries, ~2KB avg per key+value",
+                "num_entries": 10_000_000,
+                "avg_key_size": 400,
+                "avg_value_size": 1600,
+                "num_locks": 16384,
+            },
+        }
+
+        if workload_type not in workloads:
+            raise ValueError(
+                f"Unknown workload type '{workload_type}'. "
+                f"Available: {list(workloads.keys())}"
+            )
+
+        params = workloads[workload_type]
+        sizing = cls.calculate_segment_size(
+            params["num_entries"],
+            params["avg_key_size"],
+            params["avg_value_size"],
+            params["num_locks"],
+        )
+
+        return {**params, **sizing}
+
+    @classmethod
+    def analyze_sample_data(
+        cls, sample_keys: list[Any], sample_values: list[Any]
+    ) -> dict[str, Any]:
+        """
+        Analyze sample data to estimate sizing requirements.
+
+        Args:
+            sample_keys: Representative sample of keys
+            sample_values: Representative sample of values
+
+        Returns:
+            dict with analysis results and size recommendations
+        """
+        if len(sample_keys) != len(sample_values):
+            raise ValueError("sample_keys and sample_values must have same length")
+
+        if not sample_keys:
+            raise ValueError("Empty samples provided")
+
+        # Calculate pickle sizes
+        key_sizes = [cls.estimate_pickle_size(k) for k in sample_keys]
+        value_sizes = [cls.estimate_pickle_size(v) for v in sample_values]
+
+        # Statistical analysis
+        avg_key_size = sum(key_sizes) // len(key_sizes)
+        avg_value_size = sum(value_sizes) // len(value_sizes)
+        max_key_size = max(key_sizes)
+        max_value_size = max(value_sizes)
+
+        # Generate recommendations for different scales
+        recommendations = {}
+        for scale, multiplier in [
+            ("current", 1),
+            ("10x", 10),
+            ("100x", 100),
+            ("1000x", 1000),
+        ]:
+            num_entries = len(sample_keys) * multiplier
+            sizing = cls.calculate_segment_size(
+                num_entries, avg_key_size, avg_value_size
+            )
+            recommendations[scale] = sizing
+
+        return {
+            "sample_size": len(sample_keys),
+            "avg_key_size_bytes": avg_key_size,
+            "avg_value_size_bytes": avg_value_size,
+            "max_key_size_bytes": max_key_size,
+            "max_value_size_bytes": max_value_size,
+            "key_sizes": key_sizes,
+            "value_sizes": value_sizes,
+            "recommendations": recommendations,
+        }
+
+
+class LockTuner:
+    """Helper class for lock count tuning and performance optimization."""
+
+    @classmethod
+    def recommend_lock_count(
+        cls,
+        num_entries: int,
+        write_concurrency: int = 4,
+        target_entries_per_lock: int = 500,
+    ) -> dict[str, Any]:
+        """
+        Recommend optimal lock count based on workload characteristics.
+
+        Args:
+            num_entries: Expected number of dictionary entries
+            write_concurrency: Expected number of concurrent writers
+            target_entries_per_lock: Target entries per lock stripe
+
+        Returns:
+            dict with recommendations and rationale
+        """
+        # Calculate lock count based on entries per lock target
+        locks_for_entries: int = max(64, num_entries // target_entries_per_lock)
+
+        # Ensure enough locks for write concurrency (at least 4x writers)
+        locks_for_concurrency: int = write_concurrency * 4
+
+        # Take the maximum and round to next power of 2
+        min_locks: int = max(locks_for_entries, locks_for_concurrency, 64)
+        recommended_locks: int = 2 ** math.ceil(math.log2(min_locks))
+
+        # Cap at reasonable maximum (too many locks waste memory)
+        max_reasonable = 65536
+        if recommended_locks > max_reasonable:
+            recommended_locks = max_reasonable
+
+        # Calculate actual entries per lock with recommendation
+        actual_entries_per_lock = (
+            num_entries / recommended_locks if recommended_locks else 0
+        )
+
+        # Memory cost of locks
+        lock_memory_kb = (recommended_locks * SegmentSizer.MUTEX_SIZE) // 1024
+
+        return {
+            "recommended_lock_count": recommended_locks,
+            "actual_entries_per_lock": actual_entries_per_lock,
+            "lock_memory_kb": lock_memory_kb,
+            "rationale": {
+                "target_entries_per_lock": target_entries_per_lock,
+                "locks_for_entries": locks_for_entries,
+                "locks_for_concurrency": locks_for_concurrency,
+                "write_concurrency": write_concurrency,
+            },
+        }
+
+    @classmethod
+    def performance_presets(cls) -> dict[str, dict[str, Any]]:
+        """
+        Get pre-configured lock count recommendations for different performance profiles.
+
+        Returns:
+            dict mapping preset names to lock configurations
+        """
+        return {
+            "memory_optimized": {
+                "description": "Minimal memory usage, lower concurrency",
+                "base_locks": 256,
+                "max_locks": 1024,
+                "entries_per_lock": 2000,
+                "suitable_for": "Memory-constrained environments, low write concurrency",
+            },
+            "balanced": {
+                "description": "Good balance of memory and performance",
+                "base_locks": 1024,
+                "max_locks": 8192,
+                "entries_per_lock": 500,
+                "suitable_for": "Most general-purpose applications",
+            },
+            "performance_optimized": {
+                "description": "Maximum concurrency, higher memory usage",
+                "base_locks": 4096,
+                "max_locks": 32768,
+                "entries_per_lock": 100,
+                "suitable_for": "High-concurrency, write-heavy workloads",
+            },
+            "extreme_performance": {
+                "description": "Ultra-high concurrency for specialized use cases",
+                "base_locks": 16384,
+                "max_locks": 65536,
+                "entries_per_lock": 50,
+                "suitable_for": "Specialized high-throughput applications",
+            },
+        }
+
+
+def format_size(size_bytes: int) -> str:
+    """Format byte size in human-readable format."""
+    size_float: float = size_bytes
+    for unit in ["B", "KB", "MB", "GB"]:
+        if size_float < 1024:
+            return f"{size_float:.1f}{unit}"
+        size_float /= 1024
+    return f"{size_float:.1f}TB"

sharedbox-0.1.0.dist-info/METADATA

@@ -0,0 +1,155 @@
+Metadata-Version: 2.4
+Name: sharedbox
+Version: 0.1.0
+Summary: Python containers using shared memory.
+Author-email: Jacopo Abramo <jacopo.abramo@gmail.com>
+License-Expression: Apache-2.0
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: C++
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=2.2.6
+Provides-Extra: dev
+Requires-Dist: pytest>=8.4.2; extra == "dev"
+Requires-Dist: ruff>=0.13.1; extra == "dev"
+Dynamic: license-file
+
+# `sharedbox`
+
+> [!WARNING]
+> This project is a work in progress; be patient or feel free to contribute.
+
+Python inter-process shared containers leveraging `boost::interprocess`.
+
+## Installation
+
+It is reccomended to install `sharedbox` in a virtual environment; for example using `uv`:
+
+```sh
+uv venv --python 3.10
+.venv\Scripts\activate
+uv pip install sharedbox
+```
+
+## Quick Start
+
+```python
+import multiprocessing as mp
+from sharedbox import SharedDict
+
+# Use in child processes
+def worker(segment_name):
+    d = SharedDict(segment_name, create=False)  # Connect to existing
+    d["worker_data"] = "Hello from worker!"
+    d.close()  # Close in child process
+
+if __name__ == "__main__":
+    # Create a shared dictionary
+    d = SharedDict("my_segment", create=True, size=10*1024*1024)
+    d["hello"] = "world"
+    d["data"] = [1, 2, 3, 4, 5]
+
+    # Start worker
+    p = mp.Process(target=worker, args=("my_segment",))
+    p.start()
+    p.join()
+
+    print(d["worker_data"])  # "Hello from worker!"
+    d.close() # Close in the main process
+    d.unlink()  # Unlink (free resources)
+```
+
+## Initialization with Data
+
+You can initialize SharedDict with existing data for convenient setup:
+
+```python
+import numpy as np
+from sharedbox import SharedDict
+
+# Initialize with mixed data types
+config_data = {
+    "app_name": "MyApp",
+    "version": "1.0",
+    "max_users": 1000,
+    "features": ["auth", "logging"],
+    "model_weights": np.array([0.1, 0.3, 0.6])
+}
+
+# Create SharedDict with initial data
+shared_config = SharedDict("config", config_data, create=True)
+
+# Data is immediately available
+print(shared_config["app_name"])  # "MyApp"
+print(shared_config["model_weights"])  # numpy array
+
+# when a child process doesn't need the memory anymore call "close"
+shared_config.close()
+
+# the main process is in charge of cleaning up; call "unlink" to do so,
+# similarly to a regular python SharedMemory object;
+# make sure that the main process calls "close" before as well
+shared_config.unlink()
+```
+
+## Limitations
+
+- Nested dictionaries are "currently" unsupported
+- Project is quite unstable, might not provide great performance boost at this time
+- macOS unsupported
+
+## Examples
+
+The `examples/` folder contains some code examples on how to use the package.
+
+## Building locally
+### Requirements
+
+- [`git`](https://git-scm.com/downloads)
+- [`uv`](https://docs.astral.sh/uv/getting-started/installation/)
+- Python >= 3.10
+- [`vcpkg`](https://vcpkg.io/en/)
+- `cython>=3.1`
+- A C++17 compatible compiler
+
+First [install and boostrap vcpkg](https://learn.microsoft.com/en-us/vcpkg/get_started/get-started-vscode?pivots=shell-cmd) somewhere in your system.
+
+```sh
+# for Windows it is reccomended to install it in C:\
+cd C:\
+
+git clone https://github.com/microsoft/vcpkg.git
+cd vcpkg && bootstrap-vcpkg.bat
+```
+
+Don't forget to add it to your `PATH`.
+
+Also, add an environment variable pointing to the root of `vcpkg`:
+
+```sh
+set "VCPKG_ROOT=C:\path\to\vcpkg"
+```
+
+Clone this repository, then install `boost::interprocess`:
+
+```
+vcpkg install boost-interprocess
+```
+
+Finally, through `uv`, install the package in a virtual environment:
+
+```sh
+uv venv --python 3.10
+uv pip install -e .[dev]
+```
+
+## License
+
+Licensed under [Apache 2.0](./LICENSE)

sharedbox-0.1.0.dist-info/RECORD

@@ -0,0 +1,12 @@
+sharedbox/__init__.py,sha256=eXnGQaG6CewxDJ7sGZjKbQ8zRXYzllcwDnx8HK1lBQY,65
+sharedbox/_shareddict.cp313-win_amd64.pyd,sha256=n2sH_RPpyDWr9rmzMZJHrSIj96jHhe3-DpN77AD04TU,227328
+sharedbox/_shareddict.pyi,sha256=bnSxe1V2H28o4OmKYKUmWM7aOLt3PfeyN10swaixTiY,1029
+sharedbox/_shareddict.pyx,sha256=hrU1MqcozLBooxTR5gLoju-h2dvBVuYOaCcKv0sNQtA,13879
+sharedbox/_version.py,sha256=7VAwr_Y4I30gkfUzxKxH2NbnLcgKYS2eJ7xUHD2f5t4,746
+sharedbox/utils.cp313-win_amd64.pyd,sha256=5TeGVW-K8yPibdcuIBktD_own90f7JZV5G74kzhhVPQ,81408
+sharedbox/utils.py,sha256=9sTzgKYMj4dj_Ks2Q1dbTCWDKxeD-aNAkDog6CI_aNc,10893
+sharedbox-0.1.0.dist-info/licenses/LICENSE,sha256=lj5YNz2g66TL33fWs0aLU5qLgALPnOvnb0xVk6whWbA,11546
+sharedbox-0.1.0.dist-info/METADATA,sha256=LCTMS_3Wooxzx0aXsxl1E3Y3tucZdu0kL5ZxANhHO64,4239
+sharedbox-0.1.0.dist-info/WHEEL,sha256=qV0EIPljj1XC_vuSatRWjn02nZIz3N1t8jsZz7HBr2U,101
+sharedbox-0.1.0.dist-info/top_level.txt,sha256=efvcA1LLG6nAoGlbTnKFlWWtIau3MXustUMsATPQMZE,10
+sharedbox-0.1.0.dist-info/RECORD,,

sharedbox-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.9.0)
+Root-Is-Purelib: false
+Tag: cp313-cp313-win_amd64
+

sharedbox-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [2025] [Jacopo Abramo]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

sharedbox-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+sharedbox