git commit -m "SM-17 <test>"
This commit is contained in:
parent
f03325a6a2
commit
98fb313a6f
|
|
@ -1,127 +0,0 @@
|
|||
This README file provides a comprehensive guide to utilizing a Python script for interacting with S3 storage,
|
||||
specifically designed for downloading and processing data files based on a specified time range and key parameters.
|
||||
The script requires Python3 installed on your system and makes use of the s3cmd tool for accessing data in cloud storage.
|
||||
It also illustrates the process of configuring s3cmd by creating a .s3cfg file with your access credentials.
|
||||
|
||||
|
||||
############ Create the .s3cfg file in home directory ################
|
||||
|
||||
nano .s3cfg
|
||||
|
||||
Copy this lines inside the file.
|
||||
|
||||
[default]
|
||||
host_base = sos-ch-dk-2.exo.io
|
||||
host_bucket = %(bucket)s.sos-ch-dk-2.exo.io
|
||||
access_key = EXO4d838d1360ba9fb7d51648b0
|
||||
secret_key = _bmrp6ewWAvNwdAQoeJuC-9y02Lsx7NV6zD-WjljzCU
|
||||
use_https = True
|
||||
|
||||
|
||||
############ S3cmd instalation ################
|
||||
|
||||
Please install s3cmd for retrieving data from our Cloud storage.
|
||||
|
||||
sudo apt install s3cmd
|
||||
|
||||
############ Python3 instalation ################
|
||||
|
||||
To check if you have already have python3, run this command
|
||||
|
||||
python3 --version
|
||||
|
||||
|
||||
To install you can use this command:
|
||||
|
||||
1) sudo apt update
|
||||
|
||||
2) sudo apt install python3
|
||||
|
||||
3) python3 --version (to check if pyhton3 installed correctly)
|
||||
|
||||
|
||||
############ Run extractRange.py ################
|
||||
|
||||
usage: extractRange.py [-h] --key KEY --bucket-number BUCKET_NUMBER start_timestamp end_timestamp
|
||||
|
||||
KEY: the key can be a one word or a path
|
||||
|
||||
for example: /DcDc/Devices/2/Status/Dc/Battery/voltage ==> this will provide us a Dc battery Voltage of the DcDc device 2.
|
||||
example : Dc/Battery/voltage ==> This will provide all DcDc Device voltage (including the avg voltage of all DcDc device)
|
||||
example : voltage ==> This will provide all voltage of all devices in the Salimax
|
||||
|
||||
BUCKET_NUMBER: This a number of bucket name for the instalation
|
||||
|
||||
List of bucket number/ instalation:
|
||||
1: Prototype
|
||||
2: Marti Technik (Bern)
|
||||
3: Schreinerei Schönthal (Thun)
|
||||
4: Wittmann Kottingbrunn
|
||||
5: Biohof Gubelmann (Walde)
|
||||
6: Steakhouse Mettmenstetten
|
||||
7: Andreas Ballif / Lerchenhof
|
||||
8: Weidmann Oberwil (ZG)
|
||||
9: Christian Huber (EBS Elektrotechnik)
|
||||
|
||||
|
||||
start_timestamp end_timestamp: this must be a correct timestamp of 10 digits.
|
||||
The start_timestamp must be smaller than the end_timestamp.
|
||||
|
||||
PS: The data will be downloaded to a folder named S3cmdData_{Bucket_Number}. If this folder does not exist, it will be created.
|
||||
If the folder exist, it will try to download data if there is no files in the folder.
|
||||
If the folder exist and contains at least one file, it will only data extraction.
|
||||
|
||||
Example command:
|
||||
|
||||
python3 extractRange.py 1707087500 1707091260 --key ActivePowerImportT2 --bucket-number 1
|
||||
|
||||
|
||||
################################ EXTENDED FEATURES FOR MORE ADVANCED USAGE ################################
|
||||
|
||||
1) Multiple Keys Support:
|
||||
|
||||
The script supports the extraction of data using multiple keys. Users can specify one or multiple keys separated by commas with the --keys parameter.
|
||||
This feature allows for more granular data extraction, catering to diverse data analysis requirements. For example, users can extract data for different
|
||||
metrics or parameters from the same or different CSV files within the specified range.
|
||||
|
||||
2) Exact Match for Keys:
|
||||
|
||||
With the --exact_match flag, the script offers an option to enforce exact matching of keys. This means that only the rows containing a key that exactly
|
||||
matches the specified key(s) will be considered during the data extraction process. This option enhances the precision of the data extraction, making it
|
||||
particularly useful when dealing with CSV files that contain similar but distinct keys.
|
||||
|
||||
3) Dynamic Header Generation:
|
||||
|
||||
The script dynamically generates headers for the output CSV file based on the keys provided. This ensures that the output file accurately reflects the
|
||||
extracted data, providing a clear and understandable format for subsequent analysis. The headers correspond to the keys used for data extraction, making
|
||||
it easy to identify and analyze the extracted data.
|
||||
|
||||
4)Advanced Data Processing Capabilities:
|
||||
|
||||
i) Booleans as Numbers: The --booleans_as_numbers flag allows users to convert boolean values (True/False) into numeric representations (1/0). This feature
|
||||
is particularly useful for analytical tasks that require numerical data processing.
|
||||
|
||||
ii) Sampling Stepsize: The --sampling_stepsize parameter enables users to define the granularity of the time range for data extraction. By specifying the number
|
||||
of 2-second intervals, users can adjust the sampling interval, allowing for flexible data retrieval based on time.
|
||||
|
||||
Example Command:
|
||||
|
||||
python3 extractRange.py 1707087500 1707091260 --keys ActivePowerImportT2,Soc --bucket-number 1 --exact_match --booleans_as_numbers
|
||||
|
||||
|
||||
This command extracts data for ActivePowerImportT2 and TotalEnergy keys from bucket number 1, between the specified timestamps, with exact
|
||||
matching of keys and boolean values converted to numbers.
|
||||
|
||||
Visualization and Data Analysis:
|
||||
|
||||
After data extraction, the script facilitates data analysis by:
|
||||
|
||||
i) Providing a visualization function to plot the extracted data. Users can modify this function to suit their specific analysis needs, adjusting
|
||||
plot labels, titles, and other matplotlib parameters.
|
||||
|
||||
ii) Saving the extracted data in a CSV file, with dynamically generated headers based on the specified keys. This file can be used for further
|
||||
analysis or imported into data analysis tools.
|
||||
|
||||
This Python script streamlines the process of data retrieval from S3 storage, offering flexible and powerful options for data extraction, visualization,
|
||||
and analysis. Its support for multiple keys, exact match filtering, and advanced processing capabilities make it a valuable tool for data analysts and
|
||||
researchers working with time-series data or any dataset stored in S3 buckets.
|
||||
|
|
@ -1,205 +0,0 @@
|
|||
import os
|
||||
import csv
|
||||
import subprocess
|
||||
import argparse
|
||||
import matplotlib.pyplot as plt
|
||||
from collections import defaultdict
|
||||
import zipfile
|
||||
import base64
|
||||
import shutil
|
||||
|
||||
def extract_timestamp(filename):
|
||||
timestamp_str = filename[:10]
|
||||
try:
|
||||
timestamp = int(timestamp_str)
|
||||
return timestamp
|
||||
except ValueError:
|
||||
return 0
|
||||
|
||||
def extract_values_by_key(csv_file, key, exact_match):
|
||||
matched_values = defaultdict(list)
|
||||
with open(csv_file, 'r') as file:
|
||||
reader = csv.reader(file)
|
||||
for row in reader:
|
||||
if row:
|
||||
columns = row[0].split(';')
|
||||
if len(columns) > 1:
|
||||
first_column = columns[0].strip()
|
||||
path_key = first_column.split('/')[-1]
|
||||
for key_item in key:
|
||||
if exact_match:
|
||||
if key_item.lower() == row[0].split('/')[-1].split(';')[0].lower():
|
||||
matched_values[path_key].append(row[0])
|
||||
else:
|
||||
if key_item.lower() in first_column.lower():
|
||||
matched_values[path_key].append(row[0])
|
||||
final_key = ''.join(matched_values.keys())
|
||||
combined_values = []
|
||||
for values in matched_values.values():
|
||||
combined_values.extend(values)
|
||||
final_dict = {final_key: combined_values}
|
||||
return final_dict
|
||||
|
||||
def list_files_in_range(start_timestamp, end_timestamp, sampling_stepsize):
|
||||
filenames_in_range = [f"{timestamp:10d}" for timestamp in range(start_timestamp, end_timestamp + 1, 2*sampling_stepsize)]
|
||||
return filenames_in_range
|
||||
|
||||
def download_files(bucket_number, filenames_to_download, product_type):
|
||||
if product_type == 0:
|
||||
hash = "3e5b3069-214a-43ee-8d85-57d72000c19d"
|
||||
elif product_type == 1:
|
||||
hash = "c0436b6a-d276-4cd8-9c44-1eae86cf5d0e"
|
||||
else:
|
||||
raise ValueError("Invalid product type option. Use 0 or 1")
|
||||
output_directory = f"S3cmdData_{bucket_number}"
|
||||
|
||||
if not os.path.exists(output_directory):
|
||||
os.makedirs(output_directory)
|
||||
print(f"Directory '{output_directory}' created.")
|
||||
|
||||
for filename in filenames_to_download:
|
||||
stripfilename = filename.strip()
|
||||
local_path = os.path.join(output_directory, stripfilename + ".csv")
|
||||
if not os.path.exists(local_path):
|
||||
s3cmd_command = f"s3cmd get s3://{bucket_number}-{hash}/{stripfilename}.csv {output_directory}/"
|
||||
try:
|
||||
subprocess.run(s3cmd_command, shell=True, check=True)
|
||||
downloaded_files = [file for file in os.listdir(output_directory) if file.startswith(filename)]
|
||||
if not downloaded_files:
|
||||
print(f"No matching files found for prefix '{filename}'.")
|
||||
else:
|
||||
print(f"Files with prefix '{filename}' downloaded successfully.")
|
||||
except subprocess.CalledProcessError as e:
|
||||
print(f"Error downloading files: {e}")
|
||||
continue
|
||||
else:
|
||||
print(f"File '{filename}.csv' already exists locally. Skipping download.")
|
||||
|
||||
def decompress_file(compressed_file, output_directory):
|
||||
base_name = os.path.splitext(os.path.basename(compressed_file))[0]
|
||||
|
||||
with open(compressed_file, 'rb') as file:
|
||||
compressed_data = file.read()
|
||||
|
||||
# Decode the base64 encoded content
|
||||
decoded_data = base64.b64decode(compressed_data)
|
||||
|
||||
zip_path = os.path.join(output_directory, 'temp.zip')
|
||||
with open(zip_path, 'wb') as zip_file:
|
||||
zip_file.write(decoded_data)
|
||||
|
||||
with zipfile.ZipFile(zip_path, 'r') as zip_ref:
|
||||
zip_ref.extractall(output_directory)
|
||||
|
||||
# Rename the extracted data.csv file to the original timestamp-based name
|
||||
extracted_csv_path = os.path.join(output_directory, 'data.csv')
|
||||
if os.path.exists(extracted_csv_path):
|
||||
new_csv_path = os.path.join(output_directory, f"{base_name}.csv")
|
||||
os.rename(extracted_csv_path, new_csv_path)
|
||||
|
||||
os.remove(zip_path)
|
||||
#os.remove(compressed_file)
|
||||
print(f"Decompressed and renamed '{compressed_file}' to '{new_csv_path}'.")
|
||||
|
||||
|
||||
def get_last_component(path):
|
||||
path_without_slashes = path.replace('/', '')
|
||||
return path_without_slashes
|
||||
|
||||
def download_and_process_files(bucket_number, start_timestamp, end_timestamp, sampling_stepsize, key, booleans_as_numbers, exact_match, product_type):
|
||||
output_directory = f"S3cmdData_{bucket_number}"
|
||||
|
||||
if os.path.exists(output_directory):
|
||||
shutil.rmtree(output_directory)
|
||||
|
||||
if not os.path.exists(output_directory):
|
||||
os.makedirs(output_directory)
|
||||
print(f"Directory '{output_directory}' created.")
|
||||
|
||||
filenames_to_check = list_files_in_range(start_timestamp, end_timestamp, sampling_stepsize)
|
||||
existing_files = [filename for filename in filenames_to_check if os.path.exists(os.path.join(output_directory, f"{filename}.csv"))]
|
||||
files_to_download = set(filenames_to_check) - set(existing_files)
|
||||
|
||||
if os.listdir(output_directory):
|
||||
print("Files already exist in the local folder. Skipping download.")
|
||||
else:
|
||||
if files_to_download:
|
||||
download_files(bucket_number, files_to_download, product_type)
|
||||
|
||||
# Decompress all downloaded .csv files (which are actually compressed)
|
||||
compressed_files = [os.path.join(output_directory, file) for file in os.listdir(output_directory) if file.endswith('.csv')]
|
||||
for compressed_file in compressed_files:
|
||||
decompress_file(compressed_file, output_directory)
|
||||
|
||||
csv_files = [file for file in os.listdir(output_directory) if file.endswith('.csv')]
|
||||
csv_files.sort(key=extract_timestamp)
|
||||
|
||||
|
||||
keypath = ''
|
||||
for key_item in key:
|
||||
keypath += get_last_component(key_item)
|
||||
output_csv_filename = f"{keypath}_{start_timestamp}_{bucket_number}.csv"
|
||||
with open(output_csv_filename, 'w', newline='') as csvfile:
|
||||
csv_writer = csv.writer(csvfile)
|
||||
header = ['time']
|
||||
add_header = True
|
||||
|
||||
for csv_file in csv_files:
|
||||
file_path = os.path.join(output_directory, csv_file)
|
||||
extracted_values = extract_values_by_key(file_path, key, exact_match)
|
||||
if add_header:
|
||||
add_header = False
|
||||
for values in extracted_values.values():
|
||||
first_value = values
|
||||
for first_val in first_value:
|
||||
header.append(first_val.split(';')[0].strip())
|
||||
break
|
||||
csv_writer.writerow(header)
|
||||
if extracted_values:
|
||||
for first_column, values in extracted_values.items():
|
||||
if booleans_as_numbers:
|
||||
values = [1 if value.split(';')[1].strip() == "True" else 0 if value.split(';')[1].strip() == "False" else value.split(';')[1].strip() for value in values]
|
||||
values_list = []
|
||||
values_list.append(csv_file.replace(".csv", ""))
|
||||
for i, value in enumerate(values):
|
||||
if value is None:
|
||||
value = "No value provided"
|
||||
else:
|
||||
values_list.append(value.split(';')[1].strip())
|
||||
csv_writer.writerow(values_list)
|
||||
|
||||
print(f"Extracted data saved in '{output_csv_filename}'.")
|
||||
|
||||
def parse_keys(input_string):
|
||||
keys = [key.strip() for key in input_string.split(',')]
|
||||
return keys
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description='Download files from S3 using s3cmd and extract specific values from CSV files.')
|
||||
parser.add_argument('start_timestamp', type=int, help='The start timestamp for the range (even number)')
|
||||
parser.add_argument('end_timestamp', type=int, help='The end timestamp for the range (even number)')
|
||||
parser.add_argument('--keys', type=parse_keys, required=True, help='The part to match from each CSV file, can be a single key or a comma-separated list of keys')
|
||||
parser.add_argument('--bucket-number', type=int, required=True, help='The number of the bucket to download from')
|
||||
parser.add_argument('--sampling_stepsize', type=int, required=False, default=1, help='The number of 2sec intervals, which define the length of the sampling interval in S3 file retrieval')
|
||||
parser.add_argument('--booleans_as_numbers', action="store_true", required=False, help='If key used, then booleans are converted to numbers [0/1], if key not used, then booleans maintained as text [False/True]')
|
||||
parser.add_argument('--exact_match', action="store_true", required=False, help='If key used, then key has to match exactly "=", else it is enough that key is found "in" text')
|
||||
parser.add_argument('--product_type', required=True, help='Use 0 for Salimax and 1 for Salidomo')
|
||||
|
||||
args = parser.parse_args()
|
||||
start_timestamp = args.start_timestamp
|
||||
end_timestamp = args.end_timestamp
|
||||
keys = args.keys
|
||||
bucket_number = args.bucket_number
|
||||
sampling_stepsize = args.sampling_stepsize
|
||||
booleans_as_numbers = args.booleans_as_numbers
|
||||
exact_match = args.exact_match
|
||||
# new arg for product type
|
||||
product_type = int(args.product_type)
|
||||
|
||||
if start_timestamp >= end_timestamp:
|
||||
print("Error: start_timestamp must be smaller than end_timestamp.")
|
||||
return
|
||||
download_and_process_files(bucket_number, start_timestamp, end_timestamp, sampling_stepsize, keys, booleans_as_numbers, exact_match, product_type)
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Loading…
Reference in New Issue