import type { StringFormat } from "../schema.ts";

const LABEL = /^[A-Za-z0-9](?:[A-Za-z0-9-]*[A-Za-z0-9])?$/;
const labels = (parts: readonly string[]) =>
  parts.every((part) => part.length <= 63 && LABEL.test(part));

export function isNsid(value: string): boolean {
  const parts = value.split(".");
  return value.length <= 317 && parts.length >= 3 && labels(parts) &&
    /^[A-Za-z]/.test(parts[0]) && /^[A-Za-z][A-Za-z0-9]*$/.test(parts.at(-1) ?? "");
}

function isDid(value: string): boolean {
  return value.length <= 2048 &&
    /^did:[a-z]+:(?:[A-Za-z0-9._:-]|%[0-9A-Fa-f]{2})*(?:[A-Za-z0-9._-]|%[0-9A-Fa-f]{2})$/.test(
      value,
    );
}

function isHandle(value: string): boolean {
  const parts = value.split(".");
  return value.length <= 253 && parts.length >= 2 && labels(parts) &&
    /^[A-Za-z]/.test(parts.at(-1) ?? "");
}

function isAtUri(value: string): boolean {
  if (value.length > 8192) return false;
  const match = /^at:\/\/([^/?#]+)(?:\/([^/?#]+)(?:\/([^/?#]+))?)?$/.exec(value);
  if (!match) return false;
  const [, authority, collection, key] = match;
  return (isDid(authority) || isHandle(authority)) &&
    (collection === undefined || isNsid(collection)) &&
    (key === undefined ||
      (key.length <= 512 && key !== "." && key !== ".." && /^[A-Za-z0-9_~.:-]+$/.test(key)));
}

function isUri(value: string): boolean {
  if (value.length > 8192 || /%(?![0-9A-Fa-f]{2})/.test(value)) return false;
  const match = /^[A-Za-z][A-Za-z0-9+.-]*:(?:\/\/([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?$/.exec(
    value,
  );
  if (!match) return false;
  const [, authority, path, query = "", fragment = ""] = match;
  if (
    !/^[A-Za-z0-9\-._~!$&'()*+,;=:@%/]*$/.test(path) ||
    !/^[A-Za-z0-9\-._~!$&'()*+,;=:@%/?]*$/.test(query) ||
    !/^[A-Za-z0-9\-._~!$&'()*+,;=:@%/?]*$/.test(fragment)
  ) return false;
  if (authority === undefined) return true;
  const host =
    /^(?:[A-Za-z0-9\-._~!$&'()*+,;=:%]*@)?(\[[^\]]+\]|[A-Za-z0-9\-._~!$&'()*+,;=%]*)(?::[0-9]*)?$/
      .exec(authority)?.[1];
  if (host === undefined) return false;
  if (!host.startsWith("[")) return true;
  if (/^\[v[0-9a-f]+\.[A-Za-z0-9\-._~!$&'()*+,;=:]+\]$/i.test(host)) return true;
  try {
    new URL(`http://${host}/`);
    return true;
  } catch {
    return false;
  }
}

function isDatetime(value: string): boolean {
  const match = /^(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2}):(\d{2})(?:\.\d+)?(Z|[+-]\d{2}:\d{2})$/
    .exec(value);
  if (!match) return false;
  const [, y, m, d, h, min, s, zone] = match;
  const year = Number(y), month = Number(m), day = Number(d);
  const leap = year % 4 === 0 && (year % 100 !== 0 || year % 400 === 0);
  const days = [31, leap ? 29 : 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31];
  if (
    month < 1 || month > 12 || day < 1 || day > days[month - 1] ||
    Number(h) > 23 || Number(min) > 59 || Number(s) > 59 || zone === "-00:00"
  ) return false;
  if (zone !== "Z") {
    const hours = Number(zone.slice(1, 3)), minutes = Number(zone.slice(4));
    if (hours > 23 || minutes > 59) return false;
    // Offsets must not move a year-zero date into the preceding year.
    if (
      year === 0 && month === 1 && day === 1 && zone[0] === "+" &&
      Number(h) * 60 + Number(min) < hours * 60 + minutes
    ) return false;
  }
  return true;
}

function isLanguage(value: string): boolean {
  // BCP 47 grandfathered tags cannot all be represented by the modern grammar.
  if (
    /^(?:en-GB-oed|i-(?:ami|bnn|default|enochian|hak|klingon|lux|mingo|navajo|pwn|tao|tay|tsu)|sgn-(?:BE-FR|BE-NL|CH-DE)|art-lojban|cel-gaulish|no-bok|no-nyn|zh-(?:guoyu|hakka|min|min-nan|xiang))$/i
      .test(value)
  ) return true;
  if (/^x(?:-[a-z0-9]{1,8})+$/i.test(value)) return true;
  const match =
    /^([a-z]{2,3}(?:-[a-z]{3}){0,3}|[a-z]{4}|[a-z]{5,8})(?:-[a-z]{4})?(?:-(?:[a-z]{2}|[0-9]{3}))?((?:-(?:[a-z0-9]{5,8}|[0-9][a-z0-9]{3}))*)((?:-[0-9a-wy-z](?:-[a-z0-9]{2,8})+)*)(?:-x(?:-[a-z0-9]{1,8})+)?$/i
      .exec(value);
  if (!match) return false;
  const [, , variantTags, extensionTags] = match;
  const variants = variantTags.toLowerCase().split("-").filter(Boolean);
  const extensions = [...extensionTags.toLowerCase().matchAll(/-([0-9a-wy-z])(?=-)/g)].map((m) =>
    m[1]
  );
  return new Set(variants).size === variants.length &&
    new Set(extensions).size === extensions.length;
}

export function cidCodec(value: string): 0x55 | 0x71 | undefined {
  // AT Protocol's blessed form: base32 CIDv1, raw/dag-cbor, SHA-256.
  if (value.length !== 59 || value[0] !== "b") return;
  const bytes: number[] = [];
  let bits = 0, accumulator = 0;
  for (const char of value.slice(1)) {
    const digit = "abcdefghijklmnopqrstuvwxyz234567".indexOf(char);
    if (digit < 0) return;
    accumulator = (accumulator << 5) | digit;
    bits += 5;
    if (bits >= 8) {
      bits -= 8;
      bytes.push((accumulator >> bits) & 255);
      accumulator &= (1 << bits) - 1;
    }
  }
  if (
    accumulator === 0 && bytes[0] === 1 && (bytes[1] === 0x55 || bytes[1] === 0x71) &&
    bytes[2] === 0x12 && bytes[3] === 32
  ) return bytes[1];
}

export function base64Length(value: string): number | undefined {
  if (!/^(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}(?:==)?|[A-Za-z0-9+/]{3}=?)?$/.test(value)) return;
  const raw = value.replace(/=+$/, "");
  const last = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/".indexOf(
    raw.at(-1) ?? "A",
  );
  const mask = raw.length % 4 === 2 ? 15 : raw.length % 4 === 3 ? 3 : 0;
  if ((last & mask) !== 0) return;
  return Math.floor(raw.length * 6 / 8);
}

export function matchesFormat(format: StringFormat, value: string): boolean {
  switch (format) {
    case "did":
      return isDid(value);
    case "handle":
      return isHandle(value);
    case "at-identifier":
      return isDid(value) || isHandle(value);
    case "at-uri":
      return isAtUri(value);
    case "datetime":
      return isDatetime(value);
    case "nsid":
      return isNsid(value);
    case "tid":
      return /^[234567abcdefghij][234567abcdefghijklmnopqrstuvwxyz]{12}$/.test(value);
    case "cid":
      return cidCodec(value) !== undefined;
    case "language":
      return isLanguage(value);
    case "uri":
      return isUri(value);
  }
}