Add column mapping support

Add support for specifying metadata columns.

Signed-off-by: Roberto Rosario <roberto.rosario.gonzalez@gmail.com>
This commit is contained in:
Roberto Rosario
2019-06-19 17:47:32 -04:00
parent b3d59eee39
commit bc072f7b7e
4 changed files with 181 additions and 14 deletions

View File

@@ -7,18 +7,34 @@ from django.apps import apps
from django.core import management
from django.core.files import File
from mayan.apps.documents.tasks import task_upload_new_document
from ...tasks import task_upload_new_document
class Command(management.BaseCommand):
help = 'Import documents from a CSV file.'
def add_arguments(self, parser):
#parser.add_argument(
# '-l', '--link',
# action='store_true', dest='link', default=False,
# help='Create a symbolic link to each file instead of copying.',
#)
parser.add_argument(
'--document_type_column',
action='store', dest='document_type_column', default=0,
help='Column that contains the document type labels. Column '
'numbers start at 0.',
type=int
)
parser.add_argument(
'--document_path_column',
action='store', dest='document_path_column', default=1,
help='Column that contains the path to the document files. Column '
'numbers start at 0.',
type=int
)
parser.add_argument(
'--metadata_pairs_column',
action='store', dest='metadata_pairs_column',
help='Column that contains metadata name and values for the '
'documents. Use the form: <label column>:<value column>. Example: '
'2:5. Separate multiple pairs with commas. Example: 2:5,7:10',
)
parser.add_argument('filelist', nargs='?', help='File list')
def handle(self, *args, **options):
@@ -41,24 +57,44 @@ class Command(management.BaseCommand):
with open(options['filelist']) as csv_datafile:
csv_reader = csv.reader(csv_datafile)
for row in csv_reader:
with open(row[1]) as file_object:
if row[0] not in document_types:
self.stdout.write('New document type: {}. Creating and caching.'.format(row[0]))
document_type, created = DocumentType.objects.get_or_create(
label=row[0]
with open(row[options['document_path_column']]) as file_object:
document_type_label = row[options['document_type_column']]
if document_type_label not in document_types:
self.stdout.write(
'New document type: {}. Creating and caching.'.format(
document_type_label
)
)
document_types[row[0]] = document_type
document_type, created = DocumentType.objects.get_or_create(
label=document_type_label
)
document_types[document_type_label] = document_type
else:
document_type = document_types[row[0]]
document_type = document_types[document_type_label]
shared_uploaded_file = SharedUploadedFile.objects.create(
file=File(file_object)
)
extra_data = {}
if options['metadata_pairs_column']:
extra_data['metadata_pairs'] = []
for pair in options['metadata_pairs_column'].split(','):
name, value = pair.split(':')
extra_data['metadata_pairs'].append(
{
'name': row[int(name)],
'value': row[int(value)]
}
)
task_upload_new_document.apply_async(
kwargs=dict(
document_type_id=document_type.pk,
shared_uploaded_file_id=shared_uploaded_file.pk,
extra_data=extra_data
)
)

View File

@@ -0,0 +1,10 @@
from __future__ import absolute_import, unicode_literals
from django.utils.translation import ugettext_lazy as _
from mayan.apps.documents.queues import queue_uploads
queue_uploads.add_task_type(
dotted_path='mayan.apps.importer.tasks.task_upload_new_document',
label=_('Import new document')
)

View File

@@ -0,0 +1,85 @@
from __future__ import unicode_literals
import logging
from django.apps import apps
from django.db import OperationalError
from django.utils.text import slugify
from mayan.celery import app
from mayan.apps.documents.literals import UPLOAD_NEW_DOCUMENT_RETRY_DELAY
logger = logging.getLogger(__name__)
@app.task(bind=True, default_retry_delay=UPLOAD_NEW_DOCUMENT_RETRY_DELAY, ignore_result=True)
def task_upload_new_document(self, document_type_id, shared_uploaded_file_id, extra_data=None):
DocumentType = apps.get_model(
app_label='documents', model_name='DocumentType'
)
MetadataType = apps.get_model(
app_label='metadata', model_name='MetadataType'
)
SharedUploadedFile = apps.get_model(
app_label='common', model_name='SharedUploadedFile'
)
try:
document_type = DocumentType.objects.get(pk=document_type_id)
shared_file = SharedUploadedFile.objects.get(
pk=shared_uploaded_file_id
)
except OperationalError as exception:
logger.warning(
'Operational error during attempt to retrieve shared data for '
'new document of type: %s; %s. Retrying.', document_type, exception
)
raise self.retry(exc=exception)
try:
with shared_file.open() as file_object:
new_document = document_type.new_document(file_object=file_object)
except OperationalError as exception:
logger.warning(
'Operational error during attempt to create new document '
'of type: %s; %s. Retrying.', document_type, exception
)
raise self.retry(exc=exception)
except Exception as exception:
# This except and else block emulate a finally:
logger.error(
'Unexpected error during attempt to create new document '
'of type: %s; %s', document_type, exception
)
try:
shared_file.delete()
except OperationalError as exception:
logger.warning(
'Operational error during attempt to delete shared '
'file: %s; %s.', shared_file, exception
)
else:
if extra_data:
for pair in extra_data.get('metadata_pairs', []):
name = slugify(pair['name']).replace('-', '_')
metadata_type, created = MetadataType.objects.get_or_create(
label=pair['name'], defaults={'name': name}
)
if created:
new_document.document_type.metadata.create(
metadata_type=metadata_type, required=False
)
new_document.metadata.create(
metadata_type=metadata_type, value=pair['value']
)
try:
shared_file.delete()
except OperationalError as exception:
logger.warning(
'Operational error during attempt to delete shared '
'file: %s; %s.', shared_file, exception
)

View File

@@ -41,7 +41,12 @@ class ImportManagementCommandTestCase(GenericDocumentTestCase):
)
for times in range(self.test_import_count):
filewriter.writerow(
[self.test_document_type.label, TEST_SMALL_DOCUMENT_PATH]
[
self.test_document_type.label, TEST_SMALL_DOCUMENT_PATH,
'column 2', 'column 3', 'column 4', 'column 5',
'column 6', 'column 7', 'column 8', 'column 9',
'column 10', 'column 11',
]
)
def _destroy_test_csv_file(self):
@@ -56,3 +61,34 @@ class ImportManagementCommandTestCase(GenericDocumentTestCase):
self.assertTrue(DocumentType.objects.count() > 0)
self.assertTrue(Document.objects.count() > 0)
def test_import_document_type_column_mapping(self):
self.test_document_type.delete()
management.call_command(
'import', self.test_csv_path, '--document_type_column', '2'
)
self.assertTrue(DocumentType.objects.first().label == 'column 2')
self.assertTrue(Document.objects.count() > 0)
def test_import_document_path_column_mapping(self):
self.test_document_type.delete()
with self.assertRaises(IOError):
management.call_command(
'import', self.test_csv_path, '--document_path_column', '2'
)
def test_import_metadata_column_mapping(self):
self.test_document_type.delete()
management.call_command(
'import', self.test_csv_path, '--metadata_pairs_column', '2:3,4:5',
)
self.assertTrue(DocumentType.objects.count() > 0)
self.assertTrue(Document.objects.count() > 0)
self.assertTrue(Document.objects.first().metadata.count() > 0)
self.assertEqual(
Document.objects.first().metadata.get(
metadata_type__name='column_2'
).value, 'column 3'
)