def update_document(host, port, url, message_dict):
"""
Save document to core api
"""
document_uuid = message_dict['document_uuid']
url = 'http://%s:%d/%s/%s/' % (host, port, url, document_uuid)
headers = {'content-type': 'application/json; charset=utf-8'}
try:
message = {
'is_document_processed': True,
'coverImage': message_dict['cover_image'],
'previewImageWidth': message_dict['preview_image_width'],
'previewImageHeight': message_dict['preview_image_height'],
'fileType': 'pdf',
'pageCount': message_dict['metadata'].get('number_of_pages', 0),
'toc': message_dict['metadata'].get('toc', []),
'document_file_original': message_dict['original_document'],
}
response = requests.put(
url,
json=message,
headers=headers,
)
if response.status_code != 200:
sentry_client('API Error: CODE %d, CONTENT: %s' %
(response.status_code, str(response.content)))
except Exception as e:
sentry_client(e)
raise e
def save_to_seaweedfs(message):
host = CONFIG['seaweed_filler_host']
port = CONFIG['seaweed_filler_port']
try:
images_list = message['preview_images']
_dir = message['document_uuid']
original_document = message['original_document']
document_name = message['document_name']
result_list = save_files_filler_batch(host, port, _dir, images_list)
_splited = os.path.split(original_document)
_ex = _splited[1].split('.')[1]
document_file = '%s.%s' % (document_name, _ex)
new_original_document = os.path.join(_splited[0], document_file)
os.rename(original_document, new_original_document)
save_file_to_filler(host, port, _dir, new_original_document)
message['cover_image'] = result_list[0]
width, height = get_image_size(images_list[0])
message['preview_image_width'] = width
message['preview_image_height'] = height
message['original_document'] = document_file
# TODO: add /tmp to conf
shutil.rmtree(os.path.join('/tmp', _dir))
del message['preview_images']
return message
except Exception as e:
sentry_client(e)
def read_metadata(pdf_path, document_uuid, document_name):
try:
metadata_dict = get_metadata(pdf_path)
return dict(
original_document=pdf_path,
metadata=metadata_dict,
document_uuid=document_uuid,
document_name=document_name,
)
except Exception as e:
sentry_client(e)
def save_file(host, port, file_path):
"""
Save file to Seaweed FS
:param host: str, master host
:param port: int, master post
:param file_path: str, path to file
:return: str, file fid
"""
files = {'upload_file': open(file_path, 'rb')}
url = 'http://%s:%d/submit' % (host, port)
try:
response = requests.post(url, files=files)
return response.json()['fid']
except Exception as e:
sentry_client(e)
raise e
def split_document(message):
_path = message['original_document']
_splited_path = 'splited'
_sizes = dict(
large=CONFIG['preview_large'],
normal=CONFIG['preview_normal'],
small=CONFIG['preview_small']
)
_format = CONFIG['preview_format']
try:
# TODO: Add document_name to preview images
preview_images = split_pdf(_path, _splited_path, _sizes, _format)
message.update({'preview_images': preview_images})
return message
except Exception as e:
sentry_client(e)