High speed conversion of IP addresses represented in CIDR notation into their corresponding start and end IPs, along with their respective subnet masks.
The cidr_to_ip_and_subnet_mask function serves as a versatile tool for converting IP addresses represented in CIDR (Classless Inter-Domain Routing) notation into their corresponding start and end IPs, along with their respective subnet masks. This process is crucial in network management and data analysis tasks. By automating this conversion, the function significantly accelerates the handling of large datasets containing CIDR notation IP addresses. It accepts various input formats, including lists, pandas Series, and DataFrames, enhancing its adaptability. Leveraging optimized array operations through NumPy and numexpr, the function ensures efficient processing, particularly with extensive datasets. This functionality is valuable to network administrators, data scientists, security professionals, and developers alike, providing a streamlined approach for tasks involving IP address manipulation and analysis. Ultimately, it simplifies the management of network configurations and enhances the efficiency of data processing pipelines that involve IP address transformations.
It automates the process of converting CIDR notation IP addresses to start and end IP addresses along with subnet masks. This can save a significant amount of time and effort compared to manual conversion.
It can handle a large number of CIDR notation IP addresses efficiently, making it suitable for processing datasets with a large number of IP addresses.
The function can accept input in various formats, including lists, pandas Series, and DataFrames. This makes it versatile and adaptable to different data structures.
The function leverages NumPy and numexpr for efficient array operations, which can lead to improved performance, especially with large datasets.
The function is well-organized and includes meaningful variable names, making it easy for others (and the original developer) to understand and reuse the code.
from cirdhighspeedcoverter import cidr_to_ip_and_subnet_mask
df2 = pd.read_csv(
r"C:\Users\hansc\Downloads\GeoLite2-City-CSV_20230908\GeoLite2-City-CSV_20230908\GeoLite2-City-Blocks-IPv4.csv"
)
print(df2[:10].to_string())
df = cidr_to_ip_and_subnet_mask(df2[:1000].network.to_list())
df = cidr_to_ip_and_subnet_mask(df2[:1000].network)
df = cidr_to_ip_and_subnet_mask(df2[:1000], column="network")
print(df[:10].to_string())
network geoname_id registered_country_geoname_id represented_country_geoname_id is_anonymous_proxy is_satellite_provider postal_code latitude longitude accuracy_radius
0 1.0.0.0/24 2077456.0 2077456.0 NaN 0 0 NaN -33.4940 143.2104 1000.0
1 1.0.1.0/24 1814991.0 1814991.0 NaN 0 0 NaN 34.7732 113.7220 1000.0
2 1.0.2.0/23 1814991.0 1814991.0 NaN 0 0 NaN 34.7732 113.7220 1000.0
3 1.0.4.0/22 2147714.0 2077456.0 NaN 0 0 2000 -33.8715 151.2006 1000.0
4 1.0.8.0/21 1814991.0 1814991.0 NaN 0 0 NaN 34.7732 113.7220 1000.0
5 1.0.16.0/20 1861060.0 1861060.0 NaN 0 0 NaN 35.6897 139.6895 500.0
6 1.0.32.0/19 1814991.0 1814991.0 NaN 0 0 NaN 34.7732 113.7220 1000.0
7 1.0.64.0/22 1862415.0 1861060.0 NaN 0 0 730-0851 34.3927 132.4501 5.0
8 1.0.68.0/23 11818936.0 1861060.0 NaN 0 0 739-0424 34.2976 132.2898 20.0
9 1.0.70.0/25 1856520.0 1861060.0 NaN 0 0 730-0011 34.3978 132.4525 10.0
aa_startip aa_subnet aa_startip_int aa_endip aa_endip_int aa_subnetmask
0 1.0.0.0 24 16777216 1.0.0.255 16777471 255.255.255.0
1 1.0.1.0 24 16777472 1.0.1.255 16777727 255.255.255.0
2 1.0.2.0 23 16777728 1.0.3.255 16778239 255.255.254.0
3 1.0.4.0 22 16778240 1.0.7.255 16779263 255.255.252.0
4 1.0.8.0 21 16779264 1.0.15.255 16781311 255.255.248.0
5 1.0.16.0 20 16781312 1.0.31.255 16785407 255.255.240.0
6 1.0.32.0 19 16785408 1.0.63.255 16793599 255.255.224.0
7 1.0.64.0 22 16793600 1.0.67.255 16794623 255.255.252.0
8 1.0.68.0 23 16794624 1.0.69.255 16795135 255.255.254.0
9 1.0.70.0 25 16795136 1.0.70.127 16795263 255.255.255.128
Convert CIDR notation IP addresses to start and end IP addresses along with subnet masks.
This function takes a list or pandas DataFrame/Series containing CIDR notation IP addresses
and returns a DataFrame with the following columns:
- 'aa_startip': The starting IP address in string format.
- "aa_subnet": The subnet mask in integer format (uint8).
- 'aa_endip': The ending IP address in string format.
- 'aa_startip_int': The starting IP address in integer format (uint32).
- 'aa_endip_int': The ending IP address in integer format (uint32).
- 'aa_subnetmask': The subnet mask in string format.
Parameters:
-----------
df_series_list : list, pandas.Series, or pandas.DataFrame
The input data containing CIDR notation IP addresses.
column : str, optional (default="network")
The name of the column containing the CIDR notation IP addresses if df_series_list is a DataFrame.
Returns:
--------
pandas.DataFrame
A DataFrame with the converted IP addresses and subnet masks.