classifyre-cli 0.4.113__tar.gz → 0.4.114__tar.gz

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/.turbo/turbo-build.log

@@ -1,3 +1,3 @@
 $ uv sync
-Resolved 309 packages in 333ms
+Resolved 309 packages in 321ms
 Checked 51 packages in 1ms

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: classifyre-cli
-Version: 0.4.113
+Version: 0.4.114
 Summary: Classifyre CLI — scan and classify unstructured data sources
 License: MIT
 Keywords: data,ingestion,metadata,pii,secrets,unstructured

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@classifyre/cli",
-  "version": "0.4.113",
+  "version": "0.4.114",
   "private": true,
   "scripts": {
     "build": "uv sync",

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "classifyre-cli"
-version = "0.4.113"
+version = "0.4.114"
 description = "Classifyre CLI — scan and classify unstructured data sources"
 readme = "README.md"
 requires-python = ">=3.12"

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/src/outputs/rest.py

@@ -77,17 +77,32 @@ class _JitteredRetry(Retry):
 #   attempt 5 → ~16 s
 #   attempt 6 → ~32 s
 #   attempt 7 → ~60 s  (capped)
-#   attempt 8 → ~60 s  (capped)
-# Total extra wait: ~182 s (~3 min) — covers extended load spikes on a
-# single-node VPS before event-loop pressure drops.
+#   attempt 8+ → ~60 s (capped)
+#
+# Status retries are budgeted separately from — and much more generously than —
+# connect/read, because the two failures mean opposite things. A connect/read
+# failure is a broken request that will not fix itself by waiting much longer.
+# A 503 is the API *deliberately* shedding an idempotent batch it wants us to
+# send again (see AssetService.bulkIngest and CliBackpressureGuard); giving up
+# on it discards ingested work, which is the one outcome backpressure exists to
+# avoid. At 60 s per attempt past the ramp, _STATUS_RETRIES ≈ 30 minutes of
+# patience: the CLI stalls under sustained pressure but the data still lands.
+# Retry-After is honoured when the server sends it (urllib3 default), so the
+# API can shorten or lengthen this on its own.
+#
+# total=None so the status budget is not silently capped by a smaller overall
+# count — with total set, whichever limit is lowest wins.
 #
 # POST and PATCH are explicitly allowed: without this urllib3 only retries
 # idempotent methods (GET/HEAD) by default.
+_STATUS_RETRIES = 32
+
 _RETRY_POLICY = _JitteredRetry(
-    total=8,
+    total=None,
     connect=8,
     read=8,
-    status=8,
+    status=_STATUS_RETRIES,
+    other=8,
     backoff_factor=2,
     backoff_max=60,
     status_forcelist={408, 429, 502, 503, 504},

classifyre_cli-0.4.114/tests/test_rest_backpressure.py

@@ -0,0 +1,150 @@
+"""Backpressure contract between the CLI and the API.
+
+The API sheds ingestion with 503 when its heap is near the V8 ceiling
+(AssetService.bulkIngest / CliBackpressureGuard). That is only safe if the CLI
+treats a 503 as "hold, then send this again" rather than as a failure: bulk
+ingest is an idempotent upsert, so a retry costs a round trip, while giving up
+silently discards assets and findings that were already extracted.
+
+These tests pin that contract with a real socket server, because the thing that
+can break it is the urllib3 Retry configuration, not our own code — and a
+misconfiguration there (a `total` lower than `status`, a status code missing
+from the forcelist) fails open, quietly, in exactly the case that matters.
+"""
+
+from __future__ import annotations
+
+import json
+import threading
+from http.server import BaseHTTPRequestHandler, HTTPServer
+from typing import ClassVar
+
+import pytest
+import requests
+from requests.adapters import HTTPAdapter
+
+from src.outputs.rest import _RETRY_POLICY, _STATUS_RETRIES
+
+
+class _FlakyHandler(BaseHTTPRequestHandler):
+    """Answers 503 for the first `unavailable_responses` calls, then 200."""
+
+    unavailable_responses: ClassVar[int] = 0
+    received_bodies: ClassVar[list[dict]] = []
+    calls: ClassVar[int] = 0
+
+    def do_POST(self) -> None:
+        type(self).calls += 1
+        length = int(self.headers.get("Content-Length", 0))
+        body = self.rfile.read(length)
+
+        if type(self).calls <= type(self).unavailable_responses:
+            self.send_response(503)
+            # Keeps the test fast: the CLI honours Retry-After, so the server
+            # picks the schedule instead of the ~2/4/8/16s exponential ramp.
+            # Note "0" would NOT work — urllib3 treats a falsy Retry-After as
+            # absent and falls back to its own backoff.
+            self.send_header("Retry-After", "1")
+            self.send_header("Content-Length", "0")
+            self.end_headers()
+            return
+
+        type(self).received_bodies.append(json.loads(body))
+        payload = b'{"ok": true}'
+        self.send_response(200)
+        self.send_header("Content-Type", "application/json")
+        self.send_header("Content-Length", str(len(payload)))
+        self.end_headers()
+        self.wfile.write(payload)
+
+    def log_message(self, *args: object) -> None:
+        pass
+
+
+@pytest.fixture
+def flaky_server():
+    def _start(unavailable_responses: int):
+        _FlakyHandler.unavailable_responses = unavailable_responses
+        _FlakyHandler.received_bodies = []
+        _FlakyHandler.calls = 0
+        server = HTTPServer(("127.0.0.1", 0), _FlakyHandler)
+        thread = threading.Thread(target=server.serve_forever, daemon=True)
+        thread.start()
+        return server
+
+    servers: list[HTTPServer] = []
+
+    def _factory(unavailable_responses: int) -> HTTPServer:
+        server = _start(unavailable_responses)
+        servers.append(server)
+        return server
+
+    yield _factory
+
+    for server in servers:
+        server.shutdown()
+        server.server_close()
+
+
+def _session() -> requests.Session:
+    """A session wired exactly as RestOutput wires its own."""
+    session = requests.Session()
+    adapter = HTTPAdapter(max_retries=_RETRY_POLICY)
+    session.mount("http://", adapter)
+    session.mount("https://", adapter)
+    return session
+
+
+def test_batch_survives_sustained_backpressure(flaky_server) -> None:
+    """The batch still lands after more 503s than the old budget allowed.
+
+    Nine is chosen deliberately: the previous policy was `total=8`, so this
+    exact scenario used to end with the assets discarded.
+    """
+    server = flaky_server(9)
+    url = f"http://127.0.0.1:{server.server_port}/assets/bulk"
+    batch = {"assets": [{"unique_id": "a1"}, {"unique_id": "a2"}]}
+
+    response = _session().post(url, json=batch, timeout=10)
+
+    assert response.status_code == 200
+    # The point of the whole exercise: the data arrived, unchanged, exactly
+    # once as far as the server's successful handling is concerned.
+    assert _FlakyHandler.received_bodies == [batch]
+
+
+def test_status_budget_is_not_capped_by_total() -> None:
+    """`total` must not silently shorten the status budget.
+
+    urllib3 applies whichever limit is lowest, so a `total` of 8 would make a
+    `status` of 32 a lie — which is how the retry budget read as generous while
+    actually expiring after ~3 minutes. Asserted as configuration rather than
+    behaviour because waiting out 32 real backoffs would take ~28 minutes.
+    """
+    assert _RETRY_POLICY.total is None
+    assert _RETRY_POLICY.status == _STATUS_RETRIES
+    assert _STATUS_RETRIES >= 16
+
+    # The wall-clock patience that budget buys, at backoff_max per attempt
+    # once the ramp saturates.
+    patience_s = sum(
+        min(_RETRY_POLICY.backoff_factor * (2**i), _RETRY_POLICY.backoff_max)
+        for i in range(_RETRY_POLICY.status)
+    )
+    assert patience_s > 20 * 60
+
+
+def test_backpressure_codes_are_all_retried() -> None:
+    for status in (408, 429, 502, 503, 504):
+        assert status in _RETRY_POLICY.status_forcelist
+
+    # Ingest is POST; urllib3 retries only idempotent methods unless told
+    # otherwise, and bulk ingest is an upsert so replay is safe.
+    assert "POST" in _RETRY_POLICY.allowed_methods
+    assert _RETRY_POLICY.respect_retry_after_header is True
+
+
+def test_client_errors_are_not_retried() -> None:
+    """A rejected payload must fail fast, not be resent 32 times."""
+    for status in (400, 404, 422):
+        assert status not in _RETRY_POLICY.status_forcelist

{classifyre_cli-0.4.113 → classifyre_cli-0.4.114}/uv.lock

@@ -562,30 +562,30 @@ wheels = [
 
 [[package]]
 name = "boto3"
-version = "1.43.70"
+version = "1.43.71"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "botocore" },
     { name = "jmespath" },
     { name = "s3transfer" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f6/20/68434d996c767e860329d78f2aaf1d7456173ba45f8d6fb8f4a394b85667/boto3-1.43.70.tar.gz", hash = "sha256:4f1e16b9eebbad3f312bb6fbf7685d40abdd52a691485cb9f09e7df60d912b41", size = 112688, upload-time = "2026-08-12T19:23:35.485Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e0/ed/3220ca6396a7ce00a49aa6f43f20cae05ec6ac386e2b5d668adfad0b3eed/boto3-1.43.71.tar.gz", hash = "sha256:e3eddb6346ee23c895dd98e5c2cfc72cb5e5ed111aa6d4331f9e903a3e3bf4ee", size = 112630, upload-time = "2026-08-13T19:20:35.332Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6e/d1/c95cae121ef5b77ecec08cb8e045eeb2a2c8ec605a5734b6f24bfb3a6326/boto3-1.43.70-py3-none-any.whl", hash = "sha256:b9399cce51bec9552e61a6b85015c42dea68752e79c00a7005d84eaf7ee15fef", size = 140028, upload-time = "2026-08-12T19:23:33.804Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/16/7b54e7f5271756ab13beaaf6da03f62edcc3ee5ded651c2b17297e107040/boto3-1.43.71-py3-none-any.whl", hash = "sha256:3bb46661dc33121e56e3706753069d18f8dbcb644599cf098206546338602e16", size = 140024, upload-time = "2026-08-13T19:20:32.542Z" },
 ]
 
 [[package]]
 name = "botocore"
-version = "1.43.70"
+version = "1.43.71"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jmespath" },
     { name = "python-dateutil" },
     { name = "urllib3" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2f/1a/ad4ff28cd0caa8ff4f39f18491169dc9ff4b2e359589be1f7108048e1915/botocore-1.43.70.tar.gz", hash = "sha256:a7bebdd0b1c8265e72759a4f7f61e9e43d57968b2c2a5e1e7b350d0a1e3ce81c", size = 15936837, upload-time = "2026-08-12T19:23:30.58Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d6/c2/98b8ff6a21a9ed99b0699f468446f26a679443b2bca46fcd6aaa0cfbae75/botocore-1.43.71.tar.gz", hash = "sha256:2cde7a55aaac86a2e9a3bcd2ce3fad1a31a4711749e10f350af231e55ccef7ed", size = 15943777, upload-time = "2026-08-13T19:20:29.292Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/69/b2/c16b4e2e7ccab1f0887c22fbe2d9859ceb4d35fb637cce8a380d51b52b06/botocore-1.43.70-py3-none-any.whl", hash = "sha256:95bc23a6472f8876349ae334b526e66006f6cf3c87bf39d3e9f057dc8c2eb85f", size = 15628519, upload-time = "2026-08-12T19:23:27.38Z" },
+    { url = "https://files.pythonhosted.org/packages/81/2f/700d973e3f0a37a275900be1a389af5a4ab67dd9362d17782b17f1045f0b/botocore-1.43.71-py3-none-any.whl", hash = "sha256:7bc3d5ddaca6f68e2eef707714264789048a6f496d375d976c560674f0cd1d1b", size = 15635728, upload-time = "2026-08-13T19:20:26.302Z" },
 ]
 
 [[package]]
@@ -923,7 +923,7 @@ wheels = [
 
 [[package]]
 name = "classifyre-cli"
-version = "0.4.113"
+version = "0.4.114"
 source = { editable = "." }
 dependencies = [
     { name = "beautifulsoup4" },
@@ -1462,7 +1462,7 @@ youtube = [
 
 [[package]]
 name = "classifyre-schemas"
-version = "0.4.113"
+version = "0.4.114"
 source = { editable = "../../packages/schemas" }
 dependencies = [
     { name = "fastjsonschema" },
@@ -1847,14 +1847,14 @@ wheels = [
 
 [[package]]
 name = "docling"
-version = "2.119.0"
+version = "2.120.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "docling-slim", extra = ["standard"] },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/30/c2/f9fe1cc7780a41956c73e003773f002f9721671e9f76a2d19333e38f2e4c/docling-2.119.0.tar.gz", hash = "sha256:04b9aef29b9b94fc6e56fcd440824337d132a84074c7ba20b9cf055f220e72fb", size = 9016, upload-time = "2026-08-10T10:28:54.837Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/83/a3/537c6f0ee2208e61538579f3dcf3abf5df4a4d6448c5c28c8574b5291903/docling-2.120.1.tar.gz", hash = "sha256:8ff9773147e7ce3588d67594239fb3b79be5733adfbf306ba46f4310fb1f61f0", size = 9013, upload-time = "2026-08-14T11:51:21.073Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1f/ea/37626a6ee6d4b26c14701b678eae81f53615c01afaa68b6b2367c4bbb09c/docling-2.119.0-py3-none-any.whl", hash = "sha256:3ff6f1dbe9f53ac034aca03c39fae374b381dc1eda4ad85d27a458fc13fbc917", size = 5180, upload-time = "2026-08-10T10:28:53.773Z" },
+    { url = "https://files.pythonhosted.org/packages/16/90/ec73628602e3a49544f3c6a6a4fde02e0f6053d58e9b4279bc0839a55f8e/docling-2.120.1-py3-none-any.whl", hash = "sha256:04897c566285bf2df166ccae6acd285f1d41e2336755a89d05be515be2090bc3", size = 5179, upload-time = "2026-08-14T11:51:19.875Z" },
 ]
 
 [[package]]
@@ -1922,7 +1922,7 @@ wheels = [
 
 [[package]]
 name = "docling-parse"
-version = "7.12.1"
+version = "7.13.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "docling-core" },
@@ -1930,28 +1930,28 @@ dependencies = [
     { name = "pydantic" },
     { name = "pywin32", marker = "sys_platform == 'win32'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/82/00/ab56de324d844cf5862f46f5c95dc18c5ee52dbb83d794c274c8e520373d/docling_parse-7.12.1.tar.gz", hash = "sha256:25c8759b19292ad8cf31e1b133980c30f0fd8800b510c02c56a3af5dbb989232", size = 6828762, upload-time = "2026-08-12T08:36:48.18Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/ba/80/861612a7ee977ec27cf4966337cbfa9033e9a58378fbeb0449e732b6fe31/docling_parse-7.13.0.tar.gz", hash = "sha256:23575b21b8722466899135340b254124792d0c19d18cd223cad437dc4283f8f5", size = 6864422, upload-time = "2026-08-14T06:28:50.233Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ba/ff/4d4464e3e74dd54a47e0c811779a378f1f0b22b74adb2874bcc38885a3b5/docling_parse-7.12.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:ff5ac71bc2a702cbc7bd4fd7c2fcbf6af483c982af4bae36fb727a9db169a849", size = 9667688, upload-time = "2026-08-12T08:36:20.274Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/41/24b6303e0e5e79c3ea8d2daea0cdcbe95a3d71a0b1715bd5345ac7a3ba5a/docling_parse-7.12.1-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a3c91f46776ba2aa3c31667747f4f39e91f8c1b58e391663c5e44d7bb416a3f6", size = 10191724, upload-time = "2026-08-12T08:36:22.102Z" },
-    { url = "https://files.pythonhosted.org/packages/53/31/951fc8740f2ab6ab1fb9a1d223269d082eedf8a4b1dcbee42da81a395efc/docling_parse-7.12.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:33fc4afd8b46a2c760a8ca4558b4d744d6c9d207318581f3234fec3df91ca85f", size = 10577523, upload-time = "2026-08-12T08:36:23.821Z" },
-    { url = "https://files.pythonhosted.org/packages/37/b9/0c17dfa3571705f4bdd698d929c4003cfc249d117c7317ae82a294cc18a3/docling_parse-7.12.1-cp312-cp312-win_amd64.whl", hash = "sha256:344c7f1da55be7fc992045fb5ff07b411e589b1da62a8f943842ac7ac986245a", size = 11639307, upload-time = "2026-08-12T08:36:25.954Z" },
-    { url = "https://files.pythonhosted.org/packages/30/5b/5f589310d243b006d145c8e4cd0d9770accf2048ba6ad08d7608ac1f6e58/docling_parse-7.12.1-cp312-cp312-win_arm64.whl", hash = "sha256:cfddd1f3a3631e87d4c965b4ffd0dc95e249b046af0a6b8916e129285fb0e80c", size = 8964784, upload-time = "2026-08-12T08:36:27.94Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/35/510104ef0f0f01e985fb4eeb9269aaec964e80dda4b64077d19099923782/docling_parse-7.12.1-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:b3a740f373f00c87555cdbbf996b726df1e99a9de5d95a24484f71d1f7394077", size = 9667627, upload-time = "2026-08-12T08:36:29.713Z" },
-    { url = "https://files.pythonhosted.org/packages/48/7b/6bab4aba25c7356f6afff8e067e6e2ced71767db7f0b1698537e27ff8853/docling_parse-7.12.1-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:61a2f203a8e4ce5cd27fba2b6de4474eb11b5e6e1e17116e0dd349376d234d6d", size = 10191624, upload-time = "2026-08-12T08:36:31.408Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/17/4cb450252bb8a38ffb98aa7f5f42bd35d300808d42e46a96849890993e84/docling_parse-7.12.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d5c015e19e9be88ce001115a1913bf6addf4c2931c3556e9cf2a10b596bbe46d", size = 10577661, upload-time = "2026-08-12T08:36:33.227Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/e2/8ca2efcd457a2a040e9dc1af73100c17d72c20cd7bf1b1d8b20e51abc771/docling_parse-7.12.1-cp313-cp313-win_amd64.whl", hash = "sha256:24b29a5ca40071b3d5ea7494c22a5552f4d9d29bf93bdfc1925986d922feff5f", size = 11639305, upload-time = "2026-08-12T08:36:35.055Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/4b/b59a91e8e63a855b54905aaba39c1f06d4f95cad5c0c160f9995a6257701/docling_parse-7.12.1-cp313-cp313-win_arm64.whl", hash = "sha256:4f73b41fda9c4fa6beb6c3d685071dd28a2c1f00e69f0133fbc15bda8c0f7ab7", size = 8964612, upload-time = "2026-08-12T08:36:36.772Z" },
-    { url = "https://files.pythonhosted.org/packages/04/57/97ec433850b8e7706943a8de0cf65774f043ae393a5d86703e7e8658d71e/docling_parse-7.12.1-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:447ced8771f7053f9f834efec363d578d92598119b95e83bed2551c185a7355f", size = 9668271, upload-time = "2026-08-12T08:36:38.58Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/ce/6c213b6599ca8b0c93a8cf11077f2c38b2ed48a2d05e42e3c091b5245642/docling_parse-7.12.1-cp314-cp314-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:32e4c806e63f5e6dcf63643d3472b241e7786653bad82321c95b937b69942a5e", size = 10192440, upload-time = "2026-08-12T08:36:40.529Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/a7/17cd8d4c691c025a9515bd0ab81c8de502988fa257ba2fdccdb973b5c53e/docling_parse-7.12.1-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fff030c85eb85fc6589bdabc64bdc58a0c82cd6e0870ff9cf55114be525f46f5", size = 10578105, upload-time = "2026-08-12T08:36:42.238Z" },
-    { url = "https://files.pythonhosted.org/packages/63/c5/0dfb8eb4ea12f7f361beb8c342c115774524f7cea0a1feeba273f572fb27/docling_parse-7.12.1-cp314-cp314-win_amd64.whl", hash = "sha256:8a3debe619e2303442139472ec39a21cdb4714b27d026cf0c34233411779c8c4", size = 12072065, upload-time = "2026-08-12T08:36:43.985Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/45/7fdcf5c321ab3d0c2047edd844c75633d9c60abc043083f6dab95a1b1b39/docling_parse-7.12.1-cp314-cp314-win_arm64.whl", hash = "sha256:6488966275412d599aa2fcd0920cea38bb20ac937aeb5bd135ea5354c6dcb5b6", size = 9329032, upload-time = "2026-08-12T08:36:46.533Z" },
+    { url = "https://files.pythonhosted.org/packages/68/95/45b9a5cfd04853bf2a8ebe54722096ee6969a19142cfa39f320dc3254322/docling_parse-7.13.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:21eb89636a9707c484f7d73dae0e56f64807869281716ccb511c55a5adf3121f", size = 9698711, upload-time = "2026-08-14T06:28:12.256Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/72/269de56bbb4eec9255d4f5b89a0fa0996b1fd697bfc15e3f959a53adc285/docling_parse-7.13.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4a1f7d03e74e1fd6e8f11bff03c47b1300e259ea1b1ffd733166001d237baec9", size = 10223985, upload-time = "2026-08-14T06:28:14.875Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/4a/45a7ba9b8694ab552613cd4d93df1d60905023857cfee1a99289aa3f89c3/docling_parse-7.13.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f68c5dffee3e0596e0f8ee4efdb30ee52bec59b913170eaa082d0cc2f287c259", size = 10614327, upload-time = "2026-08-14T06:28:17.228Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/f7/4f18b540c6dc1f513abf38e9c7f9d1f25f664fed9af6bc52cb099ef09a78/docling_parse-7.13.0-cp312-cp312-win_amd64.whl", hash = "sha256:f683f3ad7b428b18e236e0175dbff897619f9433d39aac6d16a859263ec4125b", size = 11673968, upload-time = "2026-08-14T06:28:19.568Z" },
+    { url = "https://files.pythonhosted.org/packages/35/7a/f203bacb44f1e4113894b26479261dc400137e5675e670e0ffe2d9d75adb/docling_parse-7.13.0-cp312-cp312-win_arm64.whl", hash = "sha256:a73f3ecc446aa8832396eeb49f64924304b7eb52c473ae9a610ec6b4a1b06e2f", size = 8992588, upload-time = "2026-08-14T06:28:23.477Z" },
+    { url = "https://files.pythonhosted.org/packages/68/a6/427efea60890d2962005039260a61ecba4cd23a0e2a83b116e7064bc1c62/docling_parse-7.13.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:9cc2cdf4aed0c83d690c733259c1a7a38e9690e2b42bae39b3f6ff1d289194ae", size = 9698752, upload-time = "2026-08-14T06:28:25.835Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/d9/16593f172dce343fb9b5fbb6711070dd898bd52d05afe7fb38182d0c2588/docling_parse-7.13.0-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:979978f903bb1b387429eb7e0cbc44eb877080c4f2129c8b637e24100bd75ff5", size = 10223995, upload-time = "2026-08-14T06:28:28.518Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/da/e4bc26a0345d0651779bfb04f5cd99552a3fac1fca2681d88b67a27348d9/docling_parse-7.13.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ac73eb49b6194d40d58f01eba6e134bea857667f6aa99dc2371cf3d24d5fd817", size = 10614354, upload-time = "2026-08-14T06:28:30.871Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/b8/a40158593b3293a8364d62cf2866a29e71f0d909f46c2413e4c98f127142/docling_parse-7.13.0-cp313-cp313-win_amd64.whl", hash = "sha256:f16952a2cc2a609f444bf0d6fe7ee1d9b35f1d38f26ebb7f208652a8c7999b71", size = 11673865, upload-time = "2026-08-14T06:28:33.354Z" },
+    { url = "https://files.pythonhosted.org/packages/13/af/f255fcd2d2b5987c2e3b6dae5d42283266720c775f32e6a1503b5c6683f2/docling_parse-7.13.0-cp313-cp313-win_arm64.whl", hash = "sha256:30c612b0378233d6dbc4b76a10a4fb1492c975a50121aa3a8f691b1dd6ba8298", size = 8992516, upload-time = "2026-08-14T06:28:35.64Z" },
+    { url = "https://files.pythonhosted.org/packages/43/8f/452ea0ccdc577d7cdab49b21a97463650cf29447b381dc79866032f9a87b/docling_parse-7.13.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:1404f4292303c53c8d7822957888edb3c2ffc9029f6289862150af09cd2263d1", size = 9699328, upload-time = "2026-08-14T06:28:38.109Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/ec/f23d35ccd988e8c3e2960bf9475713387e87fe1b56747fc5700f8e5e695b/docling_parse-7.13.0-cp314-cp314-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:764a4cfa366e430ded399d505e39f91cb1f7060dece48a83c49512b2acd57928", size = 10224649, upload-time = "2026-08-14T06:28:40.433Z" },
+    { url = "https://files.pythonhosted.org/packages/92/d2/063dcaa311702a2500e4d063f08e7134d048c000813235e5e085d3d4a1e1/docling_parse-7.13.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fbe720eab2ae5b38928872953acd4ac693c7c79ef340203285b7345dd4ba3a42", size = 10614598, upload-time = "2026-08-14T06:28:42.955Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/75/9c9bace0574c2aaf2b9332083446c563dd0356ca0651d74f2c41bbdd334e/docling_parse-7.13.0-cp314-cp314-win_amd64.whl", hash = "sha256:9ff2657c8950225a894558a0ff9c8198af3ed813c4e945e5c1995b8892c17dce", size = 12107442, upload-time = "2026-08-14T06:28:45.362Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/09/ba429e3ed8566d5dc00b72064c2d223595ecd6666cd3b3a5b781cd1d9297/docling_parse-7.13.0-cp314-cp314-win_arm64.whl", hash = "sha256:28c7881ee8fec08d119d661966f6cbf581b9da5896bddd4221732b6e17993121", size = 9359125, upload-time = "2026-08-14T06:28:47.667Z" },
 ]
 
 [[package]]
 name = "docling-slim"
-version = "2.119.0"
+version = "2.120.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "certifi" },
@@ -1963,9 +1963,9 @@ dependencies = [
     { name = "requests" },
     { name = "tqdm" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/08/8b/7eb40b83cb91ecd899faeaf421c81df20054053e3cc64cafbd6ce0c530cd/docling_slim-2.119.0.tar.gz", hash = "sha256:7b4ee3891e536403f07b6ab702bc757d6c3e4fe146475656d97cf911ac96db8f", size = 582133, upload-time = "2026-08-10T10:27:35.891Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/f0/5b/3bd25d1832168ce15b788d3e995c7c0e2aa4bb3dd226aaa6adce44ef4d66/docling_slim-2.120.1.tar.gz", hash = "sha256:451622619c08fa75cb9fc7a4deb3c2f165be33503d46226f4c63024a863be64c", size = 588136, upload-time = "2026-08-14T11:49:56.896Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/51/09/74f745d6178dbec12eb503261f87b0bb947fac5b122cea2e6d3c433ce7ae/docling_slim-2.119.0-py3-none-any.whl", hash = "sha256:ca385bc6b0ca99a4f0bba07feaa4c0fc4d5fa685a22184baa79de4bd87598323", size = 723087, upload-time = "2026-08-10T10:27:34.074Z" },
+    { url = "https://files.pythonhosted.org/packages/96/dd/9f5bad579bfd712075ea4a56b7e35036cbe5bc9c39a2b668003d15a6b8a5/docling_slim-2.120.1-py3-none-any.whl", hash = "sha256:97e0275ff06865df4b9a4162d01674401e4046d85e012b923f6eb8fc53441732", size = 729536, upload-time = "2026-08-14T11:49:54.741Z" },
 ]
 
 [package.optional-dependencies]
@@ -4366,11 +4366,11 @@ wheels = [
 
 [[package]]
 name = "platformdirs"
-version = "4.11.2"
+version = "4.11.3"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e5/98/0bf930c4f97d0266b58a89e36c015f56232c52b5d2f207215d48cca9e8f7/platformdirs-4.11.2.tar.gz", hash = "sha256:3a2ae5fca3520a01ab1be8b45613537f52ddf5b5f6f53d88233892dfbf0cd82d", size = 32716, upload-time = "2026-08-10T15:48:06.092Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/b8/d7/e7bfbc86e9f99ff7807e24de7703f032e9c9ba80bb355cf26e0e9bc5a75e/platformdirs-4.11.3.tar.gz", hash = "sha256:66a73d38a849810252df809a3d8bcbda8e26f6c189920e7535ad608a48dbb5ab", size = 33050, upload-time = "2026-08-13T22:43:27.52Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/49/e2/4e6eee633809c376c024821b91ade709cbfd040ec53939ffbcc292aa7eee/platformdirs-4.11.2-py3-none-any.whl", hash = "sha256:7f89089b6ea71bda7962953edcf784b2e2d9d285b40ad88be2bb75c6e9d82ab4", size = 23361, upload-time = "2026-08-10T15:48:04.855Z" },
+    { url = "https://files.pythonhosted.org/packages/19/a9/c34aebedd3a4c9afe5101b1b8713710b3fec18087c8a36c35d2f909861bd/platformdirs-4.11.3-py3-none-any.whl", hash = "sha256:5ed065d443751de711da036041a7a214122efc4a4de393b3f4137ba5576540e7", size = 23491, upload-time = "2026-08-13T22:43:26.121Z" },
 ]
 
 [[package]]
@@ -5201,11 +5201,11 @@ wheels = [
 
 [[package]]
 name = "pyphen"
-version = "0.18.0"
+version = "0.18.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ad/fb/dac486dca9e6a82437596614073e7236aed0e73a4b7913ccdefd325b236b/pyphen-0.18.0.tar.gz", hash = "sha256:bc8718e50d80c1bfb4d38a476fc708a6b5bf904549db130f06ca20376971393e", size = 2110506, upload-time = "2026-08-13T11:57:06.995Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/94/47/8430452269cd28863d73b903d07d329d058cf762527ff211b3864ba61fc7/pyphen-0.18.1.tar.gz", hash = "sha256:dbae6fbbe4f01cb206108b43573d857c67107be9d0e38eb1b08d6fa2210634a7", size = 2116411, upload-time = "2026-08-14T11:30:12.083Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/44/3d/fd5b56a0e27e81fe884bbf3361fb69ce8d47f55de09f73dcd388926f1cab/pyphen-0.18.0-py3-none-any.whl", hash = "sha256:92d96000b81b58d9388a63604000d3787f3f7dc5174deb2a58c528b77db3a720", size = 2116143, upload-time = "2026-08-13T11:57:04.756Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/1d/23801cf008f71575f0a4800463f349afa5f04b27fe783178a45cdc4d5edf/pyphen-0.18.1-py3-none-any.whl", hash = "sha256:0aa9051e15928cecadd4c632cea0258ba57215b2a197a39baa46abcdb0f47e84", size = 2116143, upload-time = "2026-08-14T11:30:10.428Z" },
 ]
 
 [[package]]