def update_session_status(session, status, archive_id=False):
"""Update stored session item status"""
if archive_id:
ytarchive().sessionsUpdate({
'id': session.id,
'archive_id': archive_id,
'state': status
})
else:
ytarchive().sessionsUpdate({'id': session.id, 'state': status})
def store_unharvested_file(file):
"""Insert a new live file into our stored cache"""
ytarchive().filesInsert({
'id': file.id,
'session_id': file.session_id,
'url': file.url,
'type': file.type,
'state': c.FILE_NEW
})
log(file, "New File", c.FILE_NEW)
def sync():
args = get_args()
if 'site' in args and args.site:
session = ytarchive().sessionsGet(id=None,
params={
'state': c.SESSION_PROCESSED,
'site_id': args.site,
'sort': 'last_updated:asc',
'limit': 1
})
else:
session = ytarchive().sessionsGet(id=None,
params={
'state': c.SESSION_PROCESSED,
'sort': 'last_updated:asc',
'limit': 1
})
if session:
update_session_status(session, c.SESSION_SYNCING)
log(session, "Session queued for archive.org sync", c.SESSION_SYNCING)
update_success = True
delete_success = True
archive_id = session_archive_id(session)
archive_info = False
session_files = ytarchive().filesGet(id=None,
params={
'session_id': session.id,
'state': c.FILE_PROCESSED
})
# remove any unchanged files and mark them as finished
if session.archive_id:
archive_info = get_item(archive_id)
session_files = finish_unchanged_files(archive_info, session_files)
update_success = archive_update(archive_id, session, session_files,
archive_info)
# remove any files no longer present on session
# note that we never delete videos or sessions for permanent backup
removed_session_files = ytarchive().filesGetRemoved(session.id)
if removed_session_files:
delete_success = archive_delete_removed_files(
archive_id, removed_session_files, session)
if not delete_success:
update_session_status(session, c.SESSION_FAILED)
log(session, "Failed to delete items from archive.org",
c.SESSION_FAILED, c.LOG_ERROR)
def log_harvest_run_end(site, items_added, items_updated, items_deleted):
"""Stored end time and various operation summary data for finished
harvest_run
"""
message = str(items_added) + " items added, " + str(
items_updated) + " updated, " + str(items_deleted) + " deleted"
ytarchive().logsInsert({
'time': time.time(),
'site_id': site['site_id'],
'type': 'harvest_run',
'severity': c.LOG_STATUS,
'message': message,
'state': 'harvested'
})
def get_stored_sessions(site):
"""Get previously stored site sessions to compare against live"""
stored_sessions = ytarchive().sessionsGet(
id=None, params={'site_id': site['site_id']})
if not stored_sessions:
stored_sessions = []
return stored_sessions
def validate():
site_id = None
args = get_args()
if 'site' in args and args.site:
site_id = args.site
"""Check all synced files to make sure their md5 hash matches the hash
stored on Archive.org"""
synced_session = ytarchive().sessionsGetSyncedOldest(site_id)
if synced_session:
session_log = ytarchive().logsGetSynced(synced_session.id)
validation_time = time.time() - (60 * 60)
if session_log and session_log.time < validation_time:
archive_info = get_item(synced_session.archive_id)
session_files = ytarchive().filesGetSynced(synced_session.id)
valid = validate_files(session_files, archive_info)
if valid:
ytarchive().sessionsUpdate({
'id': synced_session.id,
'validated': True
})
log(synced_session, "Session files validated",
c.SESSION_SYNCED)
cleanup_files(synced_session)
else:
ytarchive().sessionsUpdate({
'id': synced_session.id,
'state': c.SESSION_FAILED,
'validated': False
})
log(synced_session, "Session files failed validation",
c.SESSION_FAILED, c.LOG_ERROR)
def update_session_file_status(session_file, status):
"""Update stored file status and properties based on state"""
if status == c.FILE_PROCESSED:
data = {
'state': status,
'md5': session_file.md5,
'id': session_file.id}
elif status == c.FILE_FETCHED:
data = {
'state': status,
'filepath': session_file.filepath,
'id': session_file.id}
else:
data = {
'state': status,
'id': session_file.id}
ytarchive().filesUpdate(data)
def mark_stored_files_for_removal(stored_files):
"""Mark stored files that have finished processing as removed that are no longer present
on the live site"""
for stored_key, stored_file in enumerate(stored_files):
current_state = stored_file.state
# avoid deleting files that are actively being processed
condition = ((current_state == c.FILE_NEW
or current_state == c.FILE_SYNCED)
and stored_file.validated)
if condition:
ytarchive().filesUpdate({
'id': stored_file.id,
'url': stored_file.url,
'type': stored_file.type,
'state': c.FILE_REMOVED
})
log(stored_file, "File removed", c.FILE_REMOVED)
def get(self, id=None):
try:
data = ytarchive().logsGet(id, request.args)
except ValueError as e:
abort(400, str(e))
schema = LogSchema() if id else LogSchema(many=True)
data = schema.dump(data)
return response(results=data, id=id, type='log')
def last_run_time(site):
"""Get the unix timestamp from the last harvest run for a given site"""
results = ytarchive().logsGet(id=None,
params={
'site_id': site['site_id'],
'type': 'harvest_run',
'state': 'harvesting',
'sort': 'time:desc',
'limit': 1
})
if results:
return results[0].time
else:
return time.time()
def process():
site_id = None
args = get_args()
if 'site' in args and args.site:
site_id = args.site
"""Download files and metadata for the oldest updated session"""
updated_session = ytarchive().sessionsGetChangedOldest(site_id)
if updated_session:
ytarchive().sessionsUpdate({'id': updated_session.id, 'state': c.SESSION_FETCHING})
log(updated_session, "Files queued for download", c.SESSION_FETCHING)
updated_session_files = ytarchive().filesGetNewChanged(updated_session.id)
downloaded_session_files = download_session_files(updated_session, updated_session_files)
log(updated_session, "Files downloaded locally", c.SESSION_FETCHED)
ytarchive().sessionsUpdate({'id': updated_session.id, 'state': c.SESSION_PROCESSING})
hash_session_files(updated_session, downloaded_session_files)
ytarchive().sessionsUpdate({'id': updated_session.id, 'state': c.SESSION_PROCESSED})
log(updated_session, "Files hashed", c.SESSION_PROCESSED)
def store_sessions(site, last_run_time, created_after):
"""Loop through all sessions that have changed since last harvest run and
update their status in the youtube_archive MySQL database
"""
# previously harvested sessions
stored_sessions = get_stored_sessions(site)
stored_sessions = dict_by_id(stored_sessions)
# live sessions that have been updated after the last harvest run
sessions = get_live_sessions(site, stored_sessions, last_run_time,
created_after)
sessions = sessions_map_sessions(sessions)
results = {'new': [], 'updated': [], 'skipped': []}
for session in sessions:
current_state = session_current_state(session, stored_sessions)
next_state = session_next_state(current_state)
message = session_state_message(next_state)
if next_state == c.SESSION_NEW:
item = session_map_item(site, session, stored_sessions,
c.SESSION_NEW)
ytarchive().sessionsInsert(item)
results['new'].append(session)
store_files(session)
# session is new but already has archive information so we avoid it
elif next_state == c.SESSION_UNMANAGED:
item = session_map_item(site, session, stored_sessions,
c.SESSION_UNMANAGED)
ytarchive().sessionsInsert(item)
results['skipped'].append(session)
elif next_state == c.SESSION_SKIPPED:
results['skipped'].append(session)
# session has been previously imported, check for updates
else:
# only update if the session has been changed since last import
if stored_sessions[session.id].last_updated < session.updated:
# only update items that are not currently being processed
if session_processing_completed(session, current_state,
stored_sessions):
item = session_map_item(site, session, stored_sessions,
next_state)
ytarchive().sessionsUpdate(item)
results['updated'].append(session)
store_files(session)
log(session, message, next_state)
return results
def update_session_file_validation(session_file, validated):
"""shortcut to update stored validated status for items"""
ytarchive().filesUpdate({'id': session_file.id, 'validated': validated})
def get_stored_files(session):
"""Get previously stored session files to compare against live"""
stored_files = ytarchive().filesGet(id=None,
params={'session_id': session.id})
return stored_files
#!/usr/bin/env python3
from sqlalchemy_declarative import Session, File, Log
from db import ytarchive
session = ytarchive().db
# sample session
new_session = Session(
id=179,
site_id=1,
group="Admin",
archive_collection_id="coloradochannel",
archive_id="om-1-179",
title="Creating a Foundation for Change new title",
category="City Council",
state="synced",
created=1489783849,
last_updated=1531858777,
validated=1)
session.add(new_session)
session.commit()
# sample session
new_session = Session(
id=180,
site_id=1,
group="Admin",
archive_collection_id="coloradochannel",
archive_id="om-1-180",
title="Another session title",
def update_file_status(file, status):
"""Update stored session file status"""
ytarchive().filesUpdate({'id': file.id, 'state': status})
def update_session_file_status(session_file, status):
"""shortcut to update stored file state"""
ytarchive().filesUpdate({'id': session_file.id, 'state': status})
