|
| 1 | +""" |
| 2 | +Plus API client handler for fetching data from the IPinfo Plus service. |
| 3 | +""" |
| 4 | + |
| 5 | +import time |
| 6 | +from ipaddress import IPv4Address, IPv6Address |
| 7 | + |
| 8 | +import requests |
| 9 | + |
| 10 | +from . import handler_utils |
| 11 | +from .bogon import is_bogon |
| 12 | +from .cache.default import DefaultCache |
| 13 | +from .data import ( |
| 14 | + continents, |
| 15 | + countries, |
| 16 | + countries_currencies, |
| 17 | + countries_flags, |
| 18 | + eu_countries, |
| 19 | +) |
| 20 | +from .details import Details |
| 21 | +from .error import APIError |
| 22 | +from .exceptions import RequestQuotaExceededError, TimeoutExceededError |
| 23 | +from .handler_utils import ( |
| 24 | + BATCH_MAX_SIZE, |
| 25 | + BATCH_REQ_TIMEOUT_DEFAULT, |
| 26 | + CACHE_MAXSIZE, |
| 27 | + CACHE_TTL, |
| 28 | + PLUS_API_URL, |
| 29 | + REQUEST_TIMEOUT_DEFAULT, |
| 30 | + cache_key, |
| 31 | +) |
| 32 | + |
| 33 | + |
| 34 | +class HandlerPlus: |
| 35 | + """ |
| 36 | + Allows client to request data for specified IP address using the Plus API. |
| 37 | + Plus API provides enhanced data including mobile carrier info and privacy detection. |
| 38 | + Instantiates and maintains access to cache. |
| 39 | + """ |
| 40 | + |
| 41 | + def __init__(self, access_token=None, **kwargs): |
| 42 | + """ |
| 43 | + Initialize the HandlerPlus object with country name list and the |
| 44 | + cache initialized. |
| 45 | + """ |
| 46 | + self.access_token = access_token |
| 47 | + |
| 48 | + # load countries file |
| 49 | + self.countries = kwargs.get("countries") or countries |
| 50 | + |
| 51 | + # load eu countries file |
| 52 | + self.eu_countries = kwargs.get("eu_countries") or eu_countries |
| 53 | + |
| 54 | + # load countries flags file |
| 55 | + self.countries_flags = kwargs.get("countries_flags") or countries_flags |
| 56 | + |
| 57 | + # load countries currency file |
| 58 | + self.countries_currencies = ( |
| 59 | + kwargs.get("countries_currencies") or countries_currencies |
| 60 | + ) |
| 61 | + |
| 62 | + # load continent file |
| 63 | + self.continents = kwargs.get("continent") or continents |
| 64 | + |
| 65 | + # setup req opts |
| 66 | + self.request_options = kwargs.get("request_options", {}) |
| 67 | + if "timeout" not in self.request_options: |
| 68 | + self.request_options["timeout"] = REQUEST_TIMEOUT_DEFAULT |
| 69 | + |
| 70 | + # setup cache |
| 71 | + if "cache" in kwargs: |
| 72 | + self.cache = kwargs["cache"] |
| 73 | + else: |
| 74 | + cache_options = kwargs.get("cache_options", {}) |
| 75 | + if "maxsize" not in cache_options: |
| 76 | + cache_options["maxsize"] = CACHE_MAXSIZE |
| 77 | + if "ttl" not in cache_options: |
| 78 | + cache_options["ttl"] = CACHE_TTL |
| 79 | + self.cache = DefaultCache(**cache_options) |
| 80 | + |
| 81 | + # setup custom headers |
| 82 | + self.headers = kwargs.get("headers", None) |
| 83 | + |
| 84 | + def getDetails(self, ip_address=None, timeout=None): |
| 85 | + """ |
| 86 | + Get Plus details for specified IP address as a Details object. |
| 87 | +
|
| 88 | + If `timeout` is not `None`, it will override the client-level timeout |
| 89 | + just for this operation. |
| 90 | + """ |
| 91 | + # If the supplied IP address uses the objects defined in the built-in |
| 92 | + # module ipaddress extract the appropriate string notation before |
| 93 | + # formatting the URL. |
| 94 | + if isinstance(ip_address, IPv4Address) or isinstance(ip_address, IPv6Address): |
| 95 | + ip_address = ip_address.exploded |
| 96 | + |
| 97 | + # check if bogon. |
| 98 | + if ip_address and is_bogon(ip_address): |
| 99 | + details = {} |
| 100 | + details["ip"] = ip_address |
| 101 | + details["bogon"] = True |
| 102 | + return Details(details) |
| 103 | + |
| 104 | + # check cache first. |
| 105 | + try: |
| 106 | + cached_data = self.cache[cache_key(ip_address)] |
| 107 | + return Details(cached_data) |
| 108 | + except KeyError: |
| 109 | + pass |
| 110 | + |
| 111 | + # prepare req http opts |
| 112 | + req_opts = {**self.request_options} |
| 113 | + if timeout is not None: |
| 114 | + req_opts["timeout"] = timeout |
| 115 | + |
| 116 | + # Build URL |
| 117 | + url = PLUS_API_URL |
| 118 | + if ip_address: |
| 119 | + url += "/" + ip_address |
| 120 | + |
| 121 | + headers = handler_utils.get_headers(self.access_token, self.headers) |
| 122 | + response = requests.get(url, headers=headers, **req_opts) |
| 123 | + |
| 124 | + if response.status_code == 429: |
| 125 | + raise RequestQuotaExceededError() |
| 126 | + if response.status_code >= 400: |
| 127 | + error_code = response.status_code |
| 128 | + content_type = response.headers.get("Content-Type") |
| 129 | + if content_type == "application/json": |
| 130 | + error_response = response.json() |
| 131 | + else: |
| 132 | + error_response = {"error": response.text} |
| 133 | + raise APIError(error_code, error_response) |
| 134 | + |
| 135 | + details = response.json() |
| 136 | + |
| 137 | + # Format and cache |
| 138 | + self._format_plus_details(details) |
| 139 | + self.cache[cache_key(ip_address)] = details |
| 140 | + |
| 141 | + return Details(details) |
| 142 | + |
| 143 | + def _format_plus_details(self, details): |
| 144 | + """ |
| 145 | + Format Plus response details. |
| 146 | + Plus has nested geo and as objects that need special formatting. |
| 147 | + """ |
| 148 | + # Format geo object if present |
| 149 | + if "geo" in details and details["geo"]: |
| 150 | + geo = details["geo"] |
| 151 | + if "country_code" in geo: |
| 152 | + country_code = geo["country_code"] |
| 153 | + geo["country_name"] = self.countries.get(country_code) |
| 154 | + geo["isEU"] = country_code in self.eu_countries |
| 155 | + geo["country_flag"] = self.countries_flags.get(country_code) |
| 156 | + geo["country_currency"] = self.countries_currencies.get(country_code) |
| 157 | + geo["continent"] = self.continents.get(country_code) |
| 158 | + geo["country_flag_url"] = ( |
| 159 | + f"{handler_utils.COUNTRY_FLAGS_URL}{country_code}.svg" |
| 160 | + ) |
| 161 | + |
| 162 | + # Top-level country_code might also exist in some responses |
| 163 | + if "country_code" in details: |
| 164 | + country_code = details["country_code"] |
| 165 | + details["country_name"] = self.countries.get(country_code) |
| 166 | + details["isEU"] = country_code in self.eu_countries |
| 167 | + details["country_flag"] = self.countries_flags.get(country_code) |
| 168 | + details["country_currency"] = self.countries_currencies.get(country_code) |
| 169 | + details["continent"] = self.continents.get(country_code) |
| 170 | + details["country_flag_url"] = ( |
| 171 | + f"{handler_utils.COUNTRY_FLAGS_URL}{country_code}.svg" |
| 172 | + ) |
| 173 | + |
| 174 | + def getBatchDetails( |
| 175 | + self, |
| 176 | + ip_addresses, |
| 177 | + batch_size=None, |
| 178 | + timeout_per_batch=BATCH_REQ_TIMEOUT_DEFAULT, |
| 179 | + timeout_total=None, |
| 180 | + raise_on_fail=True, |
| 181 | + ): |
| 182 | + """ |
| 183 | + Get Plus details for a batch of IP addresses at once. |
| 184 | +
|
| 185 | + There is no specified limit to the number of IPs this function can |
| 186 | + accept; it can handle as much as the user can fit in RAM (along with |
| 187 | + all of the response data, which is at least a magnitude larger than the |
| 188 | + input list). |
| 189 | +
|
| 190 | + The input list is broken up into batches to abide by API requirements. |
| 191 | + The batch size can be adjusted with `batch_size` but is clipped to |
| 192 | + `BATCH_MAX_SIZE`. |
| 193 | + Defaults to `BATCH_MAX_SIZE`. |
| 194 | +
|
| 195 | + For each batch, `timeout_per_batch` indicates the maximum seconds to |
| 196 | + spend waiting for the HTTP request to complete. If any batch fails with |
| 197 | + this timeout, the whole operation fails. |
| 198 | + Defaults to `BATCH_REQ_TIMEOUT_DEFAULT` seconds. |
| 199 | +
|
| 200 | + `timeout_total` is a seconds-denominated hard-timeout for the time |
| 201 | + spent in HTTP operations; regardless of whether all batches have |
| 202 | + succeeded so far, if `timeout_total` is reached, the whole operation |
| 203 | + will fail by raising `TimeoutExceededError`. |
| 204 | + Defaults to being turned off. |
| 205 | +
|
| 206 | + `raise_on_fail`, if turned off, will return any result retrieved so far |
| 207 | + rather than raise an exception when errors occur, including timeout and |
| 208 | + quota errors. |
| 209 | + Defaults to on. |
| 210 | + """ |
| 211 | + if batch_size == None: |
| 212 | + batch_size = BATCH_MAX_SIZE |
| 213 | + |
| 214 | + result = {} |
| 215 | + lookup_addresses = [] |
| 216 | + |
| 217 | + # pre-populate with anything we've got in the cache, and keep around |
| 218 | + # the IPs not in the cache. |
| 219 | + for ip_address in ip_addresses: |
| 220 | + # if the supplied IP address uses the objects defined in the |
| 221 | + # built-in module ipaddress extract the appropriate string notation |
| 222 | + # before formatting the URL. |
| 223 | + if isinstance(ip_address, IPv4Address) or isinstance( |
| 224 | + ip_address, IPv6Address |
| 225 | + ): |
| 226 | + ip_address = ip_address.exploded |
| 227 | + |
| 228 | + if ip_address and is_bogon(ip_address): |
| 229 | + details = {} |
| 230 | + details["ip"] = ip_address |
| 231 | + details["bogon"] = True |
| 232 | + result[ip_address] = Details(details) |
| 233 | + else: |
| 234 | + try: |
| 235 | + cached_data = self.cache[cache_key(ip_address)] |
| 236 | + result[ip_address] = Details(cached_data) |
| 237 | + except KeyError: |
| 238 | + lookup_addresses.append(ip_address) |
| 239 | + |
| 240 | + # all in cache - return early. |
| 241 | + if len(lookup_addresses) == 0: |
| 242 | + return result |
| 243 | + |
| 244 | + # do start timer if necessary |
| 245 | + if timeout_total is not None: |
| 246 | + start_time = time.time() |
| 247 | + |
| 248 | + # prepare req http options |
| 249 | + req_opts = {**self.request_options, "timeout": timeout_per_batch} |
| 250 | + |
| 251 | + # loop over batch chunks and do lookup for each. |
| 252 | + url = "https://api.ipinfo.io/batch" |
| 253 | + headers = handler_utils.get_headers(self.access_token, self.headers) |
| 254 | + headers["content-type"] = "application/json" |
| 255 | + |
| 256 | + for i in range(0, len(lookup_addresses), batch_size): |
| 257 | + # quit if total timeout is reached. |
| 258 | + if timeout_total is not None and time.time() - start_time > timeout_total: |
| 259 | + return handler_utils.return_or_fail( |
| 260 | + raise_on_fail, TimeoutExceededError(), result |
| 261 | + ) |
| 262 | + |
| 263 | + chunk = lookup_addresses[i : i + batch_size] |
| 264 | + |
| 265 | + # lookup |
| 266 | + try: |
| 267 | + response = requests.post(url, json=chunk, headers=headers, **req_opts) |
| 268 | + except Exception as e: |
| 269 | + return handler_utils.return_or_fail(raise_on_fail, e, result) |
| 270 | + |
| 271 | + # fail on bad status codes |
| 272 | + try: |
| 273 | + if response.status_code == 429: |
| 274 | + raise RequestQuotaExceededError() |
| 275 | + response.raise_for_status() |
| 276 | + except Exception as e: |
| 277 | + return handler_utils.return_or_fail(raise_on_fail, e, result) |
| 278 | + |
| 279 | + # Process batch response |
| 280 | + json_response = response.json() |
| 281 | + |
| 282 | + for ip_address, data in json_response.items(): |
| 283 | + # Cache and format the data |
| 284 | + if isinstance(data, dict) and not data.get("bogon"): |
| 285 | + self._format_plus_details(data) |
| 286 | + self.cache[cache_key(ip_address)] = data |
| 287 | + result[ip_address] = Details(data) |
| 288 | + |
| 289 | + return result |
0 commit comments