launch_identifiers.py

import pandas as pd
import requests

from Processing import PastStatus, PastT0s, PastName, PastID

PastT0s = PastT0s.copy()
PastStatus = PastStatus.copy().rename(
    columns={"id": "status_id", "abbrev": "status_name"}
)
PastName = PastName.copy()

data = pd.concat(
    [PastID, PastT0s, PastStatus[["status_id", "status_name"]], PastName], axis=1
)

year_selected = 1965
days_delta_limit = 1

data_year = data[data.net.dt.year == year_selected].copy().reset_index(drop=True)


def check_special_assignment(identifier_):
    if identifier_ == "1966-088A":
        return "33da85f3-08a8-4c5f-892d-8b4be3b4fae6"
    elif identifier_ == "1966-101A":
        return "7ec54e6a-8df8-468c-ad66-b990ab058502"
    return


def get_celestrak_data(identifier_):
    print(f"Getting data for {identifier_}")
    URL = f"https://celestrak.com/satcat/records.php?INTDES={identifier_}"

    celestrak_result = requests.get(URL, timeout=360)
    if celestrak_result.text == "No SATCAT records found\r\n":
        return None, None
    else:
        celestrak_json = celestrak_result.json()
    return celestrak_json[0]["OBJECT_NAME"], celestrak_json[0]["LAUNCH_DATE"]


def cleanup_string(string_):
    cleaned_string = ""
    for char in string_.lower():
        if char.isalpha():
            cleaned_string += char
        else:
            cleaned_string += " "
    return cleaned_string


def name_match(string1, string2):
    string1 = string1.split("|")[-1]
    string2 = string2.split("|")[-1]
    string1 = set(cleanup_string(string1).replace("kosmos", "cosmos").split())
    string2 = set(cleanup_string(string2).replace("kosmos", "cosmos").split())
    common_words = string1 & string2
    if not common_words:
        return False
    for word in common_words:
        if len(word) >= 2:
            return True
    return False


def set_identifier_match(index, identifier_, name_):
    print(
        f'Found match for {data_year.loc[index, "name"]} ({data_year.loc[index, "status_name"]})'
    )
    data_year.loc[index, "identifier"] = identifier_
    data_year.loc[index, "name_check"] = name_


def set_identifier_none(index):
    print(
        f'No match for {data_year.loc[index, "name"]} ({data_year.loc[index, "status_name"]})'
    )
    data_year.loc[index, "identifier"] = None
    data_year.loc[index, "name_check"] = None


count = 0
skip_next = False
for i in data_year.index:
    count += 1
    if skip_next:
        skip_next = False
        continue
    is_special = True
    while is_special:
        identifier = f'{data_year.loc[i, "net"].strftime("%Y")}-{count:03d}A'
        name, date = get_celestrak_data(identifier)
        special_case_launch_id = check_special_assignment(identifier)
        if special_case_launch_id:
            special_index = data_year[data_year.id == special_case_launch_id].index[0]
            set_identifier_match(special_index, identifier, name)
            count += 1
        else:
            is_special = False
    if date == data_year.loc[i, "net"].strftime("%Y-%m-%d"):
        # Check if identifier launch date roughly matches launch date
        if data_year.loc[i, "status_id"] == 4:
            # If launch failure, make sure name matches
            if name_match(name, data_year.loc[i, "name"]):
                # Name matches, so failed launch has an identifier
                set_identifier_match(i, identifier, name)
            else:
                # Name does not match, let's check if the next identifier matches the next launch
                identifier_2 = (
                    f'{data_year.loc[i, "net"].strftime("%Y")}-{count + 1:03d}A'
                )
                name_2, date_2 = get_celestrak_data(identifier_2)
                if name_match(
                    name_2, data_year.loc[i + 1, "name"]
                ) or date_2 == data_year.loc[i + 1, "net"].strftime("%Y-%m-%d"):
                    # Next launch matches, so failed launch has an identifier
                    set_identifier_match(i, identifier, name)
                else:
                    # Next launch does not match, so failed launch does not have an identifier
                    set_identifier_none(i)
                    count -= 1
        else:
            # If launch success, check various things
            if name_match(name, data_year.loc[i, "name"]):
                # If name matches, assign this identifier to the launch
                set_identifier_match(i, identifier, name)
            elif (
                (i < data_year.tail(1).index.values.item())
                and (
                    data_year.loc[i, "net"].strftime("%Y-%m-%d")
                    == data_year.loc[i + 1, "net"].strftime("%Y-%m-%d")
                )
                and name_match(name, data_year.loc[i + 1, "name"])
            ):
                # Next launch on the same day corresponding to the identifier
                set_identifier_match(i + 1, identifier, name)

                # Check if current launch is for next identifier
                identifier_2 = (
                    f'{data_year.loc[i, "net"].strftime("%Y")}-{(count + 1):03d}A'
                )
                name_2, date_2 = get_celestrak_data(identifier_2)
                if name_match(name_2, data_year.loc[i, "name"]):
                    # If name matches, assign this identifier to the launch
                    set_identifier_match(i, identifier_2, name_2)
                    # Skip next launch
                    skip_next = True
                else:
                    set_identifier_none(i)
                    count -= 1
            else:
                # Assign this identifier to the launch
                set_identifier_match(i, identifier, name)
    else:
        # If launch date does not match, then no identifier for this launch
        set_identifier_none(i)
        count -= 1

print("Done")