Add script to fetch and process IBAN registry data

Include a new Python script that retrieves IBAN registry data from SWIFT and processes it into a structured JSON format,
along with a PDF document for reference.

docs/international_wide_ibans/get_iban_registry.py

@@ -0,0 +1,86 @@
+#!/usr/bin/env python
+import json
+import re
+from urllib.parse import urljoin
+
+import requests
+from bs4 import BeautifulSoup
+
+
+COUNTRY_CODE_PATTERN = r"[A-Z]{2}"
+EMPTY_RANGE = (0, 0)
+URL = "https://www.swift.com/standards/data-standards/iban"
+
+
+def get_raw():
+    soup = BeautifulSoup(requests.get(URL).content, "html.parser")
+    link = soup.find("a", attrs={"data-tracking-title": "IBAN Registry (TXT)"})
+    return requests.get(urljoin(URL, link["href"])).content.decode(encoding="latin1")
+
+
+def parse_int(raw):
+    return int(re.search(r"\d+", raw).group())
+
+
+def parse_range(raw):
+    pattern = r".*?(?P<from>\d+)\s*-\s*(?P<to>\d+)"
+    match = re.search(pattern, raw)
+    if not match:
+        return EMPTY_RANGE
+    return (int(match["from"]) - 1, int(match["to"]))
+
+
+def parse(raw):
+    columns = {}
+    for line in raw.split("\r\n"):
+        header, *rows = line.split("\t")
+        if header == "IBAN prefix country code (ISO 3166)":
+            columns["country"] = [re.search(COUNTRY_CODE_PATTERN, item).group() for item in rows]
+        elif header == "Country code includes other countries/territories":
+            columns["other_countries"] = [re.findall(COUNTRY_CODE_PATTERN, item) for item in rows]
+        elif header == "BBAN structure":
+            columns["bban_spec"] = rows
+        elif header == "BBAN length":
+            columns["bban_length"] = [parse_int(item) for item in rows]
+        elif header == "Bank identifier position within the BBAN":
+            columns["bank_code_position"] = [parse_range(item) for item in rows]
+        elif header == "Branch identifier position within the BBAN":
+            columns["branch_code_position"] = [parse_range(item) for item in rows]
+        elif header == "IBAN structure":
+            columns["iban_spec"] = rows
+        elif header == "IBAN length":
+            columns["iban_length"] = [parse_int(item) for item in rows]
+    return [dict(zip(columns.keys(), row)) for row in zip(*columns.values())]
+
+
+def process(records):
+    registry = {}
+    for record in records:
+        country_codes = [record["country"]]
+        country_codes.extend(record["other_countries"])
+        for code in country_codes:
+            registry[code] = {
+                "bban_spec": record["bban_spec"],
+                "iban_spec": record["iban_spec"],
+                "bban_length": record["bban_length"],
+                "iban_length": record["iban_length"],
+                "positions": process_positions(record),
+            }
+    return registry
+
+
+def process_positions(record):
+    bank_code = record["bank_code_position"]
+    branch_code = record["branch_code_position"]
+    if branch_code == EMPTY_RANGE:
+        branch_code = (bank_code[1], bank_code[1])
+    return {
+        "account_code": (max(bank_code[1], branch_code[1]), record["bban_length"]),
+        "bank_code": bank_code,
+        "branch_code": branch_code,
+    }
+
+
+if __name__ == "__main__":
+    with open("schwifty/iban_registry/generated.json", "w+") as fp:
+        json.dump(process(parse(get_raw())), fp, indent=2)