| 1 |
c8f3051b
|
petrh
|
from Utilities import FolderProcessor, ConfigureLoader
|
| 2 |
|
|
from Utilities.Database import DatabaseLoader
|
| 3 |
|
|
|
| 4 |
|
|
|
| 5 |
|
|
CONFIG_FILES_PATH = "DatasetConfigs/"
|
| 6 |
|
|
CRAWLED_DATA_PATH = "CrawledData/"
|
| 7 |
|
|
PROCESSED_DATA_PATH = "ProcessedData/"
|
| 8 |
|
|
CRAWLER_LIB_PATH = "DatasetCrawler."
|
| 9 |
|
|
PROCESSOR_LIB_PATH = "DatasetProcessing."
|
| 10 |
|
|
|
| 11 |
|
|
|
| 12 |
|
|
def crawl_data(config):
|
| 13 |
|
|
|
| 14 |
|
|
dataset_name = config["dataset-name"]
|
| 15 |
|
|
|
| 16 |
|
|
my_function = __import__(CRAWLER_LIB_PATH + dataset_name + "Crawler",globals(),locals(),['crawl']).crawl
|
| 17 |
|
|
my_function(config)
|
| 18 |
|
|
|
| 19 |
|
|
dataset_name += '/'
|
| 20 |
|
|
|
| 21 |
|
|
|
| 22 |
|
|
def process_data(dataset_name):
|
| 23 |
|
|
dataset_path = dataset_name + '/'
|
| 24 |
|
|
|
| 25 |
|
|
process_file_func = __import__(PROCESSOR_LIB_PATH + dataset_name + "Processor",globals(),locals(),
|
| 26 |
|
|
['process_file']).process_file
|
| 27 |
|
|
|
| 28 |
|
|
# get all not processed files from dataset
|
| 29 |
|
|
not_processed_files = FolderProcessor.list_of_all_files(CRAWLED_DATA_PATH + dataset_path)
|
| 30 |
|
|
|
| 31 |
|
|
# process every file
|
| 32 |
|
|
for not_processed_file in not_processed_files:
|
| 33 |
|
|
# call processing for one file in dataset
|
| 34 |
|
|
process_file_func(CRAWLED_DATA_PATH + dataset_path + not_processed_file)
|
| 35 |
|
|
FolderProcessor.update_ignore_set(CRAWLED_DATA_PATH + dataset_path, not_processed_file)
|
| 36 |
|
|
|
| 37 |
|
|
|
| 38 |
|
|
def load_data_to_database(config):
|
| 39 |
|
|
|
| 40 |
|
|
dataset_name = config["dataset-name"]
|
| 41 |
|
|
dataset_path = dataset_name + '/'
|
| 42 |
|
|
|
| 43 |
|
|
# get all unprocessed files from dataset
|
| 44 |
|
|
not_loaded_files = FolderProcessor.list_of_all_files(PROCESSED_DATA_PATH + dataset_path)
|
| 45 |
|
|
|
| 46 |
|
|
# load every file
|
| 47 |
|
|
for not_loaded_file in not_loaded_files:
|
| 48 |
|
|
# load processed data
|
| 49 |
|
|
processed_data = DatabaseLoader.get_data_from_file(PROCESSED_DATA_PATH + dataset_path + not_loaded_file,
|
| 50 |
|
|
config["devices"])
|
| 51 |
|
|
# load processed data to database
|
| 52 |
|
|
DatabaseLoader.load_data_to_database(dataset_name, processed_data)
|
| 53 |
|
|
FolderProcessor.update_ignore_set(PROCESSED_DATA_PATH + dataset_path, not_loaded_file)
|
| 54 |
|
|
|
| 55 |
|
|
|
| 56 |
|
|
def run_full_pipeline(dataset_name):
|
| 57 |
|
|
config = ConfigureLoader.load_configuration(CONFIG_FILES_PATH + dataset_name)
|
| 58 |
|
|
crawl_data(config)
|
| 59 |
|
|
process_data(config["dataset-name"])
|
| 60 |
|
|
load_data_to_database(config)
|