{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import plyvel\n",
    "import struct\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# functions for reading IAVL tree\n",
    "\n",
    "def read_varint(x: bytes, offset: int = 0) -> int:\n",
    "    result = 0\n",
    "    factor = 1\n",
    "\n",
    "    for i, b in enumerate(x[offset:]):\n",
    "        if b >= 128:\n",
    "            result = result + (b - 128) * factor\n",
    "        else:\n",
    "            result = result + b * factor\n",
    "            return result // 2, offset+i+1\n",
    "        factor *= 128\n",
    "\n",
    "def read_uvarint(x: bytes, offset: int = 0) -> int:\n",
    "    result = 0\n",
    "    factor = 1\n",
    "\n",
    "    for i, b in enumerate(x[offset:]):\n",
    "        if b >= 128:\n",
    "            result = result + (b - 128) * factor\n",
    "        else:\n",
    "            result = result + b * factor\n",
    "            return result, offset+i+1\n",
    "        factor *= 128\n",
    "\n",
    "def read_key(key: bytes) -> tuple[int, int] | None:\n",
    "    if not key.startswith(b's'):\n",
    "        return None\n",
    "\n",
    "    version = struct.unpack_from('>Q', key[1:9])[0]\n",
    "    nonce = struct.unpack_from('>I', key[9:13])[0]\n",
    "\n",
    "    return (version, nonce)\n",
    "\n",
    "def write_key(key: tuple[int, int]) -> bytes:\n",
    "    version = struct.pack('>Q', key[0])\n",
    "    nonce = struct.pack('>I', key[1])\n",
    "\n",
    "    return b's' + version + nonce\n",
    "\n",
    "def read_node(node: bytes) -> tuple[int, int, bytes, tuple[int, int], tuple[int, int]] | tuple[int, int, list[int], bytes] | tuple[int, int]:\n",
    "\n",
    "    if node.startswith(b's'):\n",
    "        return read_key(node)\n",
    "\n",
    "    n = 0\n",
    "    height, n = read_varint(node, n)\n",
    "\n",
    "    if height == 0:\n",
    "        length, n = read_varint(node, n)\n",
    "        size, n = read_uvarint(node, n)\n",
    "        key = node[n:n+size]\n",
    "        n += size\n",
    "        valuesize, n = read_uvarint(node, n)\n",
    "        value = node[n:n+valuesize]\n",
    "\n",
    "        return (height, length, key, value)\n",
    "    else:\n",
    "        length, n = read_varint(node, n)\n",
    "        size, n = read_uvarint(node, n)\n",
    "        key = node[n:n+size]\n",
    "        n += size\n",
    "        hashsize, n = read_uvarint(node, n)\n",
    "        n += hashsize\n",
    "        mode, n = read_uvarint(node, n)\n",
    "        left_version, n = read_varint(node, n)\n",
    "        left_nonce, n = read_varint(node, n)\n",
    "        right_version, n = read_varint(node, n)\n",
    "        right_nonce, n = read_varint(node, n)\n",
    "\n",
    "        return (height, length, key, (left_version, left_nonce), (right_version, right_nonce))\n",
    "\n",
    "def walk(tree, version, searchkey):\n",
    "    if (version, 1) not in tree:\n",
    "        return None\n",
    "\n",
    "    node = tree[(version, 1)]\n",
    "    if len(node) == 2: # root copy?\n",
    "        node = tree[node]\n",
    "\n",
    "    while node[0] > 0:\n",
    "        nodekey = node[2]\n",
    "        if searchkey < nodekey:\n",
    "            next = node[3]\n",
    "        else:\n",
    "            next = node[4]\n",
    "\n",
    "        node = tree[next]\n",
    "\n",
    "    return node[3]\n",
    "\n",
    "def walk_disk_raw(db, prefix: bytes, version: int, searchkey: bytes) -> None | bytes:\n",
    "\n",
    "    root = db.get(prefix + write_key((version, 1)))\n",
    "    if root is None:\n",
    "        return None\n",
    "\n",
    "    node = read_node(root)\n",
    "\n",
    "    if len(node) == 2: # root copy?\n",
    "        node = read_node(db.get(prefix + write_key(node)))\n",
    "\n",
    "    while node[0] > 0:\n",
    "        # print(node)\n",
    "\n",
    "        nodekey = node[2]\n",
    "        if searchkey < nodekey:\n",
    "            next = node[3]\n",
    "        else:\n",
    "            next = node[4]\n",
    "\n",
    "        node = read_node(db.get(prefix + write_key(next)))\n",
    "\n",
    "    if node[2] == searchkey:\n",
    "        return node[3]\n",
    "    else:\n",
    "        return None\n",
    "\n",
    "def walk_disk_next_key_raw(db, prefix: bytes, version: int, searchkey: bytes) -> None | bytes:\n",
    "    root = db.get(prefix + write_key((version, 1)))\n",
    "    if root is None:\n",
    "        return None\n",
    "\n",
    "    node = read_node(root)\n",
    "    lowest_geq_key = node[2] if node[2] >= searchkey else None\n",
    "\n",
    "    if len(node) == 2: # root copy?\n",
    "        node = read_node(db.get(prefix + write_key(node)))\n",
    "\n",
    "    while node[0] > 0:\n",
    "        # print(node)\n",
    "\n",
    "        nodekey = node[2]\n",
    "        if searchkey < nodekey:\n",
    "            next = node[3]\n",
    "        else:\n",
    "            next = node[4]\n",
    "\n",
    "        node = read_node(db.get(prefix + write_key(next)))\n",
    "        if node[2] >= searchkey and (lowest_geq_key is None or node[2] < lowest_geq_key):\n",
    "            lowest_geq_key = node[2]\n",
    "\n",
    "    return lowest_geq_key\n",
    "\n",
    "def walk_disk(db, prefix: str, version: int, format: str, searchkey: list) -> None | bytes:\n",
    "    return walk_disk_raw(db, prefix.encode('utf-8'), version, encode_key(format, searchkey))\n",
    "\n",
    "def parse_struct(data):\n",
    "    n = 0\n",
    "    results = []\n",
    "\n",
    "    while n < len(data):\n",
    "        key, n = read_uvarint(data, n)\n",
    "        ty = key & 7\n",
    "        key >>= 3\n",
    "        if ty == 2:\n",
    "            l, n = read_uvarint(data, n)\n",
    "            val = data[n:n+l]\n",
    "            n += l\n",
    "        elif ty == 0:\n",
    "            val, n = read_uvarint(data, n)\n",
    "        else:\n",
    "            raise Exception(f'unknown type {ty}, {data[n:]}')\n",
    "        results.append((key, val))\n",
    "\n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# find max height\n",
    "\n",
    "def next_key(db, k: bytes) -> bytes | None:\n",
    "    it = db.iterator(start = k)\n",
    "    try:\n",
    "        nk, _ = next(it)\n",
    "        return nk\n",
    "    except StopIteration:\n",
    "        return None\n",
    "    finally:\n",
    "        it.close()\n",
    "\n",
    "def max_height(db) -> int:\n",
    "    testnr = 1<<63\n",
    "\n",
    "    for i in range(62, -1, -1):\n",
    "        prefix = b's/k:emissions/s'\n",
    "        n = next_key(db, prefix + struct.pack('>Q', testnr))\n",
    "\n",
    "        if n is not None and n.startswith(prefix):\n",
    "            # print(f'{testnr:16x} is low')\n",
    "            testnr += 1 << i\n",
    "        else:\n",
    "            # print(f'{testnr:16x} is high')\n",
    "            testnr -= 1 << i\n",
    "\n",
    "    n = next_key(db, prefix + struct.pack('>Q', testnr))\n",
    "    if n is not None and n.startswith(prefix):\n",
    "        return testnr\n",
    "    else:\n",
    "        return testnr - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# encode and decode keys\n",
    "\n",
    "def encode_key(format: str, key: list) -> bytes:\n",
    "    result_bytes = []\n",
    "\n",
    "    result_bytes.append(key[0])\n",
    "\n",
    "    for i, f in enumerate(format):\n",
    "        if i >= len(key) - 1:\n",
    "            break\n",
    "        if f == 's':\n",
    "            result_bytes += list(key[i+1].encode('utf-8'))\n",
    "            if i < len(format) - 1:\n",
    "                result_bytes += [0]\n",
    "        elif f == 'Q':\n",
    "            result_bytes += list(struct.pack('>Q', key[i+1]))\n",
    "        elif f == 'q':\n",
    "            result_bytes += list(struct.pack('>Q', key[i+1] + (1<<63)))\n",
    "\n",
    "    return bytes(result_bytes)\n",
    "\n",
    "def decode_key(format: str, key: bytes) -> list:\n",
    "    result = []\n",
    "\n",
    "    result.append(key[0])\n",
    "    idx = 1\n",
    "\n",
    "    for f in format:\n",
    "        if f == 's':\n",
    "            end = key[idx:].find(b'\\x00')\n",
    "            if end < 0:\n",
    "                result.append(key[idx:].decode('utf-8'))\n",
    "                break\n",
    "            else:\n",
    "                result.append(key[idx:idx+end].decode('utf-8'))\n",
    "                idx += end + 1\n",
    "        elif f == 'Q':\n",
    "            v = struct.unpack('>Q', key[idx:idx+8])[0]\n",
    "            result.append(v)\n",
    "            idx += 8\n",
    "        elif f == 'q':\n",
    "            v = struct.unpack('>Q', key[idx:idx+8])[0]\n",
    "            result.append(v - (1<<63))\n",
    "            idx += 8\n",
    "\n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "db = plyvel.DB('../node/nodedir/data/application.db')\n",
    "height = max_height(db)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get\n",
    "v = walk_disk(db, 's/k:emissions/', height, 'Qs', [11, 37, 'allo1mn4d32hwyn6grp89akek52arsw2vcdqezr0dc7'])\n",
    "parse_struct(v)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [],
   "source": [
    "class IAVLTreeIteratorRaw:\n",
    "    def __init__(self, db, prefix: bytes, version: int, start: bytes | None = None, end: bytes | None = None):\n",
    "        self.db = db\n",
    "        self.prefix = prefix\n",
    "        self.version = version\n",
    "        self.start = start\n",
    "        self.end = end\n",
    "        self.stack = []\n",
    "\n",
    "    def __iter__(self):\n",
    "        return self\n",
    "\n",
    "    def __next__(self):\n",
    "        if len(self.stack) == 0:\n",
    "            # get root node\n",
    "            root = db.get(self.prefix + write_key((self.version, 1)))\n",
    "            if root is None:\n",
    "                raise StopIteration\n",
    "            node = read_node(root)\n",
    "            if len(node) == 2: # link to other root node\n",
    "                node = read_node(db.get(self.prefix + write_key(node)))\n",
    "            self.stack.append(((self.version, 1), node))\n",
    "\n",
    "            # walk tree to either last before start or first after start\n",
    "            while node[0] > 0:\n",
    "                # print(node)\n",
    "                nodekey = node[2]\n",
    "                if self.start is None or self.start < nodekey:\n",
    "                    next = node[3]\n",
    "                else:\n",
    "                    next = node[4]\n",
    "                node = read_node(db.get(self.prefix + write_key(next)))\n",
    "                self.stack.append((next, node))\n",
    "\n",
    "            # return early if we ended up at first item after start\n",
    "            if self.start is None or node[2] >= self.start:\n",
    "                return (node[2], node[3])\n",
    "\n",
    "        # print('Stack:', [x[0] for x in self.stack])\n",
    "\n",
    "        # go up to first parent which we're a left child of\n",
    "        key = None\n",
    "        for i in range(len(self.stack)-1, 0, -1):\n",
    "            current_key = self.stack[i][0]\n",
    "            parent_node = self.stack[i-1][1]\n",
    "            self.stack.pop()\n",
    "            left = parent_node[3]\n",
    "            right = parent_node[4]\n",
    "            if current_key == left:\n",
    "                key = right\n",
    "                break\n",
    "\n",
    "        # are we at the right end of the tree?\n",
    "        if key is None:\n",
    "            raise StopIteration\n",
    "\n",
    "        # go right\n",
    "        node = read_node(db.get(self.prefix + write_key(key)))\n",
    "        self.stack.append((key, node))\n",
    "\n",
    "        # go left until at a leaf\n",
    "        while node[0] > 0:\n",
    "            key = node[3]\n",
    "            node = read_node(db.get(self.prefix + write_key(key)))\n",
    "            self.stack.append((key, node))\n",
    "\n",
    "        if self.end is not None and node[2] >= self.end:\n",
    "            raise StopIteration\n",
    "\n",
    "        return (node[2], node[3])\n",
    "\n",
    "class IAVLTreeIterator:\n",
    "    def __init__(self, db, prefix: str, version: int, format: str, start: list | None = None, end: list | None = None):\n",
    "        self.format = format\n",
    "        start_enc = encode_key(format, start) if start is not None else None\n",
    "        end_enc = encode_key(format, end) if end is not None else None\n",
    "        self.inner = IAVLTreeIteratorRaw(db, prefix.encode('utf-8'), version, start = start_enc, end = end_enc)\n",
    "\n",
    "    def __iter__(self):\n",
    "        return self\n",
    "\n",
    "    def __next__(self):\n",
    "        (k, v) = next(self.inner)\n",
    "        return (decode_key(self.format, k), v)\n",
    "\n",
    "def iterate(db, prefix, version, format, field):\n",
    "    return IAVLTreeIterator(db, prefix, version, format, start = [field], end = [field+1] if field < 255 else None)\n",
    "\n",
    "# [k for k, v in IAVLTreeIterator(db, 's/k:mint/', height, '')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[(k, parse_struct(v)) for k, v in iterate(db, 's/k:emissions/', height, 'Q', 5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[k for k, v in IAVLTreeIterator(db, 's/k:emissions/', height, 'Q', start = [5], end = [6])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "encode_key('Q', [5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[decode_key('Q', k) for k,v in IAVLTreeIteratorRaw(db, b's/k:emissions/', height, start = bytes([5]), end = bytes([6]))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "encode_key('', [1]), decode_key('', b'\\x8a')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "decode_key()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(range(5, 0, -1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "read_node(db.get(b's/k:emissions/' + write_key((height, 1))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "heights = np.arange(5776000, height+1)\n",
    "sizes = np.empty(len(heights), dtype = int)\n",
    "\n",
    "for i in range(len(heights)):\n",
    "    h = heights[i]\n",
    "    n = db.get(b's/k:emissions/s' + struct.pack('>Q', h) + struct.pack('>I', 1))\n",
    "    nd = read_node(n)\n",
    "    sizes[i] = nd[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.plot(sizes)\n",
    "# plt.xlim(0, 1000)\n",
    "# plt.ylim(19906000, 19908000)\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "n = db.get(b's/k:emissions/s' + struct.pack('>Q', 5776000) + struct.pack('>I', 1))\n",
    "nd = read_node(n)\n",
    "# decode_key('Qss', nd[2])\n",
    "nd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def has_prefix(db, prefix: bytes) -> bool:\n",
    "#     it = db.iterator(start = prefix)\n",
    "#     try:\n",
    "#         first_key, _ = next(it)\n",
    "#         return first_key.startswith(prefix)\n",
    "#     except StopIteration:\n",
    "#         return False\n",
    "#     finally:\n",
    "#         it.close()\n",
    "\n",
    "# found = []\n",
    "# prefixes = [b's/']\n",
    "\n",
    "# for i in range(20):\n",
    "#     new_prefixes = []\n",
    "\n",
    "#     for p in prefixes:\n",
    "#         for i in range(256):\n",
    "#             b = p + bytes([i])\n",
    "#             if has_prefix(db, b):\n",
    "#                 if b.endswith(b'/'):\n",
    "#                     found.append((b, db.approximate_size(b, b + b'\\xff')))\n",
    "#                 else:\n",
    "#                     new_prefixes.append(b)\n",
    "#                 # new_prefixes.append(b)\n",
    "\n",
    "#     prefixes = new_prefixes\n",
    "\n",
    "# found"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}