diff --git a/debian/control b/debian/control --- a/debian/control +++ b/debian/control @@ -40,6 +40,7 @@ Package: python3-swh.storage.archiver Architecture: all Depends: python3-swh.scheduler (>= 0.0.11~), + python3-swh.journal, python3-swh.storage (= ${binary:Version}), ${misc:Depends}, ${python3:Depends} diff --git a/swh/storage/archiver/updater.py b/swh/storage/archiver/updater.py new file mode 100644 --- /dev/null +++ b/swh/storage/archiver/updater.py @@ -0,0 +1,47 @@ +# Copyright (C) 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 logging + +from swh.journal.client import SWHJournalClient + +from .storage import ArchiverStorage + + +class SWHArchiverContentUpdater(SWHJournalClient): + """Client in charge to update new content in the content_archiver db. + + """ + CONFIG_BASE_FILENAME = 'archiver/content_updater' + + ADDITIONAL_CONFIG = { + 'archiver_storage_conn': ( + 'str', 'dbname=softwareheritage-archiver-dev user=guest'), + 'sources_present': ('list[str]', ['uffizi']) + } + + def __init__(self): + # Only interested in content here so override the configuration + super().__init__(extra_configuration={'object_types': ['content']}) + + self.sources_present = self.config['sources_present'] + + self.archiver_storage = ArchiverStorage( + self.config['archiver_storage_conn']) + + def process_objects(self, messages): + self.archiver_storage.content_archive_add( + (c[b'sha1'] for c in messages['content']), + self.sources_present) + + +if __name__ == '__main__': + logging.basicConfig( + level=logging.INFO, + format='%(asctime)s %(process)d %(levelname)s %(message)s' + ) + + content_updater = SWHArchiverContentUpdater() + content_updater.process()