def run(self):
for doc, count in iter(self.queue.get, None):
try:
if self.exclude_types and doc["doc_type"] in self.exclude_types:
print " SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, self.total, doc["doc_type"], doc["_id"])
else:
if not self.simulate:
for i in reversed(range(5)):
try:
dt = DocumentTransform(doc, self.sourcedb, self.exclude_attachments)
break
except RequestError as r:
if i == 0:
_soft_assert(False, 'Copy domain failed after 5 tries with {}'.format(r))
raise
for i in reversed(range(5)):
try:
save(dt, self.targetdb)
break
except (ResourceConflict, ParserError, TypeError) as e:
if i == 0:
_soft_assert(False, 'Copy domain failed after 5 tries with {}'.format(e))
raise
print " Synced %s/%s docs (%s: %s)" % (count, self.total, doc["doc_type"], doc["_id"])
except Exception, e:
self.err_log.write('%s\n' % doc["_id"])
print " Document %s failed! Error is: %s %s" % (doc["_id"], e.__class__.__name__, e)
def copy_domain(self, sourcedb, domain):
print "Copying domain doc"
result = sourcedb.view("domain/domains", key=domain, reduce=False, include_docs=True).first()
if result and "doc" in result:
domain_doc = Domain.wrap(result["doc"])
dt = DocumentTransform(domain_doc, sourcedb)
save(dt, self.targetdb)
else:
print "Domain doc not found for domain %s." % domain
def handle(self, *args, **options):
raise CommandError(
'copy_group_data is currently broken. '
'Ask Danny or Ethan to fix it along the lines of '
'https://github.com/dimagi/commcare-hq/pull/9180/files#diff-9d976dc051a36a028c6604581dfbce5dR95'
)
if len(args) != 2:
raise CommandError('Usage is copy_group_data %s' % self.args)
sourcedb = Database(args[0])
group_id = args[1]
exclude_user_owned = options["exclude_user_owned"]
print 'getting group'
group = Group.wrap(sourcedb.get(group_id))
group.save(force_update=True)
print 'getting domain'
domain = Domain.wrap(
sourcedb.view('domain/domains',
key=group.domain,
include_docs=True,
reduce=False,
limit=1).one()['doc'])
dt = DocumentTransform(domain._obj, sourcedb)
save(dt, Domain.get_db())
owners = [group_id]
if not exclude_user_owned:
owners.extend(group.users)
print 'getting case ids'
with OverrideDB(CommCareCase, sourcedb):
case_ids = get_case_ids_in_domain_by_owner(domain.name,
owner_id__in=owners)
xform_ids = set()
print 'copying %s cases' % len(case_ids)
for i, subset in enumerate(chunked(case_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
cases = [
CommCareCase.wrap(case['doc']) for case in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)
]
for case in cases:
def handle(self, *args, **options):
if len(args) < 2 or len(args) > 3:
raise CommandError('Usage is copy_doc %s' % self.args)
sourcedb = Database(args[0])
app_id = args[1]
domain = args[2] if len(args) == 3 else None
app_json = sourcedb.get(app_id)
if domain:
app_json['domain'] = domain
dt = DocumentTransform(app_json, sourcedb)
save(dt, get_db())
def run(self):
for doc, count in iter(self.queue.get, None):
try:
if self.exclude_types and doc["doc_type"] in self.exclude_types:
print " SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, self.total, doc["doc_type"], doc["_id"])
else:
if not self.simulate:
dt = DocumentTransform(doc, self.sourcedb)
save(dt, self.targetdb)
print " Synced %s/%s docs (%s: %s)" % (count, self.total, doc["doc_type"], doc["_id"])
except Exception, e:
print " Document %s failed! Error is: %s" % (doc["_id"], e)
def copy_domain(self, sourcedb, domain):
print "Copying domain doc"
result = sourcedb.view("domain/domains",
key=domain,
reduce=False,
include_docs=True).first()
if result and 'doc' in result:
domain_doc = Domain.wrap(result['doc'])
dt = DocumentTransform(domain_doc, sourcedb)
save(dt, self.targetdb)
else:
print "Domain doc not found for domain %s." % domain
def run(self):
for doc, count in iter(self.queue.get, None):
try:
if self.exclude_types and doc["doc_type"] in self.exclude_types:
print " SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, self.total, doc["doc_type"], doc["_id"])
else:
if not self.simulate:
dt = DocumentTransform(doc, self.sourcedb)
save(dt, self.targetdb)
print " Synced %s/%s docs (%s: %s)" % (count, self.total, doc["doc_type"], doc["_id"])
except Exception, e:
print " Document %s failed! Error is: %s" % (doc["_id"], e)
def copy_domain(self, source_couch, domain):
print "Copying domain doc"
sourcedb = source_couch.get_db_for_class(Domain)
result = sourcedb.view("domain/domains",
key=domain,
reduce=False,
include_docs=True).first()
if result and 'doc' in result:
domain_doc = Domain.wrap(result['doc'])
dt = DocumentTransform(domain_doc._obj, sourcedb)
save(dt, self.targetdb.get_db_for_doc_type(domain_doc['doc_type']))
else:
print "Domain doc not found for domain %s." % domain
def copy_doc(doc, count, sourcedb, target_couch, exclude_types, total, simulate, exclude_attachments):
if exclude_types and doc["doc_type"] in exclude_types:
print(" SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, total, doc["doc_type"], doc["_id"]))
else:
if not simulate:
dt = DocumentTransform(doc, sourcedb, exclude_attachments)
for i in reversed(list(range(5))):
try:
save(dt, target_couch)
break
except (ResourceConflict, ParserError, TypeError):
if i == 0:
raise
print(" Synced %s/%s docs (%s: %s)" % (count, total, doc["doc_type"], doc["_id"]))
def copy_doc(doc, count, sourcedb, target_couch, exclude_types, total, simulate, exclude_attachments):
if exclude_types and doc["doc_type"] in exclude_types:
print(" SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, total, doc["doc_type"], doc["_id"]))
else:
if not simulate:
dt = DocumentTransform(doc, sourcedb, exclude_attachments)
for i in reversed(list(range(5))):
try:
save(dt, target_couch)
break
except (ResourceConflict, TypeError):
if i == 0:
raise
print(" Synced %s/%s docs (%s: %s)" % (count, total, doc["doc_type"], doc["_id"]))
def copy_domain(self, source_couch, domain):
print "Copying domain doc"
sourcedb = source_couch.get_db_for_class(Domain)
result = sourcedb.view(
"domain/domains",
key=domain,
reduce=False,
include_docs=True
).first()
if result and 'doc' in result:
domain_doc = Domain.wrap(result['doc'])
dt = DocumentTransform(domain_doc._obj, sourcedb)
save(dt, self.targetdb.get_db_for_doc_type(domain_doc['doc_type']))
else:
print "Domain doc not found for domain %s." % domain
def handle(self, sourcedb, doc_ids_or_file, domain, **options):
sourcedb = Database(sourcedb)
if os.path.isfile(doc_ids_or_file):
with open(doc_ids_or_file) as f:
doc_ids = f.read().splitlines()
else:
doc_ids = doc_ids_or_file.split(',')
print("Starting copy of {} docs".format(len(doc_ids)))
for doc_id in doc_ids:
print('Copying doc: {}'.format(doc_id))
doc_json = sourcedb.get(doc_id)
if domain:
doc_json['domain'] = domain
dt = DocumentTransform(doc_json, sourcedb)
save(dt, get_db())
def handle(self, *args, **options):
if len(args) != 2:
raise CommandError('Usage is copy_group_data %s' % self.args)
sourcedb = Database(args[0])
group_id = args[1]
exclude_user_owned = options["exclude_user_owned"]
print 'getting group'
group = Group.wrap(sourcedb.get(group_id))
group.save(force_update=True)
print 'getting domain'
domain = Domain.wrap(
sourcedb.view('domain/domains', key=group.domain, include_docs=True,
reduce=False, limit=1).one()['doc']
)
dt = DocumentTransform(domain._obj, sourcedb)
save(dt, Domain.get_db())
owners = [group_id]
if not exclude_user_owned:
owners.extend(group.users)
print 'getting case ids'
with OverrideDB(CommCareCase, sourcedb):
case_ids = get_case_ids_in_domain_by_owner(
domain.name, owner_id__in=owners)
xform_ids = set()
print 'copying %s cases' % len(case_ids)
for i, subset in enumerate(chunked(case_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
cases = [CommCareCase.wrap(case['doc']) for case in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
for case in cases:
def copy_docs(self, sourcedb, domain, startkey, endkey, simulate, type=None, since=None):
doc_ids = [result["id"] for result in sourcedb.view("domain/docs", startkey=startkey,
endkey=endkey, reduce=False)]
total = len(doc_ids)
count = 0
targetdb = get_db()
msg = "Found %s matching documents in domain: %s" % (total, domain)
msg += " of type: %s" % (type) if type else ""
msg += " since: %s" % (since) if since else ""
print msg
for doc in iter_docs(sourcedb, doc_ids):
try:
count += 1
if not simulate:
dt = DocumentTransform(doc, sourcedb)
save(dt, targetdb)
print " Synced %s/%s docs (%s: %s)" % (count, total, doc["doc_type"], doc["_id"])
except Exception, e:
print " Document %s failed! Error is: %s" % (doc["_id"], e)
def handle(self, *args, **options):
if len(args) < 2 or len(args) > 3:
raise CommandError('Usage is copy_doc %s' % self.args)
sourcedb = Database(args[0])
doc_ids_or_file = args[1]
domain = args[2] if len(args) == 3 else None
if os.path.isfile(doc_ids_or_file):
with open(doc_ids_or_file) as f:
doc_ids = f.read().splitlines()
else:
doc_ids = doc_ids_or_file.split(',')
print "Starting copy of {} docs".format(len(doc_ids))
for doc_id in doc_ids:
print 'Copying doc: {}'.format(doc_id)
doc_json = sourcedb.get(doc_id)
if domain:
doc_json['domain'] = domain
dt = DocumentTransform(doc_json, sourcedb)
save(dt, get_db())
def handle(self, *args, **options):
if len(args) < 2 or len(args) > 3:
raise CommandError('Usage is copy_doc %s' % self.args)
sourcedb = Database(args[0])
doc_ids_or_file = args[1]
domain = args[2] if len(args) == 3 else None
if os.path.isfile(doc_ids_or_file):
with open(doc_ids_or_file) as f:
doc_ids = f.read().splitlines()
else:
doc_ids = doc_ids_or_file.split(',')
print("Starting copy of {} docs".format(len(doc_ids)))
for doc_id in doc_ids:
print('Copying doc: {}'.format(doc_id))
doc_json = sourcedb.get(doc_id)
if domain:
doc_json['domain'] = domain
dt = DocumentTransform(doc_json, sourcedb)
save(dt, get_db())
def copy_doc(doc, count, sourcedb, targetdb, exclude_types, total, simulate, exclude_attachments):
if exclude_types and doc["doc_type"] in exclude_types:
print " SKIPPED (excluded type: %s). Synced %s/%s docs (%s: %s)" % \
(doc["doc_type"], count, total, doc["doc_type"], doc["_id"])
else:
if not simulate:
for i in reversed(range(5)):
try:
dt = DocumentTransform(doc, sourcedb, exclude_attachments)
break
except RequestError as r:
if i == 0:
_soft_assert(False, 'Copy domain failed after 5 tries with {}'.format(r))
raise
for i in reversed(range(5)):
try:
save(dt, targetdb)
break
except (ResourceConflict, ParserError, TypeError) as e:
if i == 0:
_soft_assert(False, 'Copy domain failed after 5 tries with {}'.format(e))
raise
print " Synced %s/%s docs (%s: %s)" % (count, total, doc["doc_type"], doc["_id"])
def handle(self, *args, **options):
if len(args) != 2:
raise CommandError('Usage is copy_domain %s' % self.args)
sourcedb = Database(args[0])
domain = args[1].strip()
all_docs = sourcedb.view("domain/docs", startkey=[domain],
endkey=[domain, {}], reduce=False)
total = len(all_docs)
count = 0
targetdb = get_db()
print "found %s matching documents in domain: %s" % (total, domain)
for row in all_docs:
try:
count += 1
dt = DocumentTransform(sourcedb.get(row["id"]), sourcedb)
save(dt, targetdb)
print "Synced %s/%s docs (%s: %s)" % (count, total, row["key"][1], row["id"])
except Exception, e:
print "Document %s failed! Error is: %s" % (row["id"], e)
def handle(self, *args, **options):
raise CommandError(
'copy_group_data is currently broken. '
'Ask Danny or Ethan to fix it along the lines of '
'https://github.com/dimagi/commcare-hq/pull/9180/files#diff-9d976dc051a36a028c6604581dfbce5dR95'
)
if len(args) != 2:
raise CommandError('Usage is copy_group_data %s' % self.args)
sourcedb = Database(args[0])
group_id = args[1]
exclude_user_owned = options["exclude_user_owned"]
print 'getting group'
group = Group.wrap(sourcedb.get(group_id))
group.save(force_update=True)
print 'getting domain'
domain = Domain.wrap(
sourcedb.view('domain/domains', key=group.domain, include_docs=True,
reduce=False, limit=1).one()['doc']
)
dt = DocumentTransform(domain._obj, sourcedb)
save(dt, Domain.get_db())
owners = [group_id]
if not exclude_user_owned:
owners.extend(group.users)
print 'getting case ids'
with OverrideDB(CommCareCase, sourcedb):
case_ids = get_case_ids_in_domain_by_owner(
domain.name, owner_id__in=owners)
xform_ids = set()
print 'copying %s cases' % len(case_ids)
for i, subset in enumerate(chunked(case_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
cases = [CommCareCase.wrap(case['doc']) for case in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
for case in cases:
xform_ids.update(case.xform_ids)
self.lenient_bulk_save(CommCareCase, cases)
if not exclude_user_owned:
# also grab submissions that may not have included any case data
for user_id in group.users:
xform_ids.update(res['id'] for res in sourcedb.view(
'all_forms/view',
startkey=['submission user', domain.name, user_id],
endkey=['submission user', domain.name, user_id, {}],
reduce=False
))
print 'copying %s xforms' % len(xform_ids)
user_ids = set(group.users)
def form_wrapper(row):
doc = row['doc']
doc.pop('_attachments', None)
doc.pop('external_blobs', None)
return XFormInstance.wrap(doc)
for i, subset in enumerate(chunked(xform_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
xforms = sourcedb.all_docs(
keys=list(subset),
include_docs=True,
wrapper=form_wrapper,
).all()
self.lenient_bulk_save(XFormInstance, xforms)
for xform in xforms:
user_id = xform.metadata.userID
user_ids.add(user_id)
print 'copying %s users' % len(user_ids)
def wrap_user(row):
try:
doc = row['doc']
except KeyError:
logging.exception('trouble with user result %r' % row)
return None
try:
return CouchUser.wrap_correctly(doc)
except Exception:
logging.exception('trouble with user %s' % doc['_id'])
return None
users = sourcedb.all_docs(
keys=list(user_ids),
include_docs=True,
wrapper=wrap_user,
).all()
role_ids = set([])
for user in filter(lambda u: u is not None, users):
# if we use bulk save, django user doesn't get sync'd
domain_membership = user.get_domain_membership(domain.name)
if domain_membership and domain_membership.role_id:
role_ids.add(user.domain_membership.role_id)
user.save(force_update=True)
print 'copying %s roles' % len(role_ids)
for i, subset in enumerate(chunked(role_ids, CHUNK_SIZE)):
roles = [UserRole.wrap(role['doc']) for role in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
self.lenient_bulk_save(UserRole, roles)
if options['include_sync_logs']:
print 'copying sync logs'
for user_id in user_ids:
log_ids = [res['id'] for res in sourcedb.view("phone/sync_logs_by_user",
startkey=[user_id, {}],
endkey=[user_id],
descending=True,
reduce=False,
include_docs=True
)]
print 'user: %s, logs: %s' % (user_id, len(log_ids))
for i, subset in enumerate(chunked(log_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
logs = [SyncLog.wrap(log['doc']) for log in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
self.lenient_bulk_save(SyncLog, logs)
def handle(self, *args, **options):
if len(args) != 2:
raise CommandError('Usage is copy_group_data %s' % self.args)
sourcedb = Database(args[0])
group_id = args[1]
exclude_user_owned = options["exclude_user_owned"]
print 'getting group'
group = Group.wrap(sourcedb.get(group_id))
group.save(force_update=True)
print 'getting domain'
domain = Domain.wrap(
sourcedb.view('domain/domains', key=group.domain, include_docs=True,
reduce=False, limit=1).one()['doc']
)
dt = DocumentTransform(domain._obj, sourcedb)
save(dt, Domain.get_db())
owners = [group_id]
if not exclude_user_owned:
owners.extend(group.users)
def keys_for_owner(domain, owner_id):
return [
[domain, owner_id, False],
[domain, owner_id, True],
]
def get_case_ids(owners):
keys = list(itertools.chain(*[keys_for_owner(domain.name, owner_id) for owner_id in owners]))
results = sourcedb.view(
'hqcase/by_owner',
keys=keys,
reduce=False,
include_docs=False,
)
return [res['id'] for res in results]
print 'getting case ids'
case_ids = get_case_ids(owners)
xform_ids = set()
print 'copying %s cases' % len(case_ids)
for i, subset in enumerate(chunked(case_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
cases = [CommCareCase.wrap(case['doc']) for case in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
for case in cases:
xform_ids.update(case.xform_ids)
self.lenient_bulk_save(CommCareCase, cases)
if not exclude_user_owned:
# also grab submissions that may not have included any case data
for user_id in group.users:
xform_ids.update(res['id'] for res in sourcedb.view(
'reports_forms/all_forms',
startkey=['submission user', domain.name, user_id],
endkey=['submission user', domain.name, user_id, {}],
reduce=False
))
print 'copying %s xforms' % len(xform_ids)
user_ids = set(group.users)
def form_wrapper(row):
doc = row['doc']
doc.pop('_attachments', None)
return XFormInstance.wrap(doc)
for i, subset in enumerate(chunked(xform_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
xforms = sourcedb.all_docs(
keys=list(subset),
include_docs=True,
wrapper=form_wrapper,
).all()
self.lenient_bulk_save(XFormInstance, xforms)
for xform in xforms:
user_id = xform.metadata.userID
user_ids.add(user_id)
print 'copying %s users' % len(user_ids)
def wrap_user(row):
try:
doc = row['doc']
except KeyError:
logging.exception('trouble with user result %r' % row)
return None
try:
return CouchUser.wrap_correctly(doc)
except Exception:
logging.exception('trouble with user %s' % doc['_id'])
return None
users = sourcedb.all_docs(
keys=list(user_ids),
include_docs=True,
wrapper=wrap_user,
).all()
role_ids = set([])
for user in filter(lambda u: u is not None, users):
# if we use bulk save, django user doesn't get sync'd
if user.get_domain_membership(domain.name).role_id:
role_ids.add(user.domain_membership.role_id)
user.save(force_update=True)
print 'copying %s roles' % len(role_ids)
for i, subset in enumerate(chunked(role_ids, CHUNK_SIZE)):
roles = [UserRole.wrap(role['doc']) for role in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
self.lenient_bulk_save(UserRole, roles)
if options['include_sync_logs']:
print 'copying sync logs'
for user_id in user_ids:
log_ids = [res['id'] for res in sourcedb.view("phone/sync_logs_by_user",
startkey=[user_id, {}],
endkey=[user_id],
descending=True,
reduce=False,
include_docs=True
)]
print 'user: %s, logs: %s' % (user_id, len(log_ids))
for i, subset in enumerate(chunked(log_ids, CHUNK_SIZE)):
print i * CHUNK_SIZE
logs = [SyncLog.wrap(log['doc']) for log in sourcedb.all_docs(
keys=list(subset),
include_docs=True,
)]
self.lenient_bulk_save(SyncLog, logs)
