From 3d4131a193308698e129c4560d4b8a3dfe8782d5 Mon Sep 17 00:00:00 2001 From: Terrtia Date: Thu, 30 Apr 2020 09:54:26 +0200 Subject: [PATCH] chg: [importer] add url_extract importer --- .../Ail_feeder_url_extract.py | 43 +++++++++++++++++++ 1 file changed, 43 insertions(+) create mode 100755 bin/import/ail_json_importer/Ail_feeder_url_extract.py diff --git a/bin/import/ail_json_importer/Ail_feeder_url_extract.py b/bin/import/ail_json_importer/Ail_feeder_url_extract.py new file mode 100755 index 00000000..a71cd93d --- /dev/null +++ b/bin/import/ail_json_importer/Ail_feeder_url_extract.py @@ -0,0 +1,43 @@ +#!/usr/bin/env python3 +# -*-coding:UTF-8 -* +""" +The JSON Receiver Module +================ + +Recieve Json Items (example: Twitter feeder) + +""" +import os +import json +import sys +import datetime + +# sys.path.append(os.path.join(os.environ['AIL_BIN'], 'lib')) +# import item_basic + +sys.path.append(os.path.join(os.environ['AIL_BIN'], 'import', 'ail_json_importer')) +from Default_json import Default_json + +class Ail_feeder_twitter(Default_json): + """Twitter Feeder functions""" + + def __init__(self, name, json_item): + super().__init__(name, json_item) + + def get_feeder_name(self): + return 'urlextract' + + # define item id + def get_item_id(self): + # use twitter timestamp ? + item_date = datetime.date.today().strftime("%Y/%m/%d") + item_id = str(self.json_item['meta']['twitter:url-extracted']) + item_id = '{}{}.gz'.format(file_name_paste[:215], str(uuid.uuid4())) + return os.path.join('urlextract', item_date, item_id) + '.gz' + + # # TODO: + def process_json_meta(self, process): + ''' + Process JSON meta filed. + ''' + return None