brynq-sdk-task-scheduler 1.0.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,10 @@
1
+ Metadata-Version: 1.0
2
+ Name: brynq_sdk_task_scheduler
3
+ Version: 1.0.0
4
+ Summary: Code to execute tasks in BrynQ.com with the task scheduler
5
+ Home-page: UNKNOWN
6
+ Author: BrynQ
7
+ Author-email: support@brynq.com
8
+ License: BrynQ License
9
+ Description: Code to execute tasks in the BrynQ.com platform with the task scheduler
10
+ Platform: UNKNOWN
@@ -0,0 +1 @@
1
+ from brynq_sdk.task_scheduler.task_scheduler import TaskScheduler
@@ -0,0 +1,555 @@
1
+ import sys
2
+ import os
3
+ import datetime
4
+ import inspect
5
+ import time
6
+ import typing
7
+ import traceback
8
+ import pandas as pd
9
+ import json
10
+ import requests
11
+ from salure_helpers.mandrill import MailClient
12
+ from brynq_sdk.functions import Functions
13
+ from brynq_sdk.brynq import BrynQ
14
+ from salure_helpers.mysql import MySQL
15
+ from salure_helpers.elastic import Elastic
16
+ import warnings
17
+ import re
18
+
19
+
20
+ class TaskScheduler(BrynQ):
21
+
22
+ def __init__(self, task_id: int = None, loglevel: str = 'INFO', email_after_errors: bool = False):
23
+ """
24
+ The TaskScheduler is responsible for the logging to the database. Based on this logging, the next reload will
25
+ start or not and warning will be given or not
26
+ :param task_id: The ID from the task as saved in the task_scheduler table in the customer database
27
+ :param email_after_errors: a True or False value. When True, there will be send an email to a contactperson of the customer (as given in the database) with the number of errors
28
+ :param loglevel: Chose on which level you want to store the logs. Default is INFO. that means that a logline
29
+ with level DEBUG not is stored
30
+ """
31
+ super().__init__()
32
+ self.es = Elastic()
33
+ self.mysql = MySQL()
34
+ self.email_after_errors = email_after_errors
35
+ self.task_id = task_id
36
+ self.loglevel = loglevel
37
+ self.started_at = datetime.datetime.now()
38
+ # If the task is started via the task_scheduler, the following 3 parameters will be passed by the scheduler
39
+ if len(sys.argv[1:4]) > 0:
40
+ self.started_local = False
41
+ self.customer_db, self.task_id, self.run_id = sys.argv[1:4]
42
+ # If the task is started locally, the parameters should be set locally
43
+ else:
44
+ self.started_local = True
45
+ self.customer_db = 'placeholder'
46
+ self.run_id = int(round(time.time() * 100000))
47
+ print(self.task_id, self.run_id)
48
+ self.error_count = 0
49
+
50
+ # Check if the log tables exists in the customer database. If not, create them
51
+ # Mysql throws a warning when a table already exists. We don't care so we ignore warnings. (not exceptions!)
52
+ warnings.filterwarnings('ignore')
53
+ # self.check_if_logging_tables_exists()
54
+
55
+ # Check if the task is started on schedule or manual. store in a variable to use later in the script
56
+ self.task_manual_started = self.check_if_task_manual_started()
57
+
58
+ # Creates Elasticsearch index and data view if not exists
59
+ self.es.initialize_customer()
60
+
61
+ # Start the task and setup the data in the database
62
+ self.start_task()
63
+
64
+ def __count_keys(self, json_obj):
65
+ if not isinstance(json_obj, dict):
66
+ return 0
67
+ key_count = 0
68
+ for key, value in json_obj.items():
69
+ if not isinstance(value, dict):
70
+ key_count += 1 # Count the current key
71
+ else:
72
+ key_count += self.__count_keys(value) # Recursively count keys in nested dictionaries
73
+ return key_count
74
+
75
+ def check_if_logging_tables_exists(self):
76
+ """
77
+ This function checks if all the needed tables for the task_scheduler exists. If they don't, this function
78
+ creates the needed tables
79
+ :return: nothing
80
+ """
81
+ # Check if the table task_scheduler exists. If not, create it
82
+ new_table_query = 'CREATE TABLE IF NOT EXISTS `task_scheduler` (' \
83
+ '`id` int(11) NOT NULL AUTO_INCREMENT,' \
84
+ '`dashboard_reload` bool NOT NULL DEFAULT \'0\',' \
85
+ '`title` varchar(50) NOT NULL,' \
86
+ '`description` varchar(255) NOT NULL,' \
87
+ '`dashboard_guid` varchar(255) NULL DEFAULT NULL,' \
88
+ '`docker_image` varchar(255) DEFAULT NULL,' \
89
+ '`runfile_path` varchar(255) DEFAULT NULL,' \
90
+ '`trigger_type` enum("MANUAL", "TIME", "OTHER_TASK") NOT NULL DEFAULT \'MANUAL\',' \
91
+ '`next_reload` timestamp NULL DEFAULT NULL,' \
92
+ '`timezone` enum("Africa/Abidjan", "Africa/Accra", "Africa/Addis_Ababa", "Africa/Algiers", "Africa/Asmara", "Africa/Bamako", "Africa/Bangui", "Africa/Banjul", "Africa/Bissau", "Africa/Blantyre", "Africa/Brazzaville", "Africa/Bujumbura", "Africa/Cairo", "Africa/Casablanca", "Africa/Ceuta", "Africa/Conakry", "Africa/Dakar", "Africa/Dar_es_Salaam", "Africa/Djibouti", "Africa/Douala", "Africa/El_Aaiun", "Africa/Freetown", "Africa/Gaborone", "Africa/Harare", "Africa/Johannesburg", "Africa/Juba", "Africa/Kampala", "Africa/Khartoum", "Africa/Kigali", "Africa/Kinshasa", "Africa/Lagos", "Africa/Libreville", "Africa/Lome", "Africa/Luanda", "Africa/Lubumbashi", "Africa/Lusaka", "Africa/Malabo", "Africa/Maputo", "Africa/Maseru", "Africa/Mbabane", "Africa/Mogadishu", "Africa/Monrovia", "Africa/Nairobi", "Africa/Ndjamena", "Africa/Niamey", "Africa/Nouakchott", "Africa/Ouagadougou", "Africa/Porto-Novo", "Africa/Sao_Tome", "Africa/Tripoli", "Africa/Tunis", "Africa/Windhoek", "America/Adak", "America/Anchorage", "America/Anguilla", "America/Antigua", "America/Araguaina", "America/Argentina/Buenos_Aires", "America/Argentina/Catamarca", "America/Argentina/Cordoba", "America/Argentina/Jujuy", "America/Argentina/La_Rioja", "America/Argentina/Mendoza", "America/Argentina/Rio_Gallegos", "America/Argentina/Salta", "America/Argentina/San_Juan", "America/Argentina/San_Luis", "America/Argentina/Tucuman", "America/Argentina/Ushuaia", "America/Aruba", "America/Asuncion", "America/Atikokan", "America/Bahia", "America/Bahia_Banderas", "America/Barbados", "America/Belem", "America/Belize", "America/Blanc-Sablon", "America/Boa_Vista", "America/Bogota", "America/Boise", "America/Cambridge_Bay", "America/Campo_Grande", "America/Cancun", "America/Caracas", "America/Cayenne", "America/Cayman", "America/Chicago", "America/Chihuahua", "America/Costa_Rica", "America/Creston", "America/Cuiaba", "America/Curacao", "America/Danmarkshavn", "America/Dawson", "America/Dawson_Creek", "America/Denver", "America/Detroit", "America/Dominica", "America/Edmonton", "America/Eirunepe", "America/El_Salvador", "America/Fort_Nelson", "America/Fortaleza", "America/Glace_Bay", "America/Godthab", "America/Goose_Bay", "America/Grand_Turk", "America/Grenada", "America/Guadeloupe", "America/Guatemala", "America/Guayaquil", "America/Guyana", "America/Halifax", "America/Havana", "America/Hermosillo", "America/Indiana/Indianapolis", "America/Indiana/Knox", "America/Indiana/Marengo", "America/Indiana/Petersburg", "America/Indiana/Tell_City", "America/Indiana/Vevay", "America/Indiana/Vincennes", "America/Indiana/Winamac", "America/Inuvik", "America/Iqaluit", "America/Jamaica", "America/Juneau", "America/Kentucky/Louisville", "America/Kentucky/Monticello", "America/Kralendijk", "America/La_Paz", "America/Lima", "America/Los_Angeles", "America/Lower_Princes", "America/Maceio", "America/Managua", "America/Manaus", "America/Marigot", "America/Martinique", "America/Matamoros", "America/Mazatlan", "America/Menominee", "America/Merida", "America/Metlakatla", "America/Mexico_City", "America/Miquelon", "America/Moncton", "America/Monterrey", "America/Montevideo", "America/Montserrat", "America/Nassau", "America/New_York", "America/Nipigon", "America/Nome", "America/Noronha", "America/North_Dakota/Beulah", "America/North_Dakota/Center", "America/North_Dakota/New_Salem", "America/Ojinaga", "America/Panama", "America/Pangnirtung", "America/Paramaribo", "America/Phoenix", "America/Port-au-Prince", "America/Port_of_Spain", "America/Porto_Velho", "America/Puerto_Rico", "America/Punta_Arenas", "America/Rainy_River", "America/Rankin_Inlet", "America/Recife", "America/Regina", "America/Resolute", "America/Rio_Branco", "America/Santarem", "America/Santiago", "America/Santo_Domingo", "America/Sao_Paulo", "America/Scoresbysund", "America/Sitka", "America/St_Barthelemy", "America/St_Johns", "America/St_Kitts", "America/St_Lucia", "America/St_Thomas", "America/St_Vincent", "America/Swift_Current", "America/Tegucigalpa", "America/Thule", "America/Thunder_Bay", "America/Tijuana", "America/Toronto", "America/Tortola", "America/Vancouver", "America/Whitehorse", "America/Winnipeg", "America/Yakutat", "America/Yellowknife", "Antarctica/Casey", "Antarctica/Davis", "Antarctica/DumontDUrville", "Antarctica/Macquarie", "Antarctica/Mawson", "Antarctica/McMurdo", "Antarctica/Palmer", "Antarctica/Rothera", "Antarctica/Syowa", "Antarctica/Troll", "Antarctica/Vostok", "Arctic/Longyearbyen", "Asia/Aden", "Asia/Almaty", "Asia/Amman", "Asia/Anadyr", "Asia/Aqtau", "Asia/Aqtobe", "Asia/Ashgabat", "Asia/Atyrau", "Asia/Baghdad", "Asia/Bahrain", "Asia/Baku", "Asia/Bangkok", "Asia/Barnaul", "Asia/Beirut", "Asia/Bishkek", "Asia/Brunei", "Asia/Chita", "Asia/Choibalsan", "Asia/Colombo", "Asia/Damascus", "Asia/Dhaka", "Asia/Dili", "Asia/Dubai", "Asia/Dushanbe", "Asia/Famagusta", "Asia/Gaza", "Asia/Hebron", "Asia/Ho_Chi_Minh", "Asia/Hong_Kong", "Asia/Hovd", "Asia/Irkutsk", "Asia/Jakarta", "Asia/Jayapura", "Asia/Jerusalem", "Asia/Kabul", "Asia/Kamchatka", "Asia/Karachi", "Asia/Kathmandu", "Asia/Khandyga", "Asia/Kolkata", "Asia/Krasnoyarsk", "Asia/Kuala_Lumpur", "Asia/Kuching", "Asia/Kuwait", "Asia/Macau", "Asia/Magadan", "Asia/Makassar", "Asia/Manila", "Asia/Muscat", "Asia/Nicosia", "Asia/Novokuznetsk", "Asia/Novosibirsk", "Asia/Omsk", "Asia/Oral", "Asia/Phnom_Penh", "Asia/Pontianak", "Asia/Pyongyang", "Asia/Qatar", "Asia/Qostanay", "Asia/Qyzylorda", "Asia/Riyadh", "Asia/Sakhalin", "Asia/Samarkand", "Asia/Seoul", "Asia/Shanghai", "Asia/Singapore", "Asia/Srednekolymsk", "Asia/Taipei", "Asia/Tashkent", "Asia/Tbilisi", "Asia/Tehran", "Asia/Thimphu", "Asia/Tokyo", "Asia/Tomsk", "Asia/Ulaanbaatar", "Asia/Urumqi", "Asia/Ust-Nera", "Asia/Vientiane", "Asia/Vladivostok", "Asia/Yakutsk", "Asia/Yangon", "Asia/Yekaterinburg", "Asia/Yerevan", "Atlantic/Azores", "Atlantic/Bermuda", "Atlantic/Canary", "Atlantic/Cape_Verde", "Atlantic/Faroe", "Atlantic/Madeira", "Atlantic/Reykjavik", "Atlantic/South_Georgia", "Atlantic/St_Helena", "Atlantic/Stanley", "Australia/Adelaide", "Australia/Brisbane", "Australia/Broken_Hill", "Australia/Currie", "Australia/Darwin", "Australia/Eucla", "Australia/Hobart", "Australia/Lindeman", "Australia/Lord_Howe", "Australia/Melbourne", "Australia/Perth", "Australia/Sydney", "Canada/Atlantic", "Canada/Central", "Canada/Eastern", "Canada/Mountain", "Canada/Newfoundland", "Canada/Pacific", "Europe/Amsterdam", "Europe/Andorra", "Europe/Astrakhan", "Europe/Athens", "Europe/Belgrade", "Europe/Berlin", "Europe/Bratislava", "Europe/Brussels", "Europe/Bucharest", "Europe/Budapest", "Europe/Busingen", "Europe/Chisinau", "Europe/Copenhagen", "Europe/Dublin", "Europe/Gibraltar", "Europe/Guernsey", "Europe/Helsinki", "Europe/Isle_of_Man", "Europe/Istanbul", "Europe/Jersey", "Europe/Kaliningrad", "Europe/Kiev", "Europe/Kirov", "Europe/Lisbon", "Europe/Ljubljana", "Europe/London", "Europe/Luxembourg", "Europe/Madrid", "Europe/Malta", "Europe/Mariehamn", "Europe/Minsk", "Europe/Monaco", "Europe/Moscow", "Europe/Oslo", "Europe/Paris", "Europe/Podgorica", "Europe/Prague", "Europe/Riga", "Europe/Rome", "Europe/Samara", "Europe/San_Marino", "Europe/Sarajevo", "Europe/Saratov", "Europe/Simferopol", "Europe/Skopje", "Europe/Sofia", "Europe/Stockholm", "Europe/Tallinn", "Europe/Tirane", "Europe/Ulyanovsk", "Europe/Uzhgorod", "Europe/Vaduz", "Europe/Vatican", "Europe/Vienna", "Europe/Vilnius", "Europe/Volgograd", "Europe/Warsaw", "Europe/Zagreb", "Europe/Zaporozhye", "Europe/Zurich", "GMT", "Indian/Antananarivo", "Indian/Chagos", "Indian/Christmas", "Indian/Cocos", "Indian/Comoro", "Indian/Kerguelen", "Indian/Mahe", "Indian/Maldives", "Indian/Mauritius", "Indian/Mayotte", "Indian/Reunion", "Pacific/Apia", "Pacific/Auckland", "Pacific/Bougainville", "Pacific/Chatham", "Pacific/Chuuk", "Pacific/Easter", "Pacific/Efate", "Pacific/Enderbury", "Pacific/Fakaofo", "Pacific/Fiji", "Pacific/Funafuti", "Pacific/Galapagos", "Pacific/Gambier", "Pacific/Guadalcanal", "Pacific/Guam", "Pacific/Honolulu", "Pacific/Kiritimati", "Pacific/Kosrae", "Pacific/Kwajalein", "Pacific/Majuro", "Pacific/Marquesas", "Pacific/Midway", "Pacific/Nauru", "Pacific/Niue", "Pacific/Norfolk", "Pacific/Noumea", "Pacific/Pago_Pago", "Pacific/Palau", "Pacific/Pitcairn", "Pacific/Pohnpei", "Pacific/Port_Moresby", "Pacific/Rarotonga", "Pacific/Saipan", "Pacific/Tahiti", "Pacific/Tarawa", "Pacific/Tongatapu", "Pacific/Wake", "Pacific/Wallis", "US/Alaska", "US/Arizona", "US/Central", "US/Eastern", "US/Hawaii", "US/Mountain", "US/Pacific", "UTC") CHARACTER SET utf8mb4 COLLATE utf8mb4_bin NOT NULL DEFAULT \"Europe/Amsterdam\",' \
93
+ '`frequency` varchar(255) DEFAULT \'{"month":0,"day":0,"hour":0,"minute":0}\',' \
94
+ '`start_after_task_id` int DEFAULT NULL,' \
95
+ '`start_after_preceding_task` enum("FAILED", "SUCCESS") DEFAULT NULL,' \
96
+ '`last_reload` timestamp NULL DEFAULT NULL,' \
97
+ '`last_error_message` varchar(255) DEFAULT NULL,' \
98
+ '`status` varchar(255) DEFAULT \'IDLE\',' \
99
+ '`disabled` tinyint(4) DEFAULT \'1\',' \
100
+ '`run_instant` tinyint(1) DEFAULT \'0\',' \
101
+ '`sftp_mapping` varchar(255) NOT NULL DEFAULT \'[]\',' \
102
+ '`step_nr` int NOT NULL DEFAULT \'0\',' \
103
+ '`stopped_by_user` tinyint(1) DEFAULT \'0\',' \
104
+ '`stop_is_allowed` bool NOT NULL DEFAULT \'0\',' \
105
+ 'PRIMARY KEY (`id`),' \
106
+ 'UNIQUE KEY `task_scheduler_id_uindex` (`id`),' \
107
+ 'constraint task_scheduler_task_scheduler_id_fk foreign key (start_after_task_id) references task_scheduler (id)' \
108
+ ') ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci'
109
+ self.mysql.raw_query(new_table_query)
110
+
111
+ # Check if the table task_scheduler_log exists. If not, create it
112
+ new_table_query = 'CREATE TABLE IF NOT EXISTS `task_scheduler_log` (' \
113
+ '`reload_id` bigint NOT NULL,' \
114
+ '`task_id` int NULL,' \
115
+ '`reload_status` varchar(255) NULL,' \
116
+ '`started_at` datetime NULL,' \
117
+ '`finished_at` datetime NULL' \
118
+ ') ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci'
119
+ self.mysql.raw_query(new_table_query)
120
+
121
+ # Check if the table check_task_execution_log exists. If not, create it
122
+ new_table_query = 'CREATE TABLE IF NOT EXISTS `task_execution_log`(' \
123
+ '`reload_id` bigint NOT NULL,' \
124
+ '`task_id` int NULL,' \
125
+ '`log_level` varchar(255) NULL,' \
126
+ '`created_at` datetime NULL,' \
127
+ '`line_number` int NULL,' \
128
+ '`message` longtext NULL)' \
129
+ 'ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci'
130
+ self.mysql.raw_query(new_table_query)
131
+
132
+ # Check if the table check_task_execution_steps exists. If not, create it
133
+ new_table_query = 'CREATE TABLE IF NOT EXISTS `task_execution_steps`(' \
134
+ '`id` bigint NOT NULL AUTO_INCREMENT,' \
135
+ '`task_id` int NULL,' \
136
+ '`nr` int DEFAULT 0 NOT NULL,' \
137
+ '`description` varchar(255) DEFAULT \'ZzZzZz...\' NOT NULL,' \
138
+ 'PRIMARY KEY (`id`),' \
139
+ 'UNIQUE KEY `task_execution_steps_id_uindex` (`id`),' \
140
+ 'UNIQUE INDEX `task_execution_steps_task_id_nr_uindex` (`task_id`, `nr`))' \
141
+ 'ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci'
142
+ self.mysql.raw_query(new_table_query)
143
+
144
+ new_table_query = 'CREATE TABLE IF NOT EXISTS `task_variables`(' \
145
+ 'id INT NOT NULL AUTO_INCREMENT,' \
146
+ 'task_id INT NOT NULL,' \
147
+ 'name VARCHAR(150) NOT NULL,' \
148
+ 'description VARCHAR(255) NULL,' \
149
+ 'type ENUM(\'INT\', \'TINYINT\', \'BIGINT\', \'FLOAT\', \'DOUBLE\', \'DATETIME\', \'TIMESTAMP\', \'TIME\', \'VARCHAR\', \'BLOB\', \'TEXT\', \'LONGBLOB\') NOT NULL,' \
150
+ 'value VARCHAR(600) NULL,' \
151
+ 'temp_value VARCHAR(600) NULL,' \
152
+ 'PRIMARY KEY (`id`),' \
153
+ 'UNIQUE KEY `task_variables_id_uindex` (`id`),' \
154
+ 'UNIQUE INDEX `task_variables_name_value_uindex` (`task_id`, `name`, `value`), ' \
155
+ 'INDEX `task_variables_name_index` (`name`),' \
156
+ 'CONSTRAINT task_variables_task_scheduler_id_fk ' \
157
+ 'FOREIGN KEY (`task_id`) REFERENCES task_scheduler (`id`) ON DELETE CASCADE)' \
158
+ 'ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci'
159
+ self.mysql.raw_query(new_table_query)
160
+
161
+ # Add the variable 'email_errors_to' as default to the new added table if it doesn't exist for the current task
162
+ response = self.mysql.select('task_variables', 'temp_value',
163
+ f'WHERE name = \'email_errors_to\' AND task_id = {self.task_id}')
164
+ if len(response) == 0:
165
+ new_variables = f"-- INSERT INTO `task_variables` (`task_id`, `name`, `type`, `value`, `temp_value`) " \
166
+ f"VALUES ({self.task_id}, 'email_errors_to', 'TEXT', 'example@brynq.com, example2@brynq.com', 'example@brynq.com, example2@brynq.com')"
167
+ self.mysql.raw_query(new_variables, insert=True)
168
+
169
+ def create_task_execution_steps(self, step_details: list):
170
+ """
171
+ Check if the given steps already exists in the task_execution_steps table. If not, update or insert the values in the table
172
+ :param step_details: list of dicts. Each dict must contain task details according to required_fields.
173
+ Example: step_details = [
174
+ {'nr': 1, 'description': 'test'},
175
+ {'nr': 2, 'description': 'test2'}
176
+ ]
177
+ :return: error (str) or response of mysql
178
+ """
179
+ # Check if the required fields are available in the given list
180
+ required_fields = ['nr', 'description']
181
+ for step in step_details:
182
+ for field in required_fields:
183
+ if field not in step.keys():
184
+ return 'Field {field} is required in step {step}. Required fields are: {required_fields}'.format(
185
+ field=field, step=step, required_fields=tuple(required_fields))
186
+
187
+ # Reformat the list of dictionaries to a valid MySQL query
188
+ values = ','.join(str((self.task_id, step['nr'], step['description'])) for step in step_details)
189
+ response = self.mysql.raw_query("INSERT INTO task_execution_steps (`task_id`, `nr`, `description`) "
190
+ "VALUES {step_values} ON DUPLICATE KEY UPDATE `description` = VALUES(description)".format(
191
+ step_values=values), insert=True)
192
+ return response
193
+
194
+ def check_if_task_manual_started(self):
195
+ """
196
+ Check if the task manual is started of on schedule. If it's manual started, that's important for the variables in the db_variables function.
197
+ In that case the dynamic variables should be used instead of the static ones
198
+ :return: True of False
199
+ """
200
+ response = self.mysql.select('task_scheduler', 'run_instant', f'WHERE id = {self.task_id}')[0][0]
201
+ if response == 1:
202
+ # Reset the 1 back to 0 before sending the result
203
+ self.mysql.update('task_scheduler', ['run_instant'], [0], 'WHERE `id` = {}'.format(self.task_id))
204
+ return True
205
+ else:
206
+ return False
207
+
208
+ def start_task(self):
209
+ """
210
+ Start the task and write this to the database. While the status is running, the task will not start again
211
+ :return: if the update to the database is successful or not
212
+ """
213
+ # If the task is started from a local instance (not the task_scheduler), create a start log row in the task_scheduler_log
214
+ if self.started_local:
215
+ self.mysql.raw_query(f"INSERT INTO `task_scheduler_log` (reload_id, task_id, reload_status, started_at, finished_at) VALUES ({self.run_id}, {self.task_id}, 'Running', '{self.started_at}', null)", insert=True)
216
+ return self.mysql.update('task_scheduler', ['status', 'step_nr'], ['RUNNING', 1], 'WHERE `id` = {}'.format(self.task_id))
217
+
218
+ def db_variable(self, variable_name: str, default_value_if_temp_is_empty: bool = False):
219
+ """
220
+ Get a value from the task_variables table corresponding with the given name. If the task is manually started
221
+ (run_instant = 1), then the temp_value will be returned. This is to give the possibility for users in the frontend to run
222
+ a task once manual with other values then normal without overwriting the normal values.
223
+ :param variable_name: the name of the variable
224
+ :param default_value_if_temp_is_empty: bool to determine whether default value should be used if temp value is empty when manually started
225
+ :return: the value of the given variable.
226
+ """
227
+ if self.task_manual_started is True:
228
+ response = self.mysql.select('task_variables', 'temp_value, value',
229
+ f'WHERE name = \'{variable_name}\' AND task_id = {self.task_id}')
230
+ else:
231
+ response = self.mysql.select('task_variables', 'value',
232
+ f'WHERE name = \'{variable_name}\' AND task_id = {self.task_id}')
233
+ if len(response) == 0:
234
+ raise Exception(f'Variable with name \'{variable_name}\' does not exist')
235
+ else:
236
+ value = response[0][0]
237
+ if value is None and default_value_if_temp_is_empty is True and len(response[0]) > 0:
238
+ value = response[0][1]
239
+ return value
240
+
241
+ def write_execution_log(self, message: str, data, loglevel: str = 'INFO', full_extract: bool = False):
242
+ """
243
+ Writes messages to the database. Give the message and the level of the log
244
+ :param message: A string with a message for the log
245
+ :param loglevel: You can choose between DEBUG, INFO, ERROR or CRITICAL (DEBUG is most granulated, CRITICAL the less)
246
+ :param data: Uploaded data by the interface that has to be logged in ElasticSearch, if you have nothing to log, use None
247
+ :param full_extract: If the data is a full load, set this to True. This will prevent the payload from being logged in ElasticSearch
248
+ :return: If writing to the database is successful or not
249
+ """
250
+
251
+ # Validate if the provided loglevel is valid
252
+ allowed_loglevels = ['DEBUG', 'INFO', 'ERROR', 'CRITICAL']
253
+ if loglevel not in allowed_loglevels:
254
+ raise Exception('You\'ve entered a not allowed loglevel. Choose one of: {}'.format(allowed_loglevels))
255
+
256
+ # Handling different data types and preparing extra payload information based on the data type
257
+ if isinstance(data, pd.Series):
258
+ dataframe = pd.DataFrame(data).T
259
+ extra_payload = {
260
+ 'rows': len(dataframe),
261
+ 'columns': len(dataframe.columns),
262
+ 'cells': len(dataframe) * len(dataframe.columns),
263
+ }
264
+ if not full_extract:
265
+ extra_payload['payload'] = dataframe.to_json(orient='records')
266
+ elif isinstance(data, dict):
267
+ records = self.__count_keys(data)
268
+ extra_payload = {
269
+ 'rows': 1,
270
+ 'columns': records,
271
+ 'cells': records,
272
+ }
273
+ if not full_extract:
274
+ extra_payload['payload'] = data
275
+ elif isinstance(data, pd.DataFrame):
276
+ extra_payload = {
277
+ 'rows': len(data),
278
+ 'columns': len(data.columns),
279
+ 'cells': len(data) * len(data.columns),
280
+ }
281
+ if not full_extract:
282
+ extra_payload['payload'] = data.to_json(orient='records')
283
+ elif isinstance(data, requests.Response):
284
+ records = 1
285
+ if data.request.body is not None:
286
+ records = self.__count_keys(json.loads(data.request.body))
287
+ if isinstance(data.request.body, bytes):
288
+ data.request.body = data.request.body.decode('utf-8')
289
+ extra_payload = {
290
+ 'response': data.text,
291
+ 'status_code': data.status_code,
292
+ 'url': data.url,
293
+ 'method': data.request.method,
294
+ 'rows': 1,
295
+ 'columns': records,
296
+ 'cells': records,
297
+ }
298
+ if not full_extract:
299
+ extra_payload['payload'] = data.request.body
300
+ elif data is None:
301
+ extra_payload = {}
302
+ else:
303
+ extra_payload = {
304
+ 'data_type': str(type(data)),
305
+ }
306
+ if not full_extract:
307
+ extra_payload['payload'] = data
308
+
309
+ # Modify payload based on 'full_load' flag
310
+ if data is not None and full_extract is True:
311
+ extra_payload['full_load'] = True
312
+ elif data is not None and full_extract is False:
313
+ extra_payload['full_load'] = False
314
+
315
+ # Preparing the primary payload with log details
316
+ payload = {
317
+ 'reload_id': self.run_id,
318
+ 'task_id': self.task_id,
319
+ 'customer_id': os.getenv('BRYNQ_CUSTOMER_NAME').lower().replace(' ', '_'),
320
+ 'started_at': datetime.datetime.now().isoformat(),
321
+ 'loglevel': loglevel,
322
+ 'message': message
323
+ }
324
+ payload.update(extra_payload)
325
+
326
+ # Sending the payload to ElasticSearch
327
+ self.es.task_execution_log(payload)
328
+
329
+ # Get the linenumber from where the logline is executed. Get the stacktrace of this action, jump 1 file up and pick then the linenumber (second item)
330
+ linenumber = inspect.getouterframes(inspect.currentframe())[1][2]
331
+ # Write the logline to the database, depends on the chosen loglevel in the task
332
+ print('{} at line: {}'.format(message, linenumber))
333
+ # Remove quotes from message since these break the query
334
+ message = re.sub("[']", '', message)
335
+ timestamp = time.strftime("%Y-%m-%d %H:%M:%S")
336
+ information = {
337
+ 'reload_id': self.run_id,
338
+ 'task_id': self.task_id,
339
+ 'log_level': loglevel,
340
+ 'line_number': linenumber,
341
+ 'message': message,
342
+ 'created_at': timestamp
343
+ }
344
+ if self.loglevel == 'DEBUG':
345
+ # Count the errors
346
+ if loglevel == 'ERROR' or loglevel == 'CRITICAL':
347
+ self.error_count += 1
348
+ return self.mysql.raw_query(
349
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, '{}', '{}', {}, '{}')".format(self.run_id, self.task_id, loglevel, datetime.datetime.now(), linenumber, message), insert=True)
350
+ return self.mysql.update(table='task_execution_log',
351
+ columns=['reload_id', 'task_id', 'log_level', 'created_at', 'line_number', 'message'],
352
+ values=[self.run_id, self.task_id, loglevel, datetime.datetime.now(), linenumber, message])
353
+ elif self.loglevel == 'INFO' and (loglevel == 'INFO' or loglevel == 'ERROR' or loglevel == 'CRITICAL'):
354
+ # Count the errors
355
+ if loglevel == 'ERROR' or loglevel == 'CRITICAL':
356
+ self.error_count += 1
357
+ return self.mysql.raw_query(
358
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, '{}', '{}', {}, '{}')".format(self.run_id, self.task_id, loglevel, datetime.datetime.now(), linenumber, message), insert=True)
359
+ elif self.loglevel == 'ERROR' and (loglevel == 'ERROR' or loglevel == 'CRITICAL'):
360
+ self.error_count += 1
361
+ return self.mysql.raw_query(
362
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, '{}', '{}', {}, '{}')".format(self.run_id, self.task_id, loglevel, datetime.datetime.now(), linenumber, message), insert=True)
363
+ elif self.loglevel == 'CRITICAL' and loglevel == 'CRITICAL':
364
+ self.error_count += 1
365
+ return self.mysql.raw_query(
366
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, '{}', '{}', {}, '{}')".format(self.run_id, self.task_id, loglevel, datetime.datetime.now(), linenumber, message), insert=True)
367
+
368
+ def update_execution_step(self, step_number: int):
369
+ """
370
+ Update the current step number in the task_scheduler table so that user's in the frontend of BrynQ can see where a task is at any moment
371
+ :param step_number: Give only a number
372
+ :return: nothing
373
+ """
374
+ # Update the step number in the task_scheduler table
375
+ return self.mysql.update('task_scheduler', ['step_nr'], [step_number], 'WHERE `id` = {}'.format(self.task_id))
376
+
377
+ def error_handling(self, e: Exception, breaking=True, send_to_teams=False):
378
+ """
379
+ This function handles errors that occur in the scheduler. Logs the traceback, updates run statuses and notifies users
380
+ :param e: the Exception that is to be handled
381
+ :param task_id: The scheduler task id
382
+ :param mysql_con: The connection which is used to update the scheduler task status
383
+ :param logger: The logger that is used to write the logging status to
384
+ :param breaking: Determines if the error is breaking or code will continue
385
+ :param started_at: Give the time the task is started
386
+ :return: nothing
387
+ """
388
+
389
+ # Preparing the primary payload with error details for upload to elastic
390
+ payload = {
391
+ 'reload_id': self.run_id,
392
+ 'task_id': self.task_id,
393
+ 'customer_id': os.getenv('BRYNQ_CUSTOMER_NAME').lower().replace(' ', '_'),
394
+ 'started_at': datetime.datetime.now().isoformat(),
395
+ 'loglevel': 'CRITICAL',
396
+ 'message': str(e),
397
+ 'traceback': traceback.format_exc()
398
+ }
399
+
400
+ # Sending the payload to ElasticSearch
401
+ self.es.task_execution_log(payload)
402
+
403
+ # Format error to a somewhat readable format
404
+ exc_type, exc_obj, exc_tb = sys.exc_info()
405
+ error = str(e)[:400].replace('\'', '').replace('\"', '') + ' | Line: {}'.format(exc_tb.tb_lineno)
406
+ # Get scheduler task details for logging
407
+ task_details = \
408
+ self.mysql.select('task_scheduler', 'docker_image, runfile_path', 'WHERE id = {}'.format(self.task_id))[0]
409
+ taskname = task_details[0]
410
+ customer = task_details[1].split('/')[-1].split('.')[0]
411
+
412
+ if breaking:
413
+ # Set scheduler status to failed
414
+ self.mysql.update('task_scheduler', ['status', 'last_reload', 'last_error_message', 'step_nr'],
415
+ ['IDLE', datetime.datetime.now(), 'Failed', 0],
416
+ 'WHERE `id` = {}'.format(self.task_id))
417
+ # Log to database
418
+ timestamp = time.strftime("%Y-%m-%d %H:%M:%S")
419
+ information = {
420
+ 'reload_id': self.run_id,
421
+ 'task_id': self.task_id,
422
+ 'log_level': 'CRITICAL',
423
+ 'line_number': exc_tb.tb_lineno,
424
+ 'message': error,
425
+ 'created_at': timestamp
426
+ }
427
+ self.mysql.raw_query(
428
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, 'CRITICAL', '{}', {}, '{}')".format(self.run_id,
429
+ self.task_id,
430
+ datetime.datetime.now(),
431
+ exc_tb.tb_lineno,
432
+ error),
433
+ insert=True)
434
+ self.mysql.update(table='task_scheduler_log',
435
+ columns=['reload_status', 'finished_at'],
436
+ values=['Failed', f'{datetime.datetime.now()}'],
437
+ filter=f'WHERE `reload_id` = {self.run_id}')
438
+ # Notify users on Teams and If the variable self.send_mail_after_errors is set to True, send an email with the message that the task is failed
439
+ if send_to_teams:
440
+ Functions.send_error_to_teams(database=customer, task_number=self.task_id, task_title=taskname)
441
+ if self.email_after_errors:
442
+ self.email_errors(failed=True)
443
+ # Remove the temp values from the variables table
444
+ self.mysql.raw_query(f'UPDATE `task_variables` SET temp_value = null WHERE task_id = {self.task_id}', insert=True)
445
+
446
+ # Start the chained tasks if it there are tasks which should start if this one is failed
447
+ self.start_chained_tasks(finished_task_status='FAILED')
448
+
449
+ raise Exception(error)
450
+ else:
451
+ timestamp = time.strftime("%Y-%m-%d %H:%M:%S")
452
+ information = {
453
+ 'reload_id': self.run_id,
454
+ 'task_id': self.task_id,
455
+ 'log_level': 'CRITICAL',
456
+ 'line_number': exc_tb.tb_lineno,
457
+ 'message': error,
458
+ 'created_at': timestamp
459
+ }
460
+ self.mysql.raw_query(
461
+ "INSERT INTO `task_execution_log` (reload_id, task_id, log_level, created_at, line_number, message) VALUES ({}, {}, 'CRITICAL', '{}', {}, '{}')".format(self.run_id,
462
+ self.task_id,
463
+ datetime.datetime.now(),
464
+ exc_tb.tb_lineno,
465
+ error),
466
+ insert=True)
467
+ if send_to_teams:
468
+ Functions.send_error_to_teams(database=customer, task_number=self.task_id, task_title=taskname)
469
+ self.error_count += 1
470
+
471
+ def finish_task(self, reload_instant=False, log_limit: typing.Optional[int] = 10000, log_date_limit: datetime.date = None):
472
+ """
473
+ At the end of the script, write the outcome to the database. Write if the task is finished with or without errors, Email to a contactperson if this variable is given in the
474
+ variables table. Also clean up the execution_log table when the number of lines is more than 1000
475
+ :return:
476
+ """
477
+ # If reload instant is true, this adds an extra field 'run_instant' to the update query, and sets the value to 1. This makes the task reload immediately after it's finished
478
+ field = ['run_instant', 'next_reload'] if reload_instant else []
479
+ value = ['1', datetime.datetime.now()] if reload_instant else []
480
+ if self.error_count > 0:
481
+ self.mysql.update('task_scheduler', ['status', 'last_reload', 'last_error_message', 'step_nr'],
482
+ ['IDLE', datetime.datetime.now(), 'FinishedWithErrors', 0],
483
+ 'WHERE `id` = {}'.format(self.task_id))
484
+ self.mysql.update(table='task_scheduler_log',
485
+ columns=['reload_status', 'finished_at'],
486
+ values=['FinishedWithErrors', f'{datetime.datetime.now()}'],
487
+ filter=f'WHERE `reload_id` = {self.run_id}')
488
+ # If the variable self.send_mail_after_errors is set to True, send an email with the number of errors to the given user
489
+ if self.email_after_errors:
490
+ self.email_errors(failed=False)
491
+ else:
492
+ self.mysql.update(table='task_scheduler',
493
+ columns=['status', 'last_reload', 'last_error_message', 'step_nr', 'stopped_by_user'] + field,
494
+ values=['IDLE', datetime.datetime.now(), 'FinishedSucces', 0, 0] + value,
495
+ filter='WHERE `id` = {}'.format(self.task_id))
496
+
497
+ self.mysql.update(table='task_scheduler_log',
498
+ columns=['reload_status', 'finished_at'],
499
+ values=['FinishedSuccess', f'{datetime.datetime.now()}'],
500
+ filter=f'WHERE `reload_id` = {self.run_id}')
501
+
502
+ # Remove the temp values from the variables table
503
+ self.mysql.raw_query(f'UPDATE `task_variables` SET temp_value = null WHERE task_id = {self.task_id}', insert=True)
504
+
505
+ # Start the new task if it there is a task which should start if this one is finished
506
+ self.start_chained_tasks(finished_task_status='SUCCESS')
507
+
508
+ # Clean up execution log
509
+ # set this date filter above the actual delete filter because of the many uncooperative quotation marks involved in the whole filter
510
+ log_date_limit_filter = f"AND created_at >= \'{log_date_limit.strftime('%Y-%m-%d')}\'" if log_date_limit is not None else None
511
+ delete_filter = f"WHERE task_id = {self.task_id} " \
512
+ f"AND reload_id NOT IN (SELECT reload_id FROM (SELECT reload_id FROM `task_execution_log` WHERE task_id = {self.task_id} " \
513
+ f"AND log_level != 'CRITICAL' " \
514
+ f"AND log_level != 'ERROR' " \
515
+ f"{log_date_limit_filter if log_date_limit_filter is not None else ''} " \
516
+ f"ORDER BY created_at DESC {f' LIMIT {log_limit} ' if log_limit is not None else ''}) temp)"
517
+
518
+ resp = self.mysql.delete(table="task_execution_log",
519
+ filter=delete_filter)
520
+ print(resp)
521
+
522
+ def start_chained_tasks(self, finished_task_status: str):
523
+ filter = f'WHERE start_after_task_id = \'{self.task_id}\' AND start_after_preceding_task = \'{finished_task_status}\''
524
+ response = self.mysql.select(table='task_scheduler', selection='id', filter=filter)
525
+ if len(response) > 0:
526
+ tasks_to_run = [str(task[0]) for task in response]
527
+ self.mysql.update(table='task_scheduler', columns=['run_instant'], values=['1'], filter=f'WHERE id IN({",".join(tasks_to_run)})')
528
+
529
+ def email_errors(self, failed):
530
+ # The mails to email to should be stored in the task_variables table with the variable email_errors_to
531
+ email_variable = self.db_variable('email_errors_to')
532
+ if email_variable is not None:
533
+ email_to = email_variable.split(',')
534
+ if isinstance(email_to, list):
535
+ # The email_errors_to variable is a simple string. Convert it to a list and add a name because mandrill is asking for it
536
+ email_list = []
537
+ for i in email_to:
538
+ email_list.append({'name': 'BrynQ User', 'mail': i.strip()})
539
+ # Set the content of the mail and all other stuff
540
+ task = self.mysql.select(table='task_scheduler', selection='title', filter=f'WHERE id = {self.task_id}')[0][
541
+ 0]
542
+ finished_at = \
543
+ self.mysql.select(table='task_scheduler', selection='last_reload', filter=f'WHERE id = {self.task_id}')[0][
544
+ 0]
545
+ if failed:
546
+ subject = f'Task \'{task}\' has failed'
547
+ content = f'Task \'{task}\' with task ID \'{self.task_id}\' failed during its last run and was stopped at {finished_at}. ' \
548
+ f'The task is failed. ' \
549
+ f'to visit the BrynQ scheduler, click here: <a href="https://app.brynq.com/interfaces/">here</a>. Here you can find the logs and find more information on why this task had failed.'
550
+ else:
551
+ subject = f'Task \'{task}\' is finished with errors'
552
+ content = f'Task \'{task}\' with ID \'{self.task_id}\' has runned and is finished at {finished_at}. ' \
553
+ f'The task is finished with {self.error_count} errors. ' \
554
+ f'to visit the BrynQ scheduler, click here: <a href="https://app.brynq.com/interfaces/">here</a>. Here you can find the logs and find more information on why this task had some errors.'
555
+ MailClient().send_mail(email_to=email_list, subject=subject, content=content, language='EN')
@@ -0,0 +1,10 @@
1
+ Metadata-Version: 1.0
2
+ Name: brynq-sdk-task-scheduler
3
+ Version: 1.0.0
4
+ Summary: Code to execute tasks in BrynQ.com with the task scheduler
5
+ Home-page: UNKNOWN
6
+ Author: BrynQ
7
+ Author-email: support@brynq.com
8
+ License: BrynQ License
9
+ Description: Code to execute tasks in the BrynQ.com platform with the task scheduler
10
+ Platform: UNKNOWN
@@ -0,0 +1,9 @@
1
+ setup.py
2
+ brynq_sdk/task_scheduler/__init__.py
3
+ brynq_sdk/task_scheduler/task_scheduler.py
4
+ brynq_sdk_task_scheduler.egg-info/PKG-INFO
5
+ brynq_sdk_task_scheduler.egg-info/SOURCES.txt
6
+ brynq_sdk_task_scheduler.egg-info/dependency_links.txt
7
+ brynq_sdk_task_scheduler.egg-info/not-zip-safe
8
+ brynq_sdk_task_scheduler.egg-info/requires.txt
9
+ brynq_sdk_task_scheduler.egg-info/top_level.txt
@@ -0,0 +1,5 @@
1
+ brynq-sdk-brynq>=1
2
+ brynq-sdk-functions>=1
3
+ salure-helpers-mysql>=1
4
+ salure-helpers-mandrill>=0
5
+ salure-helpers-elastic>=1
@@ -0,0 +1,4 @@
1
+ [egg_info]
2
+ tag_build =
3
+ tag_date = 0
4
+
@@ -0,0 +1,21 @@
1
+ from setuptools import setup
2
+
3
+
4
+ setup(
5
+ name='brynq_sdk_task_scheduler',
6
+ version='1.0.0',
7
+ description='Code to execute tasks in BrynQ.com with the task scheduler',
8
+ long_description='Code to execute tasks in the BrynQ.com platform with the task scheduler',
9
+ author='BrynQ',
10
+ author_email='support@brynq.com',
11
+ packages=["brynq_sdk.task_scheduler"],
12
+ license='BrynQ License',
13
+ install_requires=[
14
+ 'brynq-sdk-brynq>=1',
15
+ 'brynq-sdk-functions>=1',
16
+ 'salure-helpers-mysql>=1',
17
+ 'salure-helpers-mandrill>=0',
18
+ 'salure-helpers-elastic>=1'
19
+ ],
20
+ zip_safe=False,
21
+ )