diff --git a/PKG-INFO b/PKG-INFO index b24f080..2b76874 100644 --- a/PKG-INFO +++ b/PKG-INFO @@ -1,72 +1,72 @@ Metadata-Version: 2.1 Name: swh.journal -Version: 0.6.0 +Version: 0.6.1 Summary: Software Heritage Journal utilities Home-page: https://forge.softwareheritage.org/diffusion/DJNL/ Author: Software Heritage developers Author-email: swh-devel@inria.fr License: UNKNOWN Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-journal Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-journal/ Description: swh-journal =========== Persistent logger of changes to the archive, with publish-subscribe support. See the [documentation](https://docs.softwareheritage.org/devel/swh-journal/index.html#software-heritage-journal) for more details. # Local test As a pre-requisite, you need a kakfa installation path. The following target will take care of this: ``` make install ``` Then, provided you are in the right virtual environment as described in the [swh getting-started](https://docs.softwareheritage.org/devel/developer-setup.html#developer-setup): ``` pytest ``` or: ``` tox ``` # Running ## publisher Command: ``` $ swh-journal --config-file ~/.config/swh/journal/publisher.yml \ publisher ``` # Auto-completion To have the completion, add the following in your ~/.virtualenvs/swh/bin/postactivate: ``` eval "$(_SWH_JOURNAL_COMPLETE=$autocomplete_cmd swh-journal)" ``` Platform: UNKNOWN Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing diff --git a/debian/changelog b/debian/changelog index 7e9c51d..6ac4287 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,464 +1,468 @@ -swh-journal (0.6.0-1~swh1~bpo10+1) buster-swh; urgency=medium +swh-journal (0.6.1-1~swh1) unstable-swh; urgency=medium - * Rebuild for buster-swh + * New upstream release 0.6.1 - (tagged by Valentin Lorentz + on 2020-12-21 11:29:15 +0100) + * Upstream changes: - v0.6.1 - * serializers: Deserialize as + lists instead of tuples - * Fix dependency on msgpack (>= 1.0.0) + - * Blacklist msgpack 1.0.1. - -- Software Heritage autobuilder (on jenkins-debian1) Tue, 08 Dec 2020 14:32:15 +0000 + -- Software Heritage autobuilder (on jenkins-debian1) Mon, 21 Dec 2020 10:31:53 +0000 swh-journal (0.6.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.6.0 - (tagged by David Douard on 2020-12-08 15:27:18 +0100) * Upstream changes: - v0.6.0 -- Software Heritage autobuilder (on jenkins-debian1) Tue, 08 Dec 2020 14:31:08 +0000 swh-journal (0.5.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.1 - (tagged by Valentin Lorentz on 2020-11-05 15:17:23 +0100) * Upstream changes: - v0.5.1 - This release makes the JournalWriter usable by swh-indexer-storage: - * Make value_sanitizer an argument of JournalWriter. - * Make the type of values of JournalWriter generic, so it works with types not from swh-model. -- Software Heritage autobuilder (on jenkins-debian1) Thu, 05 Nov 2020 14:20:17 +0000 swh-journal (0.5.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.0 - (tagged by Nicolas Dandrimont on 2020-10-27 16:01:22 +0100) * Upstream changes: - Release swh.journal v0.5.0 - Use the model-provided unique_key as deduplication key - Drop long- deprecated swh.journal.cli module - Use RawExtrinsicMetadata.target attribute instead of .id - CI changes: pin black version, use upstream flake8 hook -- Software Heritage autobuilder (on jenkins-debian1) Tue, 27 Oct 2020 15:05:44 +0000 swh-journal (0.4.3-1~swh1) unstable-swh; urgency=medium * New upstream release 0.4.3 - (tagged by David Douard on 2020-09-25 11:51:00 +0200) * Upstream changes: - v0.4.3 -- Software Heritage autobuilder (on jenkins-debian1) Fri, 25 Sep 2020 09:53:37 +0000 swh-journal (0.4.2-1~swh2) unstable-swh; urgency=medium * Fix debian dependencies -- Antoine R. Dumont (@ardumont)) Fri, 07 Aug 2020 09:38:25 +0000 swh-journal (0.4.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.4.2 - (tagged by Antoine R. Dumont (@ardumont) on 2020-08-07 10:58:45 +0200) * Upstream changes: - v0.4.2 - pytest_plugin: Deal with the case when ctime is dropped - setup.py: Migrate from vcversioner to setuptools-scm -- Software Heritage autobuilder (on jenkins-debian1) Fri, 07 Aug 2020 09:01:25 +0000 swh-journal (0.4.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.4.1 - (tagged by Valentin Lorentz on 2020-07-31 11:25:19 +0200) * Upstream changes: - v0.4.1 - * Remove TEST_OBJECT_DICTS, use only TEST_OBJECTS. - * Add support for MetadataAuthority, MetadataFetcher, and RawExtrinsicMetadata. -- Software Heritage autobuilder (on jenkins-debian1) Fri, 31 Jul 2020 09:31:32 +0000 swh-journal (0.4.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.4.0 - (tagged by David Douard on 2020-07-06 13:49:40 +0200) * Upstream changes: - v0.4.0 -- Software Heritage autobuilder (on jenkins-debian1) Mon, 06 Jul 2020 11:58:06 +0000 swh-journal (0.3.5-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.5 - (tagged by Antoine R. Dumont (@ardumont) on 2020-07-01 15:55:31 +0200) * Upstream changes: - v0.3.5 - journal_data: Drop obsolete origin_visit fields - Use proper hash ids in tests' journal_data -- Software Heritage autobuilder (on jenkins-debian1) Wed, 01 Jul 2020 14:00:27 +0000 swh-journal (0.3.4-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.4 - (tagged by Antoine R. Dumont (@ardumont) on 2020-06-25 10:10:16 +0200) * Upstream changes: - v0.3.4 - Drop datetime conversion indirection -- Software Heritage autobuilder (on jenkins-debian1) Thu, 25 Jun 2020 08:13:02 +0000 swh-journal (0.3.3-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.3 - (tagged by Antoine R. Dumont (@ardumont) on 2020-06-25 09:34:32 +0200) * Upstream changes: - v0.3.3 - journal_data: Make origin-visit optional fields to None -- Software Heritage autobuilder (on jenkins-debian1) Thu, 25 Jun 2020 07:36:28 +0000 swh-journal (0.3.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.2 - (tagged by David Douard on 2020-06-17 09:29:44 +0200) * Upstream changes: - v0.3.2 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 17 Jun 2020 07:37:10 +0000 swh-journal (0.3.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.1 - (tagged by Antoine R. Dumont (@ardumont) on 2020-06-10 10:54:44 +0200) * Upstream changes: - v0.3.1 - pytest_plugin: pprint key when assertion is not respected -- Software Heritage autobuilder (on jenkins-debian1) Wed, 10 Jun 2020 08:56:57 +0000 swh-journal (0.3.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.0 - (tagged by Antoine R. Dumont (@ardumont) on 2020-06-09 10:42:30 +0200) * Upstream changes: - v0.3.0 - Allow journal to deal with origin_visit_status - test: Use origin-visit date field as datetime to phase out iso8601 str -- Software Heritage autobuilder (on jenkins-debian1) Tue, 09 Jun 2020 08:44:44 +0000 swh-journal (0.2.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.2.0 - (tagged by David Douard on 2020-06-03 13:50:54 +0200) * Upstream changes: - v0.2.0 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 03 Jun 2020 12:00:31 +0000 swh-journal (0.1.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.1.0 - (tagged by David Douard on 2020-05-07 10:33:18 +0200) * Upstream changes: - v0.1.0 -- Software Heritage autobuilder (on jenkins-debian1) Thu, 07 May 2020 08:35:49 +0000 swh-journal (0.0.32-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.32 - (tagged by Antoine R. Dumont (@ardumont) on 2020-05-04 18:02:45 +0200) * Upstream changes: - v0.0.32 - serializers: Make kafka_to_key implem compatible with stable version - setup.py: add documentation link - Remove the content replayer code - Remove the backfiller and the (storage) replayer -- Software Heritage autobuilder (on jenkins-debian1) Mon, 04 May 2020 16:04:38 +0000 swh-journal (0.0.31-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.31 - (tagged by David Douard on 2020-04-23 12:27:47 +0200) * Upstream changes: - v0.0.31 -- Software Heritage autobuilder (on jenkins-debian1) Thu, 23 Apr 2020 10:37:28 +0000 swh-journal (0.0.30-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.30 - (tagged by Nicolas Dandrimont on 2020-04-14 15:57:34 +0200) * Upstream changes: - Release swh.journal v0.0.30 - various test refactorings - enable black on the source code - accept swh.model objects in the journal writers - listen to delivery notifications after writing messages - default to message.max.bytes = 100 MB in the kafka writer -- Software Heritage autobuilder (on jenkins-debian1) Tue, 14 Apr 2020 14:10:54 +0000 swh-journal (0.0.29-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.29 - (tagged by Antoine R. Dumont (@ardumont) on 2020-03-27 10:40:46 +0100) * Upstream changes: - v0.0.29 - replayer: Allow legacy origin to be replayed - tests.utils: Test from modules using this need to call close method - tests: Adapt model according to latest change -- Software Heritage autobuilder (on jenkins-debian1) Fri, 27 Mar 2020 09:51:08 +0000 swh-journal (0.0.28-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.28 - (tagged by Antoine R. Dumont (@ardumont) on 2020-03-25 10:14:47 +0100) * Upstream changes: - v0.0.28 - journal.replay: Migrate to latest HashCollision change - replayer: factor out legacy objects fixers - journal.replay: Inline `_fix_origin_visit` for loop in insert_object - journal.replay: Align _fix_content with other fix methods - journal.replay: Align fix revision behavior to other fix methods - Remove extra 'perms' key of contents when replaying. -- Software Heritage autobuilder (on jenkins-debian1) Wed, 25 Mar 2020 09:23:55 +0000 swh-journal (0.0.27-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.27 - (tagged by Antoine R. Dumont (@ardumont) on 2020-03-16 16:09:33 +0100) * Upstream changes: - v0.0.27 - Migrate to latest origin_visit_upsert/add api changes - journal: Use swh-model objects instead of dicts in replay and writer - replay: Filter out colliding contents when replaying - Use better kafka producer semantics in the journal writers - Make the number of messages processed at a time by journal clients - configurable - Drop deprecated cli options - Replace deprecated options with a config file override in cli tests - Clean up the signature of test_cli's invoke method - Migrate test cli config to a dict instead of raw yaml - kafka: normalize KafkaJournalWriter.write_addition[s] API - Rename JournalClient.max_messages to JournalClient.stop_after_objects - Be more careful with content generation in test_write_replay - Add type annotations to swh.journal.client arguments - Unify tense for content replay statistics log entry - Add missing log4j.properties file from MANIFEST.in - Unify retry/error handling for content replay -- Software Heritage autobuilder (on jenkins-debian1) Mon, 16 Mar 2020 15:17:43 +0000 swh-journal (0.0.26-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.26 - (tagged by David Douard on 2020-03-06 15:36:23 +0100) * Upstream changes: - v0.0.26 -- Software Heritage autobuilder (on jenkins-debian1) Fri, 06 Mar 2020 14:47:18 +0000 swh-journal (0.0.25-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.25 - (tagged by Valentin Lorentz on 2020-01-30 15:35:41 +0100) * Upstream changes: - v0.0.25 - * Add support for swh-storage v0.0.168. - * Accept None dates when validating revisions. -- Software Heritage autobuilder (on jenkins-debian1) Thu, 30 Jan 2020 14:43:25 +0000 swh-journal (0.0.24-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.24 - (tagged by Antoine Lambert on 2020-01-06 16:14:32 +0100) * Upstream changes: - version 0.0.24 -- Software Heritage autobuilder (on jenkins-debian1) Mon, 06 Jan 2020 15:20:54 +0000 swh-journal (0.0.23-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.23 - (tagged by Nicolas Dandrimont on 2020-01-03 20:02:39 +0100) * Upstream changes: - Release swh.journal v0.0.23 - support short-hand syntax for initializing the journal direct writer -- Software Heritage autobuilder (on jenkins-debian1) Fri, 03 Jan 2020 19:06:31 +0000 swh-journal (0.0.21-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.21 - (tagged by David Douard on 2019-11-29 15:37:49 +0100) * Upstream changes: - v0.0.21 -- Software Heritage autobuilder (on jenkins-debian1) Mon, 02 Dec 2019 14:12:49 +0000 swh-journal (0.0.20-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.20 - (tagged by David Douard on 2019-11-29 12:06:13 +0100) * Upstream changes: - v0.0.20 -- Software Heritage autobuilder (on jenkins-debian1) Fri, 29 Nov 2019 11:10:44 +0000 swh-journal (0.0.19-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.19 - (tagged by Nicolas Dandrimont on 2019-11-07 14:33:58 +0100) * Upstream changes: - Release swh.journal v0.0.19 - Merge several reliability fixes -- Software Heritage autobuilder (on jenkins-debian1) Thu, 07 Nov 2019 13:37:23 +0000 swh-journal (0.0.18-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.18 - (tagged by Stefano Zacchiroli on 2019-10-01 10:19:26 +0200) * Upstream changes: - v0.0.18 - * tox: anticipate mypy run to just after flake8 - * init.py: switch to documented way of extending path - * typing: minimal changes to make a no-op mypy run pass - * writer: Normalize 'cls' value to 'memory' for in- memory instantiation - * Add a test directly for the journal client. -- Software Heritage autobuilder (on jenkins-debian1) Tue, 01 Oct 2019 11:12:03 +0000 swh-journal (0.0.17-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.17 - (tagged by Nicolas Dandrimont on 2019-09-18 18:01:50 +0200) * Upstream changes: - Release swh.journal v0.0.17 - Cleanup fallout from confluent_kafka migration - Better error handling and logging in direct_writer - Backfiller fixups - More extensive mock for KafkaConsumer -- Software Heritage autobuilder (on jenkins-debian1) Wed, 18 Sep 2019 16:08:20 +0000 swh-journal (0.0.16-1~swh2) unstable-swh; urgency=medium * Migrate to confluent-kafka -- Nicolas Dandrimont Fri, 13 Sep 2019 20:11:24 +0200 swh-journal (0.0.16-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.16 - (tagged by Nicolas Dandrimont on 2019-09-13 14:28:50 +0200) * Upstream changes: - Release swh.journal v0.0.16: - Migrate to confluent-kafka from python-kafka -- Software Heritage autobuilder (on jenkins-debian1) Fri, 13 Sep 2019 12:34:35 +0000 swh-journal (0.0.15-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.15 - (tagged by David Douard on 2019-09-10 16:50:42 +0200) * Upstream changes: - v0.0.15 -- Software Heritage autobuilder (on jenkins-debian1) Tue, 10 Sep 2019 14:53:55 +0000 swh-journal (0.0.14-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.14 - (tagged by David Douard on 2019-07-18 13:38:39 +0200) * Upstream changes: - 0.0.14 - Code of conduct - fix the backfiller - fix compatibility with click < 7 - make the replayer robust against old formats -- Software Heritage autobuilder (on jenkins-debian1) Thu, 18 Jul 2019 11:44:40 +0000 swh-journal (0.0.13-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.13 - (tagged by Antoine R. Dumont (@ardumont) on 2019-07-03 10:26:29 +0200) * Upstream changes: - v0.0.13 - cli: Document depreated options -- Software Heritage autobuilder (on jenkins-debian1) Wed, 03 Jul 2019 08:33:57 +0000 swh-journal (0.0.12-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.12 - (tagged by Valentin Lorentz on 2019-07-02 11:58:00 +0200) * Upstream changes: - v0.0.12 - More CLI option - Replay parallelism - Fix build on Debian 9 -- Software Heritage autobuilder (on jenkins-debian1) Tue, 02 Jul 2019 10:08:07 +0000 swh-journal (0.0.11-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.11 - (tagged by David Douard on 2019-06-12 13:58:14 +0200) * Upstream changes: - v0.0.11 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 12 Jun 2019 12:10:52 +0000 swh-journal (0.0.10-1~swh2) unstable-swh; urgency=medium * Disable tests at build-time -- Nicolas Dandrimont Thu, 09 May 2019 14:42:24 +0200 swh-journal (0.0.10-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.10 - (tagged by Nicolas Dandrimont on 2019-05-09 14:29:52 +0200) * Upstream changes: - Release swh.journal v0.0.10 - Remove the publisher component, introduce the backfiller component. -- Software Heritage autobuilder (on jenkins-debian1) Thu, 09 May 2019 12:34:36 +0000 swh-journal (0.0.9-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.9 - (tagged by David Douard on 2019-04-10 13:42:32 +0200) * Upstream changes: - v0.0.9 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 10 Apr 2019 11:48:39 +0000 swh-journal (0.0.8-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.8 - (tagged by Antoine R. Dumont (@ardumont) on 2019-03-15 13:56:57 +0100) * Upstream changes: - v0.0.8 - Add swh-journal cli -- Software Heritage autobuilder (on jenkins-debian1) Fri, 15 Mar 2019 13:00:11 +0000 swh-journal (0.0.7-1~swh2) unstable-swh; urgency=low * New release fixing build dependencies -- Antoine Romain Dumont (@ardumont) Tue, 19 Feb 2019 14:18:06 +0100 swh-journal (0.0.7-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.7 - (tagged by Antoine R. Dumont (@ardumont) on 2019-01-11 11:53:44 +0100) * Upstream changes: - v0.0.7 - Fix off-by-one error when checking max_messages. - tests: Adapt tests according to latest in-memory storage changes -- Software Heritage autobuilder (on jenkins-debian1) Fri, 11 Jan 2019 10:56:46 +0000 swh-journal (0.0.4-1~swh1) unstable-swh; urgency=medium * Release swh.journal version 0.0.4 * Update packaging runes -- Nicolas Dandrimont Thu, 12 Oct 2017 19:01:53 +0200 swh-journal (0.0.3-1~swh1) unstable-swh; urgency=medium * Release swh.journal v0.0.3 * Prepare building for stretch -- Nicolas Dandrimont Fri, 30 Jun 2017 17:29:15 +0200 swh-journal (0.0.2-1~swh1) unstable-swh; urgency=medium * v0.0.2 * Adapt swh.journal.publisher * Adapt swh.journal.client * Add swh.journal.checker basic implementation (reads and sends all * objects to publisher's subscribed queues). -- Antoine R. Dumont (@ardumont) Fri, 24 Mar 2017 12:54:16 +0100 swh-journal (0.0.1-1~swh1) unstable-swh; urgency=medium * Initial release * v0.0.1 * Add a journal publisher * Add a base class interface for journal clients -- Antoine R. Dumont (@ardumont) Tue, 21 Mar 2017 14:38:13 +0100 diff --git a/requirements.txt b/requirements.txt index b83b81a..91043c5 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,6 +1,6 @@ # Add here external Python modules dependencies, one per line. Module names # should match https://pypi.python.org/pypi names. For the full spec or # dependency lines, see https://pip.readthedocs.org/en/1.1/requirements.html confluent-kafka -msgpack +msgpack >= 1.0.0, != 1.0.1 tenacity diff --git a/swh.journal.egg-info/PKG-INFO b/swh.journal.egg-info/PKG-INFO index b24f080..2b76874 100644 --- a/swh.journal.egg-info/PKG-INFO +++ b/swh.journal.egg-info/PKG-INFO @@ -1,72 +1,72 @@ Metadata-Version: 2.1 Name: swh.journal -Version: 0.6.0 +Version: 0.6.1 Summary: Software Heritage Journal utilities Home-page: https://forge.softwareheritage.org/diffusion/DJNL/ Author: Software Heritage developers Author-email: swh-devel@inria.fr License: UNKNOWN Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-journal Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-journal/ Description: swh-journal =========== Persistent logger of changes to the archive, with publish-subscribe support. See the [documentation](https://docs.softwareheritage.org/devel/swh-journal/index.html#software-heritage-journal) for more details. # Local test As a pre-requisite, you need a kakfa installation path. The following target will take care of this: ``` make install ``` Then, provided you are in the right virtual environment as described in the [swh getting-started](https://docs.softwareheritage.org/devel/developer-setup.html#developer-setup): ``` pytest ``` or: ``` tox ``` # Running ## publisher Command: ``` $ swh-journal --config-file ~/.config/swh/journal/publisher.yml \ publisher ``` # Auto-completion To have the completion, add the following in your ~/.virtualenvs/swh/bin/postactivate: ``` eval "$(_SWH_JOURNAL_COMPLETE=$autocomplete_cmd swh-journal)" ``` Platform: UNKNOWN Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing diff --git a/swh.journal.egg-info/requires.txt b/swh.journal.egg-info/requires.txt index c1a5a7d..ffa69af 100644 --- a/swh.journal.egg-info/requires.txt +++ b/swh.journal.egg-info/requires.txt @@ -1,8 +1,8 @@ confluent-kafka -msgpack +msgpack!=1.0.1,>=1.0.0 tenacity swh.model>=0.7.2 [testing] pytest hypothesis diff --git a/swh/journal/pytest_plugin.py b/swh/journal/pytest_plugin.py index 9f8d458..f226f2a 100644 --- a/swh/journal/pytest_plugin.py +++ b/swh/journal/pytest_plugin.py @@ -1,239 +1,258 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import random import string -from typing import Collection, Dict, Iterator, Optional +from typing import Any, Collection, Dict, Iterator, Optional import attr from confluent_kafka import Consumer, KafkaException, Producer from confluent_kafka.admin import AdminClient import pytest from swh.journal.serializers import kafka_to_key, kafka_to_value, pprint_key from swh.journal.tests.journal_data import TEST_OBJECTS +def ensure_lists(value: Any) -> Any: + """ + >>> ensure_lists(["foo", 42]) + ['foo', 42] + >>> ensure_lists(("foo", 42)) + ['foo', 42] + >>> ensure_lists({"a": ["foo", 42]}) + {'a': ['foo', 42]} + >>> ensure_lists({"a": ("foo", 42)}) + {'a': ['foo', 42]} + """ + if isinstance(value, (tuple, list)): + return list(map(ensure_lists, value)) + elif isinstance(value, dict): + return dict(ensure_lists(list(value.items()))) + else: + return value + + def consume_messages(consumer, kafka_prefix, expected_messages): """Consume expected_messages from the consumer; Sort them all into a consumed_objects dict""" consumed_messages = defaultdict(list) fetched_messages = 0 retries_left = 1000 while fetched_messages < expected_messages: if retries_left == 0: raise ValueError( "Timed out fetching messages from kafka. " f"Only {fetched_messages}/{expected_messages} fetched" ) msg = consumer.poll(timeout=0.01) if not msg: retries_left -= 1 continue error = msg.error() if error is not None: if error.fatal(): raise KafkaException(error) retries_left -= 1 continue fetched_messages += 1 topic = msg.topic() assert topic.startswith(f"{kafka_prefix}.") or topic.startswith( f"{kafka_prefix}_privileged." ), "Unexpected topic" object_type = topic[len(kafka_prefix + ".") :] consumed_messages[object_type].append( (kafka_to_key(msg.key()), kafka_to_value(msg.value())) ) return consumed_messages def assert_all_objects_consumed( consumed_messages: Dict, exclude: Optional[Collection] = None ): """Check whether all objects from TEST_OBJECTS have been consumed `exclude` can be a list of object types for which we do not want to compare the values (eg. for anonymized object). """ for object_type, known_objects in TEST_OBJECTS.items(): known_keys = [obj.unique_key() for obj in known_objects] if not consumed_messages[object_type]: return (received_keys, received_values) = zip(*consumed_messages[object_type]) if object_type in ("content", "skipped_content"): for value in received_values: value.pop("ctime", None) if object_type == "content": known_objects = [attr.evolve(o, data=None) for o in known_objects] for key in known_keys: assert key in received_keys, ( f"expected {object_type} key {pprint_key(key)} " "absent from consumed messages" ) if exclude and object_type in exclude: continue for value in known_objects: expected_value = value.to_dict() if value.object_type in ("content", "skipped_content"): expected_value.pop("ctime", None) - assert expected_value in received_values, ( + assert ensure_lists(expected_value) in received_values, ( f"expected {object_type} value {value!r} is " "absent from consumed messages" ) @pytest.fixture(scope="function") def kafka_prefix(): """Pick a random prefix for kafka topics on each call""" return "".join(random.choice(string.ascii_lowercase) for _ in range(10)) @pytest.fixture(scope="function") def kafka_consumer_group(kafka_prefix: str): """Pick a random consumer group for kafka consumers on each call""" return "test-consumer-%s" % kafka_prefix @pytest.fixture(scope="function") def object_types(): """Set of object types to precreate topics for.""" return set(TEST_OBJECTS.keys()) @pytest.fixture(scope="function") def privileged_object_types(): """Set of object types to precreate privileged topics for.""" return {"revision", "release"} @pytest.fixture(scope="function") def kafka_server( kafka_server_base: str, kafka_prefix: str, object_types: Iterator[str], privileged_object_types: Iterator[str], ) -> str: """A kafka server with existing topics Unprivileged topics are built as ``{kafka_prefix}.{object_type}`` with object_type from the ``object_types`` list. Privileged topics are built as ``{kafka_prefix}_privileged.{object_type}`` with object_type from the ``privileged_object_types`` list. """ topics = [f"{kafka_prefix}.{obj}" for obj in object_types] + [ f"{kafka_prefix}_privileged.{obj}" for obj in privileged_object_types ] # unfortunately, the Mock broker does not support the CreatTopic admin API, so we # have to create topics using a Producer. producer = Producer( { "bootstrap.servers": kafka_server_base, "client.id": "bootstrap producer", "acks": "all", } ) for topic in topics: producer.produce(topic=topic, value=None) for i in range(10): if producer.flush(0.1) == 0: break return kafka_server_base @pytest.fixture(scope="session") def kafka_server_base() -> Iterator[str]: """Create a mock kafka cluster suitable for tests. Yield a connection string. Note: this is a generator to keep the mock broker alive during the whole test session. see https://github.com/edenhill/librdkafka/blob/master/src/rdkafka_mock.h """ admin = AdminClient({"test.mock.num.brokers": "1"}) metadata = admin.list_topics() brokers = [str(broker) for broker in metadata.brokers.values()] assert len(brokers) == 1, "More than one broker found in the kafka cluster?!" broker_connstr, broker_id = brokers[0].split("/") yield broker_connstr TEST_CONFIG = { "consumer_id": "swh.journal.consumer", "stop_after_objects": 1, # will read 1 object and stop "storage": {"cls": "memory", "args": {}}, } @pytest.fixture def test_config( kafka_server_base: str, kafka_prefix: str, object_types: Iterator[str], privileged_object_types: Iterator[str], ): """Test configuration needed for producer/consumer """ return { **TEST_CONFIG, "object_types": object_types, "privileged_object_types": privileged_object_types, "brokers": [kafka_server_base], "prefix": kafka_prefix, } @pytest.fixture def consumer( kafka_server: str, test_config: Dict, kafka_consumer_group: str ) -> Consumer: """Get a connected Kafka consumer. """ consumer = Consumer( { "bootstrap.servers": kafka_server, "auto.offset.reset": "earliest", "enable.auto.commit": True, "group.id": kafka_consumer_group, } ) prefix = test_config["prefix"] kafka_topics = [ f"{prefix}.{object_type}" for object_type in test_config["object_types"] ] + [ f"{prefix}_privileged.{object_type}" for object_type in test_config["privileged_object_types"] ] consumer.subscribe(kafka_topics) yield consumer consumer.close() diff --git a/swh/journal/serializers.py b/swh/journal/serializers.py index 41c48e8..d39d2db 100644 --- a/swh/journal/serializers.py +++ b/swh/journal/serializers.py @@ -1,123 +1,113 @@ # Copyright (C) 2016-2017 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import datetime from enum import Enum from typing import Any, Union import msgpack from swh.model.model import KeyType class MsgpackExtTypeCodes(Enum): LONG_INT = 1 LONG_NEG_INT = 2 # this as been copied from swh.core.api.serializer # TODO refactor swh.core to make this function available def _msgpack_encode_longint(value): # needed because msgpack will not handle long integers with more than 64 bits # which we unfortunately happen to have to deal with from time to time if value > 0: code = MsgpackExtTypeCodes.LONG_INT.value else: code = MsgpackExtTypeCodes.LONG_NEG_INT.value value = -value length, rem = divmod(value.bit_length(), 8) if rem: length += 1 return msgpack.ExtType(code, int.to_bytes(value, length, "big")) def msgpack_ext_encode_types(obj): if isinstance(obj, int): return _msgpack_encode_longint(obj) return obj def msgpack_ext_hook(code, data): if code == MsgpackExtTypeCodes.LONG_INT.value: return int.from_bytes(data, "big") if code == MsgpackExtTypeCodes.LONG_NEG_INT.value: return -int.from_bytes(data, "big") raise ValueError("Unknown msgpack extended code %s" % code) # for BW compat def decode_types_bw(obj): if set(obj.keys()) == {b"d", b"swhtype"} and obj[b"swhtype"] == "datetime": return datetime.datetime.fromisoformat(obj[b"d"]) return obj def stringify_key_item(k: str, v: Union[str, bytes]) -> str: """Turn the item of a dict key into a string""" if isinstance(v, str): return v if k == "url": return v.decode("utf-8") return v.hex() def pprint_key(key: KeyType) -> str: """Pretty-print a kafka key""" if isinstance(key, dict): return "{%s}" % ", ".join( f"{k}: {stringify_key_item(k, v)}" for k, v in key.items() ) elif isinstance(key, bytes): return key.hex() else: return key def key_to_kafka(key: KeyType) -> bytes: """Serialize a key, possibly a dict, in a predictable way""" p = msgpack.Packer(use_bin_type=True) if isinstance(key, dict): return p.pack_map_pairs(sorted(key.items())) else: return p.pack(key) def kafka_to_key(kafka_key: bytes) -> KeyType: """Deserialize a key""" return msgpack.loads(kafka_key, raw=False) def value_to_kafka(value: Any) -> bytes: """Serialize some data for storage in kafka""" return msgpack.packb( value, use_bin_type=True, datetime=True, # encode datetime as msgpack.Timestamp default=msgpack_ext_encode_types, ) def kafka_to_value(kafka_value: bytes) -> Any: """Deserialize some data stored in kafka""" - value = msgpack.unpackb( + return msgpack.unpackb( kafka_value, raw=False, object_hook=decode_types_bw, ext_hook=msgpack_ext_hook, strict_map_key=False, timestamp=3, # convert Timestamp in datetime objects (tz UTC) ) - return ensure_tuples(value) - - -def ensure_tuples(value: Any) -> Any: - if isinstance(value, (tuple, list)): - return tuple(map(ensure_tuples, value)) - elif isinstance(value, dict): - return dict(ensure_tuples(list(value.items()))) - else: - return value diff --git a/swh/journal/tests/test_client.py b/swh/journal/tests/test_client.py index 3984c46..63f8646 100644 --- a/swh/journal/tests/test_client.py +++ b/swh/journal/tests/test_client.py @@ -1,329 +1,329 @@ # Copyright (C) 2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Dict, List from unittest.mock import MagicMock from confluent_kafka import Producer import pytest from swh.journal.client import JournalClient from swh.journal.serializers import key_to_kafka, value_to_kafka from swh.model.model import Content REV = { "message": b"something cool", "author": {"fullname": b"Peter", "name": None, "email": b"peter@ouiche.lo"}, "committer": {"fullname": b"Stephen", "name": b"From Outer Space", "email": None}, "date": { "timestamp": {"seconds": 123456789, "microseconds": 123}, "offset": 120, "negative_utc": False, }, "committer_date": { "timestamp": {"seconds": 123123456, "microseconds": 0}, "offset": 0, "negative_utc": False, }, "type": "git", "directory": ( b"\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00" b"\x01\x02\x03\x04\x05" ), "synthetic": False, "metadata": None, - "parents": (), + "parents": [], "id": b"\x8b\xeb\xd1\x9d\x07\xe2\x1e0\xe2 \x91X\x8d\xbd\x1c\xa8\x86\xdeB\x0c", } def test_client(kafka_prefix: str, kafka_consumer_group: str, kafka_server: str): producer = Producer( { "bootstrap.servers": kafka_server, "client.id": "test producer", "acks": "all", } ) # Fill Kafka producer.produce( topic=kafka_prefix + ".revision", key=REV["id"], value=value_to_kafka(REV), ) producer.flush() client = JournalClient( brokers=[kafka_server], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=1, ) worker_fn = MagicMock() client.process(worker_fn) worker_fn.assert_called_once_with({"revision": [REV]}) def test_client_eof(kafka_prefix: str, kafka_consumer_group: str, kafka_server: str): producer = Producer( { "bootstrap.servers": kafka_server, "client.id": "test producer", "acks": "all", } ) # Fill Kafka producer.produce( topic=kafka_prefix + ".revision", key=REV["id"], value=value_to_kafka(REV), ) producer.flush() client = JournalClient( brokers=[kafka_server], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=None, stop_on_eof=True, ) worker_fn = MagicMock() client.process(worker_fn) worker_fn.assert_called_once_with({"revision": [REV]}) @pytest.mark.parametrize("batch_size", [1, 5, 100]) def test_client_batch_size( kafka_prefix: str, kafka_consumer_group: str, kafka_server: str, batch_size: int, ): num_objects = 2 * batch_size + 1 assert num_objects < 256, "Too many objects, generation will fail" producer = Producer( { "bootstrap.servers": kafka_server, "client.id": "test producer", "acks": "all", } ) contents = [Content.from_data(bytes([i])) for i in range(num_objects)] # Fill Kafka for content in contents: producer.produce( topic=kafka_prefix + ".content", key=key_to_kafka(content.sha1), value=value_to_kafka(content.to_dict()), ) producer.flush() client = JournalClient( brokers=[kafka_server], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=num_objects, batch_size=batch_size, ) collected_output: List[Dict] = [] def worker_fn(objects): received = objects["content"] assert len(received) <= batch_size collected_output.extend(received) client.process(worker_fn) expected_output = [content.to_dict() for content in contents] assert len(collected_output) == len(expected_output) for output in collected_output: assert output in expected_output @pytest.fixture() def kafka_producer(kafka_prefix: str, kafka_server_base: str): producer = Producer( { "bootstrap.servers": kafka_server_base, "client.id": "test producer", "acks": "all", } ) # Fill Kafka producer.produce( topic=kafka_prefix + ".something", key=key_to_kafka(b"key1"), value=value_to_kafka("value1"), ) producer.produce( topic=kafka_prefix + ".else", key=key_to_kafka(b"key1"), value=value_to_kafka("value2"), ) producer.flush() return producer def test_client_subscribe_all( kafka_producer: Producer, kafka_prefix: str, kafka_server_base: str ): client = JournalClient( brokers=[kafka_server_base], group_id="whatever", prefix=kafka_prefix, stop_after_objects=2, ) assert set(client.subscription) == { f"{kafka_prefix}.something", f"{kafka_prefix}.else", } worker_fn = MagicMock() client.process(worker_fn) worker_fn.assert_called_once_with( {"something": ["value1"], "else": ["value2"],} ) def test_client_subscribe_one_topic( kafka_producer: Producer, kafka_prefix: str, kafka_server_base: str ): client = JournalClient( brokers=[kafka_server_base], group_id="whatever", prefix=kafka_prefix, stop_after_objects=1, object_types=["else"], ) assert client.subscription == [f"{kafka_prefix}.else"] worker_fn = MagicMock() client.process(worker_fn) worker_fn.assert_called_once_with({"else": ["value2"]}) def test_client_subscribe_absent_topic( kafka_producer: Producer, kafka_prefix: str, kafka_server_base: str ): with pytest.raises(ValueError): JournalClient( brokers=[kafka_server_base], group_id="whatever", prefix=kafka_prefix, stop_after_objects=1, object_types=["really"], ) def test_client_subscribe_absent_prefix( kafka_producer: Producer, kafka_prefix: str, kafka_server_base: str ): with pytest.raises(ValueError): JournalClient( brokers=[kafka_server_base], group_id="whatever", prefix="wrong.prefix", stop_after_objects=1, ) with pytest.raises(ValueError): JournalClient( brokers=[kafka_server_base], group_id="whatever", prefix="wrong.prefix", stop_after_objects=1, object_types=["else"], ) def test_client_subscriptions_with_anonymized_topics( kafka_prefix: str, kafka_consumer_group: str, kafka_server_base: str ): producer = Producer( { "bootstrap.servers": kafka_server_base, "client.id": "test producer", "acks": "all", } ) # Fill Kafka with revision object on both the regular prefix (normally for # anonymized objects in this case) and privileged one producer.produce( topic=kafka_prefix + ".revision", key=REV["id"], value=value_to_kafka(REV), ) producer.produce( topic=kafka_prefix + "_privileged.revision", key=REV["id"], value=value_to_kafka(REV), ) producer.flush() # without privileged "channels" activated on the client side client = JournalClient( brokers=[kafka_server_base], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=1, privileged=False, ) # we only subscribed to "standard" topics assert client.subscription == [kafka_prefix + ".revision"] # with privileged "channels" activated on the client side client = JournalClient( brokers=[kafka_server_base], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=1, privileged=True, ) # we only subscribed to "privileged" topics assert client.subscription == [kafka_prefix + "_privileged.revision"] def test_client_subscriptions_without_anonymized_topics( kafka_prefix: str, kafka_consumer_group: str, kafka_server_base: str ): producer = Producer( { "bootstrap.servers": kafka_server_base, "client.id": "test producer", "acks": "all", } ) # Fill Kafka with revision objects only on the standard prefix producer.produce( topic=kafka_prefix + ".revision", key=REV["id"], value=value_to_kafka(REV), ) producer.flush() # without privileged channel activated on the client side client = JournalClient( brokers=[kafka_server_base], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=1, privileged=False, ) # we only subscribed to the standard prefix assert client.subscription == [kafka_prefix + ".revision"] # with privileged channel activated on the client side client = JournalClient( brokers=[kafka_server_base], group_id=kafka_consumer_group, prefix=kafka_prefix, stop_after_objects=1, privileged=True, ) # we also only subscribed to the standard prefix, since there is no priviled prefix # on the kafka broker assert client.subscription == [kafka_prefix + ".revision"]