tailcove 0.1.0__py3-none-any.whl

tailcove/__init__.py

@@ -0,0 +1,5 @@
+from .handler import TailcoveHandler
+from .transport import DeliveryError
+
+__all__ = ["TailcoveHandler", "DeliveryError"]
+__version__ = "0.1.0"

tailcove/context.py

@@ -0,0 +1,25 @@
+from __future__ import annotations
+
+import logging
+from typing import Any, Dict, FrozenSet
+
+# The stdlib `logging` module doesn't nest `extra={...}` under a key — each
+# entry lands as a top-level attribute directly on the LogRecord. To recover
+# "the extra dict" we diff a record's __dict__ against the attributes a
+# vanilla LogRecord always carries; whatever's left over is caller-supplied
+# context. Computed once from a throwaway record rather than hardcoded, so
+# it stays correct if a future Python version adds/removes a built-in field.
+_DUMMY_RECORD = logging.LogRecord(
+    name="tailcove",
+    level=logging.INFO,
+    pathname=__file__,
+    lineno=0,
+    msg="",
+    args=(),
+    exc_info=None,
+)
+RESERVED_ATTRS: FrozenSet[str] = frozenset(vars(_DUMMY_RECORD).keys()) | {"message", "asctime"}
+
+
+def extra_context(record: logging.LogRecord) -> Dict[str, Any]:
+    return {key: value for key, value in vars(record).items() if key not in RESERVED_ATTRS}

tailcove/exceptions.py

@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+import traceback
+from types import TracebackType
+from typing import Any, Dict, List, Optional, Tuple, Type
+
+ExcInfo = Tuple[Optional[Type[BaseException]], Optional[BaseException], Optional[TracebackType]]
+
+
+def format_exception(exc_info: ExcInfo, trace_limit: int = 10) -> Dict[str, Any]:
+    """
+    Renders a caught exception as structured data instead of a flat string,
+    matching the tailcove/laravel package's `{class, message, location,
+    trace}` shape so fingerprinting (Milestone 11) works identically
+    regardless of which client sent the line. `location` is the file:line
+    of the frame where the exception was actually raised (the last frame in
+    the traceback); `trace` is trimmed to `trace_limit` frames the same way
+    the Laravel package trims its stringified trace.
+    """
+    exc_type, exc_value, exc_tb = exc_info
+
+    frames = traceback.extract_tb(exc_tb)
+    location = f"{frames[-1].filename}:{frames[-1].lineno}" if frames else ""
+
+    trace: List[str] = [
+        f"{frame.filename}:{frame.lineno} in {frame.name}" for frame in frames
+    ][:trace_limit]
+
+    class_name = "UnknownException"
+    if exc_type is not None:
+        class_name = exc_type.__qualname__
+        if exc_type.__module__ not in ("builtins", "__main__"):
+            class_name = f"{exc_type.__module__}.{class_name}"
+
+    return {
+        "class": class_name,
+        "message": str(exc_value) if exc_value is not None else "",
+        "location": location,
+        "trace": trace,
+    }

tailcove/handler.py

@@ -0,0 +1,160 @@
+from __future__ import annotations
+
+import logging
+import queue
+import socket
+import sys
+import threading
+import time
+from datetime import datetime, timezone
+from typing import Any, Dict, List, Optional
+
+from .context import extra_context
+from .exceptions import format_exception
+from .levels import normalize_level
+from .transport import send_batch
+
+DEFAULT_ENDPOINT = "https://ingest.tailcove.dev/api/v1/logs"
+DEFAULT_BATCH_SIZE = 50
+DEFAULT_BATCH_SECONDS = 2.0
+DEFAULT_TIMEOUT = 2.0
+DEFAULT_TRACE_LIMIT = 10
+
+# Queue item kinds. A plain string tag plus payload, rather than subclasses,
+# keeps the background loop's dispatch a single if/elif chain.
+_LOG = "log"
+_FLUSH = "flush"
+_CLOSE = "close"
+
+
+class TailcoveHandler(logging.Handler):
+    """
+    Batches records and ships them to a Tailcove source's ingest endpoint on
+    a background thread — up to `batch_size` lines or `batch_seconds`,
+    whichever comes first, matching the wire contract and thresholds of the
+    tailcove/laravel and @tailcove/node clients (see the log-ingestion
+    skill). Never raises into the host app: `emit()` catches its own
+    failures and reports them via `self.handleError()`, the stdlib
+    `Handler` convention; delivery failures on the background thread are
+    caught and printed to stderr rather than propagated anywhere.
+
+    Flushing at interpreter exit is handled by the stdlib itself: every
+    `logging.Handler` is registered in `logging._handlerList`, and the
+    `logging` module registers its own `atexit` hook (`logging.shutdown`)
+    that calls `flush()` then `close()` on every handler. Subclassing
+    `Handler` and implementing `close()` to drain the queue is therefore
+    sufficient — no separate `atexit.register()` call is needed here, and
+    adding one would risk a double flush racing the stdlib's own.
+    """
+
+    def __init__(
+        self,
+        token: str,
+        endpoint: str = DEFAULT_ENDPOINT,
+        batch_size: int = DEFAULT_BATCH_SIZE,
+        batch_seconds: float = DEFAULT_BATCH_SECONDS,
+        timeout: float = DEFAULT_TIMEOUT,
+        trace_limit: int = DEFAULT_TRACE_LIMIT,
+        level: int = logging.NOTSET,
+    ) -> None:
+        super().__init__(level=level)
+        self.token = token
+        self.endpoint = endpoint
+        self.batch_size = batch_size
+        self.batch_seconds = batch_seconds
+        self.timeout = timeout
+        self.trace_limit = trace_limit
+
+        self._hostname = socket.gethostname()
+        self._queue: "queue.Queue[Any]" = queue.Queue()
+        self._closed = False
+        self._thread = threading.Thread(
+            target=self._run, name="tailcove-sender", daemon=True
+        )
+        self._thread.start()
+
+    def emit(self, record: logging.LogRecord) -> None:
+        try:
+            self._queue.put_nowait((_LOG, self._normalize(record)))
+        except Exception:
+            self.handleError(record)
+
+    def flush(self) -> None:
+        """Sends whatever is buffered right now, regardless of size or age. Safe to call directly; blocks until the background thread confirms the flush."""
+        self._signal(_FLUSH)
+
+    def close(self) -> None:
+        if self._closed:
+            super().close()
+            return
+
+        self._closed = True
+        self._signal(_CLOSE)
+        self._thread.join(timeout=self.timeout + 1)
+        super().close()
+
+    def _signal(self, kind: str) -> None:
+        if not self._thread.is_alive():
+            return
+
+        done = threading.Event()
+        self._queue.put((kind, done))
+        done.wait(timeout=self.timeout + 1)
+
+    def _normalize(self, record: logging.LogRecord) -> Dict[str, Any]:
+        context = extra_context(record)
+
+        if record.exc_info:
+            context["exception"] = format_exception(record.exc_info, self.trace_limit)
+
+        return {
+            "message": record.getMessage(),
+            "level": normalize_level(record.levelname),
+            "context": context,
+            "host": self._hostname,
+            "ts": datetime.fromtimestamp(record.created, tz=timezone.utc).isoformat(
+                timespec="milliseconds"
+            ),
+        }
+
+    def _run(self) -> None:
+        batch: List[Dict[str, Any]] = []
+        deadline: Optional[float] = None
+
+        while True:
+            wait = None if deadline is None else max(0.0, deadline - time.monotonic())
+
+            try:
+                kind, payload = self._queue.get(timeout=wait)
+            except queue.Empty:
+                kind, payload = None, None
+
+            if kind == _LOG:
+                batch.append(payload)
+                if deadline is None:
+                    deadline = time.monotonic() + self.batch_seconds
+            elif kind == _FLUSH:
+                self._send(batch)
+                batch, deadline = [], None
+                payload.set()
+                continue
+            elif kind == _CLOSE:
+                self._send(batch)
+                payload.set()
+                return
+
+            if batch and (
+                len(batch) >= self.batch_size
+                or (deadline is not None and time.monotonic() >= deadline)
+            ):
+                self._send(batch)
+                batch, deadline = [], None
+
+    def _send(self, batch: List[Dict[str, Any]]) -> None:
+        if not batch:
+            return
+
+        try:
+            send_batch(self.endpoint, self.token, batch, self.timeout)
+        except Exception as exc:
+            print(f"[tailcove] failed to send log batch: {exc}", file=sys.stderr)

tailcove/levels.py

@@ -0,0 +1,22 @@
+from __future__ import annotations
+
+#: Syslog-style levels, matching the set the ingest endpoint normalizes to
+#: (see the log-ingestion skill). Standard library level names map onto
+#: these directly; anything unrecognized (including a custom level name the
+#: server has never heard of) falls back to "info", same as the server.
+_LEVEL_MAP = {
+    "DEBUG": "debug",
+    "INFO": "info",
+    "NOTICE": "notice",
+    "WARNING": "warning",
+    "WARN": "warning",
+    "ERROR": "error",
+    "CRITICAL": "critical",
+    "FATAL": "critical",
+    "ALERT": "alert",
+    "EMERGENCY": "emergency",
+}
+
+
+def normalize_level(levelname: str) -> str:
+    return _LEVEL_MAP.get(levelname.upper(), "info")

tailcove/transport.py

@@ -0,0 +1,51 @@
+from __future__ import annotations
+
+import json
+import urllib.error
+import urllib.request
+from typing import Any, Dict, List
+
+
+class DeliveryError(Exception):
+    """Raised when a batch fails to reach the ingest endpoint (network error or non-2xx response)."""
+
+
+def send_batch(
+    endpoint: str,
+    token: str,
+    batch: List[Dict[str, Any]],
+    timeout: float,
+) -> None:
+    """
+    Ships a batch as NDJSON — one JSON object per line, per the ingest
+    endpoint's payload sniffing (see the log-ingestion skill). Uses only
+    `urllib` from the standard library so the package has zero required
+    dependencies. Raises DeliveryError on a network failure or non-2xx
+    response; callers decide whether that's swallowed or surfaced.
+    """
+    body = "\n".join(json.dumps(line, default=str) for line in batch).encode("utf-8")
+
+    request = urllib.request.Request(
+        endpoint,
+        data=body,
+        method="POST",
+        headers={
+            "Content-Type": "application/x-ndjson",
+            "Authorization": f"Bearer {token}",
+        },
+    )
+
+    try:
+        with urllib.request.urlopen(request, timeout=timeout) as response:
+            status = response.status
+    except urllib.error.HTTPError as exc:
+        raise DeliveryError(
+            f"tailcove: ingest endpoint responded with status {exc.code}"
+        ) from exc
+    except urllib.error.URLError as exc:
+        raise DeliveryError(
+            f"tailcove: request to ingest endpoint failed ({exc.reason})"
+        ) from exc
+
+    if status >= 300:
+        raise DeliveryError(f"tailcove: ingest endpoint responded with status {status}")

tailcove-0.1.0.dist-info/METADATA

@@ -0,0 +1,201 @@
+Metadata-Version: 2.5
+Name: tailcove
+Version: 0.1.0
+Summary: Python client for shipping application logs to Tailcove — a stdlib logging.Handler that batches and ships NDJSON.
+Project-URL: Homepage, https://tailcove.dev
+Project-URL: Documentation, https://tailcove.dev/docs/sending-logs/python
+Project-URL: Source, https://github.com/tailcove/python
+Project-URL: Issues, https://github.com/tailcove/python/issues
+Author-email: Akcybex <developers@akcybex.com>
+License-Expression: MIT
+License-File: LICENSE
+Keywords: logging,monolog,ndjson,tailcove
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: System :: Logging
+Classifier: Typing :: Typed
+Requires-Python: >=3.9
+Provides-Extra: dev
+Requires-Dist: pytest>=7; extra == 'dev'
+Description-Content-Type: text/markdown
+
+<h1 align="center">tailcove</h1>
+
+<p align="center">
+  Send Python logs to <a href="https://tailcove.dev">Tailcove</a> with the standard <code>logging</code> module.<br>
+  One handler, no dependencies, and it never slows down your code.
+</p>
+
+<p align="center">
+  <a href="https://pypi.org/project/tailcove/"><img alt="PyPI version" src="https://img.shields.io/pypi/v/tailcove?color=2DD4BF&labelColor=0A0A0A"></a>
+  <a href="https://pypi.org/project/tailcove/"><img alt="Python versions" src="https://img.shields.io/pypi/pyversions/tailcove?color=2DD4BF&labelColor=0A0A0A"></a>
+  <a href="https://pypi.org/project/tailcove/"><img alt="PyPI downloads" src="https://img.shields.io/pypi/dm/tailcove?color=2DD4BF&labelColor=0A0A0A"></a>
+  <a href="LICENSE"><img alt="MIT license" src="https://img.shields.io/badge/license-MIT-2DD4BF?labelColor=0A0A0A"></a>
+</p>
+
+<p align="center">
+  <a href="https://tailcove.dev/docs/sending-logs/python">Documentation</a> ·
+  <a href="https://github.com/tailcove/python/issues">Report a bug</a> ·
+  <a href="https://app.tailcove.dev/register">Get started free</a>
+</p>
+
+---
+
+## Why use it
+
+- **Plain `logging`.** `TailcoveHandler` is a normal `logging.Handler`. Keep the loggers you already have.
+- **Never blocks.** Records are batched and sent from a background thread, so a request or a hot loop never waits on the network.
+- **No dependencies.** Delivery uses `urllib` from the standard library.
+- **Safe.** A failed delivery is caught and never raised into your code.
+- **Structured exceptions.** Log an exception and it arrives as data (class, message, location, trace), so identical errors group together.
+
+## Install
+
+```bash
+pip install tailcove
+```
+
+Requires Python 3.9 or newer.
+
+## Quick start
+
+**1. Get a token.** Create a source in your [Tailcove dashboard](https://app.tailcove.dev) and copy its ingest token. It is shown only once, so save it somewhere safe (an environment variable is ideal).
+
+**2. Add the handler.**
+
+```python
+import logging
+import os
+
+from tailcove import TailcoveHandler
+
+logger = logging.getLogger("myapp")
+logger.addHandler(TailcoveHandler(token=os.environ["TAILCOVE_TOKEN"]))
+logger.setLevel(logging.INFO)
+
+logger.info("user signed up", extra={"user_id": 42})
+logger.error("payment failed", extra={"order_id": 99, "reason": "card_declined"})
+```
+
+**3. Open Tailcove.** Your records appear in search and live tail within a couple of seconds.
+
+Buffered records are flushed when your program exits normally, because Python's own `logging.shutdown()` closes every handler.
+
+## Options
+
+| Argument | Default | What it does |
+|---|---|---|
+| `token` | — | Your source's ingest token. **Required.** |
+| `endpoint` | `https://ingest.tailcove.dev/api/v1/logs` | The ingest URL. Only change this if you self-host. |
+| `batch_size` | `50` | Send once this many records are buffered. |
+| `batch_seconds` | `2.0` | Send once this many seconds have passed since the first buffered record. |
+| `timeout` | `2.0` | Seconds a delivery attempt may take before giving up. |
+| `trace_limit` | `10` | Most stack frames kept when a logged exception is rendered. |
+| `level` | `logging.NOTSET` | Lowest level this handler processes, on top of the logger's own level. |
+
+Records are sent once `batch_size` is reached or `batch_seconds` has passed, whichever comes first. They are never sent one HTTP request per log call. To send whatever is buffered right now, call `handler.flush()`. It blocks until the background thread confirms.
+
+## Structured context
+
+Anything you pass in `extra={...}` is shipped as the line's context, with the same keys:
+
+```python
+logger.info("order placed", extra={"order_id": 99, "total_cents": 4200})
+```
+
+## Exceptions
+
+Log with `exc_info=True`, or use `logger.exception(...)` inside an `except` block, and the exception is sent as structured data and not as a flattened string:
+
+```python
+try:
+    charge_card(order)
+except CardDeclined:
+    logger.error("payment failed", exc_info=True)
+```
+
+```json
+"exception": {
+    "class": "billing.errors.CardDeclined",
+    "message": "insufficient funds",
+    "location": "/app/billing/charge.py:42",
+    "trace": [
+        "/app/billing/charge.py:42 in charge_card",
+        "/app/orders/service.py:17 in place_order"
+    ]
+}
+```
+
+`location` is where the exception was raised, and `trace` is limited to `trace_limit` frames. It's the same shape the Laravel, Node and browser libraries use, so the same error groups into one issue whichever app reported it.
+
+## Frameworks
+
+### Django
+
+Add the handler to `LOGGING` in `settings.py`:
+
+```python
+LOGGING = {
+    "version": 1,
+    "disable_existing_loggers": False,
+    "handlers": {
+        "tailcove": {
+            "class": "tailcove.TailcoveHandler",
+            "token": env("TAILCOVE_TOKEN"),
+        },
+    },
+    "root": {
+        "handlers": ["console", "tailcove"],
+        "level": "INFO",
+    },
+    "loggers": {
+        "django.request": {
+            "handlers": ["tailcove"],
+            "level": "ERROR",
+            "propagate": True,
+        },
+    },
+}
+```
+
+### Flask
+
+Attach the handler to the app logger, or to the root logger if you want every `logging.getLogger(__name__)` in the app to ship too:
+
+```python
+from flask import Flask
+from tailcove import TailcoveHandler
+
+app = Flask(__name__)
+app.logger.addHandler(TailcoveHandler(token=app.config["TAILCOVE_TOKEN"]))
+app.logger.setLevel("INFO")
+
+
+@app.errorhandler(Exception)
+def handle_exception(exc):
+    app.logger.error("unhandled exception", exc_info=exc)
+    raise
+```
+
+## When delivery fails
+
+Network errors and non-2xx responses never raise into your code. `emit()` follows the standard `Handler` convention and routes its own failures through `handleError()`. Errors on the background thread are caught and printed to stderr. Logging is best-effort by design.
+
+## Help and feedback
+
+- **Docs:** [tailcove.dev/docs/sending-logs/python](https://tailcove.dev/docs/sending-logs/python)
+- **Bugs and feature requests:** [open an issue](https://github.com/tailcove/python/issues/new/choose)
+- **Security problems:** please email **security@tailcove.dev** instead of opening a public issue.
+
+## Other Tailcove libraries
+
+[Laravel](https://github.com/tailcove/laravel) · [Node.js](https://github.com/tailcove/node) · [Browser](https://github.com/tailcove/browser)
+
+## License
+
+[MIT](LICENSE) © Akcybex

tailcove-0.1.0.dist-info/RECORD

@@ -0,0 +1,11 @@
+tailcove/__init__.py,sha256=M6jEyWQipvkW18ELB11Q4S_3DMJadjwqy5nD92aF_-w,144
+tailcove/context.py,sha256=ePdDScnCQeY7ngXjDJM-rcC0W-vvEUSyVSropf73DcA,964
+tailcove/exceptions.py,sha256=kjpZL4YUyoXpuD0WAH9LDFlAM6Q6kthVMXa61iyP8cc,1533
+tailcove/handler.py,sha256=669QGPvOvEl2mbqHbdNkLXCBH0vDUwpIGjbe5uFppqs,5503
+tailcove/levels.py,sha256=-y7b2vvz2klWQ7VnAkqWK1fcIhiQE6LMp8Sm5e0ZcVg,689
+tailcove/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+tailcove/transport.py,sha256=zySSxWSgNsnL4mVSCBNzPMIbcJlfZzZb45xWlY3kKdU,1631
+tailcove-0.1.0.dist-info/METADATA,sha256=1_BWAco6_iX7lkpcPIN2LqquIGESuWThMjUWW3SSxEc,7431
+tailcove-0.1.0.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
+tailcove-0.1.0.dist-info/licenses/LICENSE,sha256=mHEELvGtWbKsucOMiKAMnUeQMOZY3R4-IGGc1IEGUWE,1064
+tailcove-0.1.0.dist-info/RECORD,,

tailcove-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.32.4
+Root-Is-Purelib: true
+Tag: py3-none-any

tailcove-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Akcybex
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.