beaver-db 0.1.0__py3-none-any.whl → 0.2.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of beaver-db might be problematic. Click here for more details.

beaver/core.py CHANGED
@@ -2,162 +2,386 @@ import asyncio
2
2
  import json
3
3
  import sqlite3
4
4
  import time
5
- from typing import Any, AsyncIterator
5
+ from typing import Any, AsyncIterator, Union
6
6
 
7
- # --- SQL Schema ---
8
- # These statements are executed once to set up the database.
9
7
 
10
- PUBSUB_TABLE_SQL = """
11
- CREATE TABLE IF NOT EXISTS _beaver_pubsub_log (
12
- timestamp REAL PRIMARY KEY,
13
- channel_name TEXT NOT NULL,
14
- message_payload TEXT NOT NULL
15
- );
16
- """
8
+ class BeaverDB:
9
+ """
10
+ An embedded, multi-modal database in a single SQLite file.
11
+ Currently supports async pub/sub and a synchronous key-value store.
12
+ """
17
13
 
18
- PUBSUB_INDEX_SQL = """
19
- CREATE INDEX IF NOT EXISTS _beaver_idx_pubsub_channel_timestamp
20
- ON _beaver_pubsub_log (channel_name, timestamp);
21
- """
14
+ def __init__(self, db_path: str):
15
+ """
16
+ Initializes the database connection and creates necessary tables.
22
17
 
18
+ Args:
19
+ db_path: The path to the SQLite database file.
20
+ """
21
+ self._db_path = db_path
22
+ # Enable WAL mode for better concurrency between readers and writers
23
+ self._conn = sqlite3.connect(self._db_path, check_same_thread=False)
24
+ self._conn.execute("PRAGMA journal_mode=WAL;")
25
+ self._conn.row_factory = sqlite3.Row
26
+ self._create_pubsub_table()
27
+ self._create_kv_table()
28
+ self._create_list_table()
23
29
 
24
- class Subscriber(AsyncIterator):
25
- """
26
- A stateful, async context manager for receiving messages from a channel.
30
+ def _create_pubsub_table(self):
31
+ """Creates the pub/sub log table if it doesn't exist."""
32
+ with self._conn:
33
+ self._conn.execute("""
34
+ CREATE TABLE IF NOT EXISTS beaver_pubsub_log (
35
+ timestamp REAL PRIMARY KEY,
36
+ channel_name TEXT NOT NULL,
37
+ message_payload TEXT NOT NULL
38
+ )
39
+ """)
40
+ self._conn.execute("""
41
+ CREATE INDEX IF NOT EXISTS idx_pubsub_channel_timestamp
42
+ ON beaver_pubsub_log (channel_name, timestamp)
43
+ """)
27
44
 
28
- This object is returned by `BeaverDB.subscribe()` and is designed to
29
- be used with `async with` and `async for`.
30
- """
31
- def __init__(self, db_path: str, channel_name: str, poll_interval: float = 0.1):
32
- self._db_path = db_path
33
- self._channel = channel_name
34
- self._poll_interval = poll_interval
35
- self._queue = asyncio.Queue()
36
- self._last_seen_timestamp = time.time()
37
- self._polling_task = None
45
+ def _create_kv_table(self):
46
+ """Creates the key-value store table if it doesn't exist."""
47
+ with self._conn:
48
+ self._conn.execute("""
49
+ CREATE TABLE IF NOT EXISTS _beaver_kv_store (
50
+ key TEXT PRIMARY KEY,
51
+ value TEXT NOT NULL
52
+ )
53
+ """)
38
54
 
39
- async def _poll_for_messages(self):
40
- """A background task that polls SQLite for new messages."""
41
- while True:
42
- try:
43
- # Run the synchronous DB query in a thread to avoid blocking asyncio
44
- new_messages = await asyncio.to_thread(
45
- self._fetch_new_messages_from_db
55
+ def _create_list_table(self):
56
+ """Creates the lists table if it doesn't exist."""
57
+ with self._conn:
58
+ self._conn.execute("""
59
+ CREATE TABLE IF NOT EXISTS beaver_lists (
60
+ list_name TEXT NOT NULL,
61
+ item_order REAL NOT NULL,
62
+ item_value TEXT NOT NULL,
63
+ PRIMARY KEY (list_name, item_order)
46
64
  )
65
+ """)
47
66
 
48
- if new_messages:
49
- for timestamp, payload_str in new_messages:
50
- payload = json.loads(payload_str)
51
- await self._queue.put(payload)
52
- self._last_seen_timestamp = timestamp
67
+ def close(self):
68
+ """Closes the database connection."""
69
+ if self._conn:
70
+ self._conn.close()
53
71
 
54
- await asyncio.sleep(self._poll_interval)
55
- except asyncio.CancelledError:
56
- # Gracefully exit when the task is cancelled
57
- break
58
- except Exception as e:
59
- # In a real app, you'd add more robust logging/error handling
60
- print(f"ERROR: Pub/sub polling task failed: {e}")
61
- await asyncio.sleep(self._poll_interval * 5) # Back off on error
62
-
63
- def _fetch_new_messages_from_db(self) -> list[tuple[float, str]]:
64
- """The actual synchronous database query using sqlite3."""
65
- # Each call in a separate thread gets its own connection object.
66
- with sqlite3.connect(self._db_path) as conn:
67
- cursor = conn.cursor()
68
- cursor.execute(
69
- "SELECT timestamp, message_payload FROM _beaver_pubsub_log "
70
- "WHERE channel_name = ? AND timestamp > ? "
71
- "ORDER BY timestamp ASC",
72
- (self._channel, self._last_seen_timestamp)
72
+ # --- Key-Value Store Methods ---
73
+
74
+ def set(self, key: str, value: Any):
75
+ """
76
+ Stores a JSON-serializable value for a given key.
77
+ This operation is synchronous.
78
+
79
+ Args:
80
+ key: The unique string identifier for the value.
81
+ value: A JSON-serializable Python object (dict, list, str, int, etc.).
82
+
83
+ Raises:
84
+ TypeError: If the key is not a string or the value is not JSON-serializable.
85
+ """
86
+ if not isinstance(key, str):
87
+ raise TypeError("Key must be a string.")
88
+
89
+ try:
90
+ json_value = json.dumps(value)
91
+ except TypeError as e:
92
+ raise TypeError("Value must be JSON-serializable.") from e
93
+
94
+ with self._conn:
95
+ self._conn.execute(
96
+ "INSERT OR REPLACE INTO _beaver_kv_store (key, value) VALUES (?, ?)",
97
+ (key, json_value)
73
98
  )
74
- return cursor.fetchall()
75
99
 
76
- async def __aenter__(self):
77
- """Starts the background task when entering an 'async with' block."""
78
- if not self._polling_task:
79
- self._polling_task = asyncio.create_task(self._poll_for_messages())
80
- return self
100
+ def get(self, key: str) -> Any:
101
+ """
102
+ Retrieves a value for a given key.
103
+ This operation is synchronous.
81
104
 
82
- async def __aexit__(self, exc_type, exc_val, exc_tb):
83
- """Stops the background task when exiting the 'async with' block."""
84
- if self._polling_task:
85
- self._polling_task.cancel()
86
- await asyncio.gather(self._polling_task, return_exceptions=True)
105
+ Args:
106
+ key: The string identifier for the value.
87
107
 
88
- def __aiter__(self):
89
- return self
108
+ Returns:
109
+ The deserialized Python object, or None if the key is not found.
90
110
 
91
- async def __anext__(self) -> Any:
92
- """Allows 'async for' to pull messages from the internal queue."""
93
- return await self._queue.get()
111
+ Raises:
112
+ TypeError: If the key is not a string.
113
+ """
114
+ if not isinstance(key, str):
115
+ raise TypeError("Key must be a string.")
94
116
 
117
+ cursor = self._conn.cursor()
118
+ cursor.execute("SELECT value FROM _beaver_kv_store WHERE key = ?", (key,))
119
+ result = cursor.fetchone()
120
+ cursor.close()
95
121
 
96
- class BeaverDB:
97
- """
98
- A single-file, multi-modal database for Python.
122
+ if result:
123
+ return json.loads(result['value'])
124
+ return None
99
125
 
100
- Currently provides asynchronous pub/sub functionality using the standard
101
- sqlite3 library.
102
- """
103
- def __init__(self, db_path: str = "beaver.db"):
126
+ # --- List Methods ---
127
+
128
+ def list(self, name: str) -> "ListWrapper":
104
129
  """
105
- Initializes the database.
130
+ Returns a wrapper object for interacting with a specific list.
106
131
 
107
132
  Args:
108
- db_path: The path to the SQLite database file.
133
+ name: The name of the list.
134
+
135
+ Returns:
136
+ A ListWrapper instance bound to the given list name.
109
137
  """
110
- self.db_path = db_path
111
- self._setup_database()
138
+ if not isinstance(name, str) or not name:
139
+ raise TypeError("List name must be a non-empty string.")
140
+ return ListWrapper(name, self._conn)
112
141
 
113
- def _setup_database(self):
114
- """Creates the necessary tables and indices if they don't exist."""
115
- with sqlite3.connect(self.db_path) as conn:
116
- cursor = conn.cursor()
117
- cursor.execute(PUBSUB_TABLE_SQL)
118
- cursor.execute(PUBSUB_INDEX_SQL)
119
- conn.commit()
142
+ # --- Asynchronous Pub/Sub Methods ---
120
143
 
121
144
  async def publish(self, channel_name: str, payload: Any):
122
145
  """
123
146
  Publishes a JSON-serializable message to a channel.
124
-
125
- Args:
126
- channel_name: The name of the channel.
127
- payload: A JSON-serializable Python object (e.g., dict, list).
147
+ This operation is asynchronous.
128
148
  """
129
149
  if not isinstance(channel_name, str) or not channel_name:
130
150
  raise ValueError("Channel name must be a non-empty string.")
131
-
132
151
  try:
133
152
  json_payload = json.dumps(payload)
134
153
  except TypeError as e:
135
154
  raise TypeError("Message payload must be JSON-serializable.") from e
136
155
 
137
- # Run the blocking DB write in a thread to keep this method non-blocking
138
156
  await asyncio.to_thread(
139
157
  self._write_publish_to_db, channel_name, json_payload
140
158
  )
141
159
 
142
- def _write_publish_to_db(self, channel_name: str, json_payload: str):
143
- """The synchronous part of the publish operation using sqlite3."""
144
- with sqlite3.connect(self.db_path) as conn:
145
- cursor = conn.cursor()
146
- cursor.execute(
147
- "INSERT INTO _beaver_pubsub_log (timestamp, channel_name, message_payload) "
148
- "VALUES (?, ?, ?)",
160
+ def _write_publish_to_db(self, channel_name, json_payload):
161
+ """The synchronous part of the publish operation."""
162
+ with self._conn:
163
+ self._conn.execute(
164
+ "INSERT INTO beaver_pubsub_log (timestamp, channel_name, message_payload) VALUES (?, ?, ?)",
149
165
  (time.time(), channel_name, json_payload)
150
166
  )
151
- conn.commit()
152
167
 
153
- def subscribe(self, channel_name: str) -> Subscriber:
168
+ def subscribe(self, channel_name: str) -> "Subscriber":
154
169
  """
155
- Subscribes to a channel.
170
+ Subscribes to a channel, returning an async iterator.
171
+ """
172
+ return Subscriber(self._conn, channel_name)
156
173
 
157
- Returns a 'Subscriber' object that can be used in an 'async with'
158
- block and `async for` loop to receive messages.
159
174
 
160
- Args:
161
- channel_name: The name of the channel to subscribe to.
175
+ class ListWrapper:
176
+ """A wrapper providing a Pythonic interface to a list in the database."""
177
+
178
+ def __init__(self, name: str, conn: sqlite3.Connection):
179
+ self._name = name
180
+ self._conn = conn
181
+
182
+ def __len__(self) -> int:
183
+ """Returns the number of items in the list (e.g., `len(my_list)`)."""
184
+ cursor = self._conn.cursor()
185
+ cursor.execute("SELECT COUNT(*) FROM beaver_lists WHERE list_name = ?", (self._name,))
186
+ count = cursor.fetchone()[0]
187
+ cursor.close()
188
+ return count
189
+
190
+ def __getitem__(self, key: Union[int, slice]) -> Any:
162
191
  """
163
- return Subscriber(self.db_path, channel_name)
192
+ Retrieves an item or slice from the list (e.g., `my_list[0]`, `my_list[1:3]`).
193
+ """
194
+ if isinstance(key, slice):
195
+ start, stop, step = key.indices(len(self))
196
+ if step != 1:
197
+ raise ValueError("Slicing with a step is not supported.")
198
+
199
+ limit = stop - start
200
+ if limit <= 0:
201
+ return []
202
+
203
+ cursor = self._conn.cursor()
204
+ cursor.execute(
205
+ "SELECT item_value FROM beaver_lists WHERE list_name = ? ORDER BY item_order ASC LIMIT ? OFFSET ?",
206
+ (self._name, limit, start)
207
+ )
208
+ results = [json.loads(row['item_value']) for row in cursor.fetchall()]
209
+ cursor.close()
210
+ return results
211
+
212
+ elif isinstance(key, int):
213
+ list_len = len(self)
214
+ if key < -list_len or key >= list_len:
215
+ raise IndexError("List index out of range.")
216
+
217
+ offset = key if key >= 0 else list_len + key
218
+
219
+ cursor = self._conn.cursor()
220
+ cursor.execute(
221
+ "SELECT item_value FROM beaver_lists WHERE list_name = ? ORDER BY item_order ASC LIMIT 1 OFFSET ?",
222
+ (self._name, offset)
223
+ )
224
+ result = cursor.fetchone()
225
+ cursor.close()
226
+ return json.loads(result['item_value']) if result else None
227
+
228
+ else:
229
+ raise TypeError("List indices must be integers or slices.")
230
+
231
+ def _get_order_at_index(self, index: int) -> float:
232
+ """Helper to get the float `item_order` at a specific index."""
233
+ cursor = self._conn.cursor()
234
+ cursor.execute(
235
+ "SELECT item_order FROM beaver_lists WHERE list_name = ? ORDER BY item_order ASC LIMIT 1 OFFSET ?",
236
+ (self._name, index)
237
+ )
238
+ result = cursor.fetchone()
239
+ cursor.close()
240
+
241
+ if result:
242
+ return result[0]
243
+
244
+ raise IndexError(f"{index} out of range.")
245
+
246
+ def push(self, value: Any):
247
+ """Pushes an item to the end of the list."""
248
+ with self._conn:
249
+ cursor = self._conn.cursor()
250
+ cursor.execute("SELECT MAX(item_order) FROM beaver_lists WHERE list_name = ?", (self._name,))
251
+ max_order = cursor.fetchone()[0] or 0.0
252
+ new_order = max_order + 1.0
253
+
254
+ cursor.execute(
255
+ "INSERT INTO beaver_lists (list_name, item_order, item_value) VALUES (?, ?, ?)",
256
+ (self._name, new_order, json.dumps(value))
257
+ )
258
+
259
+ def prepend(self, value: Any):
260
+ """Prepends an item to the beginning of the list."""
261
+ with self._conn:
262
+ cursor = self._conn.cursor()
263
+ cursor.execute("SELECT MIN(item_order) FROM beaver_lists WHERE list_name = ?", (self._name,))
264
+ min_order = cursor.fetchone()[0] or 0.0
265
+ new_order = min_order - 1.0
266
+
267
+ cursor.execute(
268
+ "INSERT INTO beaver_lists (list_name, item_order, item_value) VALUES (?, ?, ?)",
269
+ (self._name, new_order, json.dumps(value))
270
+ )
271
+
272
+ def insert(self, index: int, value: Any):
273
+ """Inserts an item at a specific index."""
274
+ list_len = len(self)
275
+ if index <= 0:
276
+ self.prepend(value)
277
+ return
278
+ if index >= list_len:
279
+ self.push(value)
280
+ return
281
+
282
+ # Midpoint insertion
283
+ order_before = self._get_order_at_index(index - 1)
284
+ order_after = self._get_order_at_index(index)
285
+
286
+ new_order = order_before + (order_after - order_before) / 2.0
287
+
288
+ with self._conn:
289
+ self._conn.execute(
290
+ "INSERT INTO beaver_lists (list_name, item_order, item_value) VALUES (?, ?, ?)",
291
+ (self._name, new_order, json.dumps(value))
292
+ )
293
+
294
+ def pop(self) -> Any:
295
+ """Removes and returns the last item from the list."""
296
+ with self._conn:
297
+ cursor = self._conn.cursor()
298
+ cursor.execute(
299
+ "SELECT rowid, item_value FROM beaver_lists WHERE list_name = ? ORDER BY item_order DESC LIMIT 1",
300
+ (self._name,)
301
+ )
302
+ result = cursor.fetchone()
303
+ if not result:
304
+ return None
305
+
306
+ rowid_to_delete, value_to_return = result
307
+ cursor.execute("DELETE FROM beaver_lists WHERE rowid = ?", (rowid_to_delete,))
308
+ return json.loads(value_to_return)
309
+
310
+ def deque(self) -> Any:
311
+ """Removes and returns the first item from the list."""
312
+ with self._conn:
313
+ cursor = self._conn.cursor()
314
+ cursor.execute(
315
+ "SELECT rowid, item_value FROM beaver_lists WHERE list_name = ? ORDER BY item_order ASC LIMIT 1",
316
+ (self._name,)
317
+ )
318
+ result = cursor.fetchone()
319
+ if not result:
320
+ return None
321
+
322
+ rowid_to_delete, value_to_return = result
323
+ cursor.execute("DELETE FROM beaver_lists WHERE rowid = ?", (rowid_to_delete,))
324
+ return json.loads(value_to_return)
325
+
326
+
327
+ class Subscriber(AsyncIterator):
328
+ """
329
+ An async iterator that polls a channel for new messages.
330
+ Designed to be used with 'async with'.
331
+ """
332
+
333
+ def __init__(self, conn: sqlite3.Connection, channel_name: str, poll_interval: float = 0.1):
334
+ self._conn = conn
335
+ self._channel = channel_name
336
+ self._poll_interval = poll_interval
337
+ self._queue = asyncio.Queue()
338
+ self._last_seen_timestamp = time.time()
339
+ self._polling_task = None
340
+
341
+ async def _poll_for_messages(self):
342
+ """Background task that polls the database for new messages."""
343
+ while True:
344
+ try:
345
+ new_messages = await asyncio.to_thread(
346
+ self._fetch_new_messages_from_db
347
+ )
348
+ if new_messages:
349
+ for msg in new_messages:
350
+ payload = json.loads(msg["message_payload"])
351
+ await self._queue.put(payload)
352
+ self._last_seen_timestamp = msg["timestamp"]
353
+ await asyncio.sleep(self._poll_interval)
354
+ except asyncio.CancelledError:
355
+ break
356
+ except Exception:
357
+ # In a real app, add more robust error logging
358
+ await asyncio.sleep(self._poll_interval * 5)
359
+
360
+ def _fetch_new_messages_from_db(self) -> list:
361
+ """The actual synchronous database query."""
362
+ cursor = self._conn.cursor()
363
+ cursor.execute(
364
+ "SELECT timestamp, message_payload FROM beaver_pubsub_log WHERE channel_name = ? AND timestamp > ? ORDER BY timestamp ASC",
365
+ (self._channel, self._last_seen_timestamp)
366
+ )
367
+ results = cursor.fetchall()
368
+ cursor.close()
369
+ return results
370
+
371
+ async def __aenter__(self):
372
+ """Starts the background task."""
373
+ self._polling_task = asyncio.create_task(self._poll_for_messages())
374
+ return self
375
+
376
+ async def __aexit__(self, exc_type, exc_val, exc_tb):
377
+ """Stops the background task."""
378
+ if self._polling_task:
379
+ self._polling_task.cancel()
380
+ await asyncio.gather(self._polling_task, return_exceptions=True)
381
+
382
+ def __aiter__(self):
383
+ return self
384
+
385
+ async def __anext__(self) -> Any:
386
+ """Allows 'async for' to pull messages from the internal queue."""
387
+ return await self._queue.get()
@@ -0,0 +1,109 @@
1
+ Metadata-Version: 2.4
2
+ Name: beaver-db
3
+ Version: 0.2.0
4
+ Summary: Asynchronous, embedded, modern DB based on SQLite.
5
+ Requires-Python: >=3.13
6
+ Description-Content-Type: text/markdown
7
+
8
+ # beaver 🦫
9
+
10
+ A fast, single-file, multi-modal database for Python, built with the standard sqlite3 library.
11
+
12
+ `beaver` is the Backend for Embedded Asynchronous Vector & Event Retrieval. It's an industrious, all-in-one database designed to manage complex, modern data types without requiring a database server.
13
+
14
+ Design Philosophy
15
+ `beaver` is built with a minimalistic philosophy for small, local use cases where a full-blown database server would be overkill.
16
+
17
+ - **Minimalistic & Zero-Dependency**: Uses only Python's standard libraries (sqlite3, asyncio). No external packages are required, making it incredibly lightweight and portable.
18
+ - **Async-First (When It Matters)**: The pub/sub system is fully asynchronous for high-performance, real-time messaging. Simpler features like key-value and list operations remain synchronous for ease of use.
19
+ - **Built for Local Applications**: Perfect for local AI tools, chatbots (streaming tokens), task management apps, desktop utilities, and prototypes that need persistent, structured data without network overhead.
20
+ - **Fast by Default**: It's built on SQLite, which is famously fast, reliable, and will likely serve your needs for a long way before you need a "professional" database.
21
+
22
+ ## Core Features
23
+
24
+ - **Asynchronous Pub/Sub**: A fully asynchronous, Redis-like publish-subscribe system for real-time messaging.
25
+ - **Persistent Key-Value Store**: A simple set/get interface for storing configuration, session data, or any other JSON-serializable object.
26
+ - **Pythonic List Management**: A fluent, Redis-like interface (db.list("name").push()) for managing persistent, ordered lists with support for indexing and slicing.
27
+ - **Single-File & Portable**: All data is stored in a single SQLite file, making it incredibly easy to move, back up, or embed in your application.
28
+
29
+ ## Installation
30
+
31
+ ```bash
32
+ pip install beaver-db
33
+ ```
34
+
35
+ ## Quickstart & API Guide
36
+
37
+ ### 1. Initialization
38
+
39
+ All you need to do is import and instantiate the BeaverDB class with a file path.
40
+
41
+ ```python
42
+ from beaver import BeaverDB
43
+
44
+ db = BeaverDB("my_application.db")
45
+ ```
46
+
47
+ ### 2. Key-Value Store
48
+
49
+ Use `set()` and `get()` for simple data storage. The value can be any JSON-encodable object.
50
+
51
+ ```python
52
+ # Set a value
53
+ db.set("app_config", {"theme": "dark", "user_id": 123})
54
+
55
+ # Get a value
56
+ config = db.get("app_config")
57
+ print(f"Theme: {config['theme']}") # Output: Theme: dark
58
+ ```
59
+
60
+ ### 3. List Management
61
+
62
+ Get a list wrapper with `db.list()` and use Pythonic methods to manage it.
63
+
64
+ ```python
65
+ # Get a wrapper for the 'tasks' list
66
+ tasks = db.list("daily_tasks")
67
+
68
+ # Push items to the list
69
+ tasks.push("Write the project report")
70
+ tasks.push("Send follow-up emails")
71
+ tasks.prepend("Plan the day's agenda") # Push to the front
72
+
73
+ # Use len() and indexing (including slices!)
74
+ print(f"There are {len(tasks)} tasks.")
75
+ print(f"The first task is: {tasks[0]}")
76
+ print(f"The rest is: {tasks[1:]}")
77
+ ```
78
+
79
+ ### 4. Asynchronous Pub/Sub
80
+
81
+ Publish events from one part of your app and listen in another using asyncio.
82
+
83
+ ```python
84
+ import asyncio
85
+
86
+ async def listener():
87
+ async with db.subscribe("system_events") as sub:
88
+ async for message in sub:
89
+ print(f"LISTENER: Received event -> {message['event']}")
90
+
91
+ async def publisher():
92
+ await asyncio.sleep(1)
93
+ await db.publish("system_events", {"event": "user_login", "user": "alice"})
94
+
95
+ # To run them concurrently:
96
+ # asyncio.run(asyncio.gather(listener(), publisher()))
97
+ ```
98
+
99
+ ## Roadmap
100
+
101
+ `beaver` aims to be a complete, self-contained data toolkit. The following features are planned:
102
+
103
+ - **Vector Storage & Search**: Store NumPy vector embeddings and perform efficient k-nearest neighbor (k-NN) searches using `scipy.spatial.cKDTree`.
104
+ - **JSON Document Store with Full-Text Search**: Store flexible JSON documents and get powerful full-text search across all text fields, powered by SQLite's FTS5 extension.
105
+ - **Standard Relational Interface**: While `beaver` provides high-level features, you can always use the same SQLite file for normal relational tasks (e.g., managing users, products) with standard SQL.
106
+
107
+ ## License
108
+
109
+ This project is licensed under the MIT License.
@@ -0,0 +1,6 @@
1
+ beaver/__init__.py,sha256=pE1JdpHVni2Hv6igs5VrKPlHkkKMik3ZiwhR23KRBkk,38
2
+ beaver/core.py,sha256=NQizE87jkR7DwbcZoIX61rdVe7z6bAarBUbo-oUc8SI,13720
3
+ beaver_db-0.2.0.dist-info/METADATA,sha256=DGdk2Il2CFRtfM2BzY7Z2dYUxy1UQeKdVKnIVa_5wzg,4295
4
+ beaver_db-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
5
+ beaver_db-0.2.0.dist-info/top_level.txt,sha256=FxA4XnX5Qm5VudEXCduFriqi4dQmDWpQ64d7g69VQKI,7
6
+ beaver_db-0.2.0.dist-info/RECORD,,
@@ -1,117 +0,0 @@
1
- Metadata-Version: 2.4
2
- Name: beaver-db
3
- Version: 0.1.0
4
- Summary: Asynchronous, embedded, modern DB based on SQLite.
5
- Requires-Python: >=3.13
6
- Description-Content-Type: text/markdown
7
-
8
- # beaver 🦫
9
-
10
- A single-file, multi-modal database for Python, built with the standard sqlite3 library.
11
-
12
- `beaver` is the **B**ackend for **E**mbedded **A**synchronous **V**ector & **E**vent **R**etrieval. It's an industrious, all-in-one database designed to manage complex, modern data types without leaving the comfort of a single file.
13
-
14
- This project is currently in its initial phase, with the core asynchronous pub/sub functionality fully implemented.
15
-
16
- ## Core Features (Current)
17
-
18
- - **Zero Dependencies:** Built using only the standard Python `sqlite3` and `asyncio` libraries. No external packages to install.
19
- - **Async Pub/Sub:** A fully asynchronous, Redis-like publish-subscribe system for real-time messaging between components of your application.
20
- - **Single-File & Persistent:** All data is stored in a single SQLite file, making it incredibly portable and easy to back up. Your event log persists across application restarts.
21
- - **Works with Existing Databases:** `beaver` can be pointed at an existing SQLite file and will create its tables without disturbing other data.
22
-
23
- ## Use Cases
24
-
25
- I built `beaver` to have a local, embedded database for building small AI-powered projects without having to pay for a server-based database.
26
-
27
- Examples include:
28
-
29
- - Streaming messages and tokens from a local FastAPI to a local Streamlit app.
30
- - Storing user files for Retrieval Augmented Generation in single-user applications.
31
-
32
- ## Installation
33
-
34
- To use `beaver`, just run `pip install beaver-db` and import the main class.
35
-
36
- ```python
37
- import asyncio
38
- from beaver import BeaverDB
39
-
40
-
41
- # --- Example Usage ---
42
- async def listener(db: BeaverDB):
43
- """A sample task that listens for messages."""
44
- print("LISTENER: Waiting for messages on the 'system_events' channel...")
45
- try:
46
- async with db.subscribe("system_events") as subscriber:
47
- async for message in subscriber:
48
- print(f"LISTENER: Received -> {message}")
49
- except asyncio.CancelledError:
50
- print("LISTENER: Shutting down.")
51
-
52
-
53
- async def publisher(db: BeaverDB):
54
- """A sample task that publishes messages."""
55
- print("PUBLISHER: Ready to send events.")
56
- await asyncio.sleep(1) # Give the listener a moment to start
57
-
58
- print("PUBLISHER: Sending user login event.")
59
- await db.publish(
60
- "system_events",
61
- {"event": "user_login", "username": "alice", "status": "success"}
62
- )
63
-
64
- await asyncio.sleep(2)
65
-
66
- print("PUBLISHER: Sending system alert.")
67
- await db.publish(
68
- "system_events",
69
- {"event": "system_alert", "level": "warning", "detail": "CPU usage at 85%"}
70
- )
71
- await asyncio.sleep(1)
72
-
73
-
74
- async def main():
75
- """Runs the listener and publisher concurrently."""
76
- db = BeaverDB("demo.db")
77
-
78
- # Run both tasks and wait for them to complete
79
- listener_task = asyncio.create_task(listener(db))
80
- publisher_task = asyncio.create_task(publisher(db))
81
-
82
- await asyncio.sleep(5) # Let them run for a bit
83
- listener_task.cancel() # Cleanly shut down the listener
84
- await asyncio.gather(listener_task, publisher_task, return_exceptions=True)
85
- print("\nDemo finished.")
86
-
87
-
88
- if __name__ == "__main__":
89
- # To run this demo, save the file as beaver.py and run `python beaver.py`
90
- print("--- BeaverDB Pub/Sub Demo ---")
91
- asyncio.run(main())
92
- ```
93
-
94
- ## Roadmap
95
-
96
- `beaver` aims to be a complete, self-contained data toolkit for modern Python applications. The following features are planned for future releases, all accessible through a high-level API while still allowing direct SQL access:
97
-
98
- - **Vector Storage & Search:** Store numpy vector embeddings alongside your data and perform efficient k-nearest neighbor (k-NN) searches.
99
- - **Persistent Key-Value Store:** A simple get/set interface for storing configuration, session data, or any other JSON-serializable object.
100
- - **JSON Document Store with Full-Text Search:** Store flexible JSON documents and get powerful full-text search across all text fields by default, powered by SQLite's FTS5 extension.
101
- - **Standard Relational Interface:** While beaver provides high-level features, you will always be able to use the underlying SQLite connection for normal relational tasks, such as creating and managing users or products tables with standard SQL.
102
-
103
- ## Performance
104
-
105
- Despite its local, embedded nature, `beaver` is highly performant by small use cases. Here are some metrics, measured on a single laptop, Intel Core i7, 7th generation.
106
-
107
- - Process 100,000 messages (1000 messages times 100 asynchronous clients) in less than 30 seconds, giving over 3K messages per second with an average latency of only 100 ms (time elapsed between message generation and client processing).
108
-
109
- ## Why Beaver?
110
-
111
- Beavers are nature's engineers. They build a single, robust, and complex home—the lodge—from many different materials.
112
-
113
- Similarly, beaver builds a powerful, multi-modal database but contains it all within a single, self-contained file. It's an industrious, no-nonsense tool for building modern applications.
114
-
115
- ## License
116
-
117
- This project is licensed under the MIT License.
@@ -1,6 +0,0 @@
1
- beaver/__init__.py,sha256=pE1JdpHVni2Hv6igs5VrKPlHkkKMik3ZiwhR23KRBkk,38
2
- beaver/core.py,sha256=gWuuPwpT7yWgOulv_uwhOvJniltFO8K5-dfqjfA0jNk,5888
3
- beaver_db-0.1.0.dist-info/METADATA,sha256=b0gaV6IuXw2F_B4NOjKAP6Wj7dzHi7DjIYYg-0R_Q1Q,5243
4
- beaver_db-0.1.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
5
- beaver_db-0.1.0.dist-info/top_level.txt,sha256=FxA4XnX5Qm5VudEXCduFriqi4dQmDWpQ64d7g69VQKI,7
6
- beaver_db-0.1.0.dist-info/RECORD,,