Python library to interact with the Astreum blockchain and its virtual machine.
Project description
lib
Python library to interact with the Astreum blockchain and its virtual machine.
Configuration
When initializing an astreum.Node, pass a dictionary with any of the options below. Only the parameters you want to override need to be present – everything else falls back to its default.
Core Configuration
| Parameter | Type | Default | Description |
|---|---|---|---|
chain |
string | "test" |
Chain name ("main" or "test"). If chain_id is omitted, main maps to 1; all other values default to test (0). |
chain_id |
int | 0 |
Chain identifier used for validation (0 = test, 1 = main). If chain is omitted, chain is derived from this value (1 => main, otherwise test). |
hot_storage_limit |
int | 1073741824 |
Maximum bytes kept in the hot cache before new atoms are skipped (1 GiB). |
cold_storage_limit |
int | 10737418240 |
Cold storage write threshold (10 GiB by default); set to 0 to skip the limit. |
cold_storage_path |
string | None |
Directory where persisted atoms live; Astreum creates it on startup and skips cold storage when unset. |
cold_storage_scale |
string | "MB" |
Base unit for cold storage roll-up thresholds (KB, MB, or GB). This sets the derived cold_storage_base_size used for level_0 collation and higher-level merges. |
atom_fetch_interval |
float | 0.25 |
Poll interval (seconds) while waiting for missing atoms in get_atom_list_from_storage; 0 disables waiting. |
atom_fetch_retries |
int | 8 |
Number of poll attempts for missing atoms; max wait is roughly interval * retries, 0 disables waiting. |
verify_blockchain_interval |
float | 10.0 |
Delay (seconds) between consensus verification worker iterations. Defaults to peer_timeout_interval when not explicitly set. |
verification_max_stale_seconds |
int | 10 |
Ignore otherwise-valid candidate heads whose block timestamp is older than this many seconds when selecting the latest verified chain head. |
verification_max_future_skew |
int | 2 |
Ignore candidate heads whose block timestamp is more than this many seconds in the future when selecting the latest verified chain head. |
latest_block_hash |
hex string | None |
Optional 32-byte block-hash override used to preload the node's starting latest_block_hash from config. |
verified_up_to |
hex string | None |
Optional 32-byte hash override used to preload the verification anchor (node.verified_up_to) from config. |
logging_enabled |
bool | True |
When False, disable logger setup entirely, including file creation and the background logging listener thread. |
logging_retention_days |
int | 7 |
Number of days to keep rotated log files (daily gzip). |
verbose |
bool | False |
When True, also mirror JSON logs to stdout with a human-readable format. |
Communication
| Parameter | Type | Default | Description |
|---|---|---|---|
relay_secret_key |
hex string | Auto-generated | X25519 private key used for the relay route; a new keypair is created when this field is omitted. |
relay_payment_secret_key |
hex string | None |
Optional Ed25519 private key used for relay/storage payment channels; when set, the node can advertise a relay payment public key for paid objects. |
validation_secret_key |
hex string | None |
Optional Ed25519 key that lets the node join the validation route; leave blank to opt out of validation. |
use_ipv6 |
bool | False |
Bind the incoming/outgoing sockets on IPv6 (the OS still listens on IPv4 if a peer speaks both). |
incoming_port |
int | 52780 |
UDP port the relay binds to; pass 0 or omit to let the OS pick an ephemeral port. |
default_seed |
string | "bootstrap.astreum.org:52780" |
Default address to ping before joining; set to None to disable the built-in default. |
additional_seeds |
list[str] | [] |
Extra addresses appended to the bootstrap list; each must look like host:port or [ipv6]:port. |
peer_timeout |
int | 900 |
Evict peers that have not been seen within this many seconds (15 minutes). |
peer_timeout_interval |
int | 10 |
How often (seconds) the peer manager checks for stale peers. |
bootstrap_retry_interval |
int | 30 |
How often (seconds) to retry bootstrapping when the peer list is empty. |
storage_index_interval |
int | 600 |
How often (seconds) to re-advertise entries in node.atom_advertisments to the closest known peer. |
storage_request_minimum_price |
int | 1 |
Floor price for storage/object requests; the dynamic storage request price never drops below this value. |
storage_request_price_interval |
float | 5.0 |
How often (seconds) the storage thread recomputes request pricing from inbound queue pressure. |
fair_use_limit |
int | 1048576 |
Bytes a peer may receive via shared object uploads before fair-use ratio enforcement begins (1 MiB by default). |
fair_use_ratio |
float | 0.5 |
Minimum download/upload ratio a peer must maintain after fair_use_limit is exceeded; set 0 to disable the fair-use gate. |
incoming_queue_size_limit |
int | 67108864 |
Soft cap (bytes) for inbound queue usage tracked by enqueue_incoming; set to 0 to disable. |
incoming_queue_timeout |
float | 1.0 |
When > 0, enqueue_incoming waits up to this many seconds for space before dropping the payload. |
Advertisements: node.atom_advertisments holds (atom_id, payload_type, expires_at) tuples. Use node.add_atom_advertisement or node.add_atom_advertisements to enqueue entries (expires_at=None keeps them indefinite). Validators automatically advertise block, transaction (main and detail lists), receipt, and account trie lists for 15 minutes by default.
Note The peer‑to‑peer route used for object discovery is always enabled. If
validation_secret_keyis provided the node automatically joins the validation route too.
Usage
from astreum.node import Node
config = {
"relay_secret_key": "ab…cd", # optional – hex encoded
"validation_secret_key": "12…34", # optional – validator
"hot_storage_limit": 1073741824, # cap hot cache at 1 GiB
"cold_storage_limit": 10737418240, # cap cold storage at 10 GiB
"cold_storage_path": "./data/node1",
"incoming_port": 52780,
"use_ipv6": False,
"default_seed": None,
"additional_seeds": [
"127.0.0.1:7374"
]
}
node = Node(config)
# … your code …
Validation Overview
Call node.verify() to connect the node, initialize fork tracking, and start the background consensus verification worker. The worker watches peer-reported block heads, verifies candidate forks, merges fully verified forks, and updates node.latest_block_hash / node.latest_block when a better verified head is available.
node.verify()
node.verify() is idempotent while the verification thread is already running.
To start creating blocks, call node.validate(validation_secret_key). Validation connects the node, prepares validator state, creates a genesis block when no latest block is configured, and starts the consensus validation worker.
from cryptography.hazmat.primitives.asymmetric.ed25519 import Ed25519PrivateKey
validation_secret_key = Ed25519PrivateKey.generate()
node.validate(validation_secret_key)
The validation worker only creates blocks when this node is the scheduled validator for the current head. It applies queued transactions when available, can create empty blocks when the queue is empty, stores the new block atoms locally, advertises them to peers, and updates node.latest_block_hash / node.latest_block.
Transaction Overview
Use send_transaction(...) to atomize, store, advertise, and forward an already-signed transaction to available validators.
from astreum.consensus.transaction import (
create_transaction,
send_transaction,
)
tx = create_transaction(
chain_id=node.config["chain_id"],
amount=100,
counter=sender_account.counter + 1,
recipient=recipient_public_key,
sender=sender_public_key,
)
tx.sign(sender_key)
tx_hash = send_transaction(node, tx)
print(tx_hash.hex())
The node must already be connected and have a latest_block; otherwise the function raises RuntimeError. It writes the transaction's atoms to local storage, advertises them on the P2P network, and sends the transaction hash to peers on the validation route.
Query API
Query functions let you fetch blocks and search transactions by height or attribute from the chain tip.
from astreum import get_block, find_transactions
get_block(node, *, height)
Fetch a single block by its chain height. Returns the Block object or None if the block hasn't been mined yet or isn't reachable.
| Parameter | Type | Description |
|---|---|---|
node |
Node |
An initialised, connected Astreum node. |
height |
int |
The target block height. Must be ≤ the node's latest block. |
block = get_block(node, height=100_000)
if block:
print(f"block hash: {block.expr_id.hex()[:16]}...")
print(f"tx count: {len(block.transactions) if block.transactions else 0}")
Internally walks the previous_block chain to the target era, then binary-descents the bloom tree by offset — approximately 11 storage fetches (O(log N)) plus the chain walk.
find_transactions(node, *, sender, receiver, …)
Search for transactions matching the given filters. All filter parameters are optional — leave a filter at its default (32 zero bytes) to match anything.
When multiple filters are set, only transactions matching all of them are returned (AND semantics).
| Parameter | Type | Default | Description |
|---|---|---|---|
node |
Node |
— | An initialised, connected Astreum node. |
tx_hash |
bytes |
ZERO32 |
Match a specific transaction hash. |
sender |
bytes |
ZERO32 |
Filter by sender public key. |
receiver |
bytes |
ZERO32 |
Filter by recipient public key. |
key |
bytes |
ZERO32 |
Filter by contract bloom key (from bloom.put). |
start_height |
int |
node.latest_block.height |
Search backward from this height. |
end_height |
int |
0 |
Stop when blocks drop below this height. |
limit |
int |
1 |
Max results; pass 0 for no limit. |
# Find up to 10 transactions from a specific sender
txs = find_transactions(node, sender=addr, limit=10)
# Find transactions in a specific height range
txs = find_transactions(
node,
receiver=addr,
start_height=50_000,
end_height=40_000,
limit=5,
)
Each returned Transaction has its block_hash field set to the expr hash of the block that included it, so you can locate the containing block.
Internally uses the bloom tree index to skip eras that can't contain a match, then walks individual blocks inside candidate eras.
Language Syntax
Astreum uses S-expressions with prefix notation. Expressions are either atoms or parenthesised lists. Lists are right-linked Link chains — (a b c) parses as Link(a, Link(b, c)).
Tokens
| Token | Meaning |
|---|---|
( ) |
Delimit a list expression. |
' |
Quote token — when alone, parses as the symbol '. Inside a list it's a regular symbol. |
123 -5 |
Integer literals. Parsed to Expr.Int. |
3.14 -2.5 |
Float literals. Parsed to Expr.Float. |
"hello world" |
String literals. Everything between double quotes is one token (spaces, parens preserved). Parsed to Expr.String. |
0x1f 0Xab |
Hex bytes. Raw hex digits, no two's complement. Parsed to Expr.Bytes. |
add def |
Everything else is a symbol. Parsed to Expr.Symbol. |
; |
Line comment — skips to end of line. |
#; |
Expression skip — skips the next complete expression (including nested lists). |
Expression types
| Type | Class | Description |
|---|---|---|
| Symbol | Expr.Symbol(value: str) |
A named identifier. |
| Bytes | Expr.Bytes(value: bytes) |
Raw byte data. |
| Int | Expr.Int(value: int) |
Arbitrary-precision signed integer. |
| Float | Expr.Float(value: float) |
IEEE 754 double-precision float. |
| String | Expr.String(value: str) |
UTF-8 string content. |
| Link | Expr.Link(head, tail) |
A pair — the building block for lists. Link(None, None) is NIL. |
Links form right-associated chains. (1 2 3) becomes:
Link(Int(1), Link(Int(2), Int(3)))
Environment
Env(data={}, parent=None) is a string-keyed binding store with parent-chain lookup. env.get(key) walks up parent environments. env.put(key, value) writes to the local environment only.
Machine Overview
The machine evaluates an expression tree against an environment, producing a result stack.
from astreum.machine.main import Machine
from astreum.machine import Env, Expr, tokenize, parse
from astreum.node import Node
node = Node()
machine = Machine(node)
# Parse source text and evaluate
tokens = tokenize("(1 2 +)")
expr, _ = parse(tokens)
env = Env()
result = machine.run(expr, env)
# result = Int(3)
machine.run(expr, env) walks the expression tree and pushes results onto a stack. Symbols that match operators pop arguments and push results. Non-operator symbols are looked up in the environment. Bytes, Int, Float, and String values are pushed as-is.
The Machine constructor accepts a mode parameter ("dynamic" or "deterministic", default "dynamic"). In deterministic mode the operators spawn, send, receive, and eval push NIL instead of executing — this ensures reproducible evaluation for contexts such as block validation.
Metering
Every Machine carries a Meter that tracks computation cost in bytes read:
machine = Machine(node, meter_enabled=True, meter_limit=1_000_000)
# MeterExceededError raised if limit is exceeded
machine.meter.used # bytes consumed so far
Operators
Operators are symbols that pop arguments from the stack and push a result.
| Operator | Stack effect | Description |
|---|---|---|
+ |
a b → sum |
Addition. Int/Int → Int, Float/Float → Float, Int+Float → Float (promotion). Overflow on int→float conversion pushes NIL. |
- |
a b → diff |
Subtraction. Same type rules as +. |
* |
a b → prod |
Multiplication. Same type rules as +. |
/ |
a b → quot |
Division. Int/Int → integer division (//), Float/Float → float division. Int division by zero pushes NIL. |
% |
a b → rem |
Modulo (Int only). Pushes NIL on non-Int. |
sqrt |
a → sqrt(a) |
Square root (Float only). Pushes NIL on non-Float or negative. |
& |
a b → a&b |
Bitwise AND (Bytes). |
| |
a b → a|b |
Bitwise OR (Bytes). |
^ |
a b → a^b |
Bitwise XOR (Bytes). |
~ |
a → ~a |
Bitwise NOT (Bytes, one's complement within the operand's byte width). |
<< |
value shifts → result |
Bitwise left shift (Bytes). |
>>> |
value shifts → result |
Logical right shift (Bytes, zero-fill). |
>> |
value shifts → result |
Arithmetic right shift (Bytes, sign-extend). |
rol |
value shifts → result |
Rotate left by shifts bits (Bytes, within the value's bit-width). |
ror |
value shifts → result |
Rotate right by shifts bits (Bytes, within the value's bit-width). |
dip |
v (expr) → ... v |
Temporarily remove v, evaluate (expr) on the remaining stack, then push v back. |
drop |
a → — |
Pop and discard one value. |
dup |
a → a a |
Pop and push the same value twice. |
swap |
a b → b a |
Pop two values and push them back in reversed order. |
link |
head tail → Link(head, tail) |
Construct a Link pair. |
head |
Link(h, t) → h |
Extract the head of a Link; pushes NIL on non-Link. |
tail |
Link(h, t) → t |
Extract the tail of a Link; pushes NIL on non-Link. |
is_atom |
expr → 0|1 |
Pushes Bytes(b"\\x01") if the value is not a Link (i.e. Bytes, Int, Float, String, or Symbol), else Bytes(b"\\x00"). |
is_eq |
a b → 0|1 |
Structural equality: atoms compared by value; Link by recursive head+tail. Different types are never equal. |
eval |
expr → result|nil |
Pop an expression and evaluate it as code in the current environment. In deterministic mode pushes NIL. |
if |
(cond) then else → result |
Evaluate cond quotation; if truthy (non-zero Bytes/Int/Float or non-NIL Link) evaluate then, otherwise evaluate else. |
fn |
argN … arg1 params body → result |
Pops params (a Link chain of Symbols), body, and N args. Binds each arg to its param name in a child environment (parent = call-site env) and evaluates body. |
lambda |
argN … arg1 params body → result |
Same as fn but with parent=None — the body can only access its parameters and built-in operators, not the caller's environment. |
def |
name value → — |
Binds name (a Symbol) to value in the current environment. Write‑once: if the name already exists in the target environment, def is a no‑op (pushes NIL). |
' |
(' X) → X |
Quote special form — wraps a single unevaluated expression. (' 42) pushes Int(42). (' (1 2 3)) pushes the entire list unevaluated as a Link chain. |
quote |
a → (' a) |
Stack operator — pops a value and pushes it back wrapped in a (' …) quotation. |
symbol |
a → symbol|nil |
Convert Bytes (UTF-8 decoded), String, Int, or Float to Expr.Symbol. Pushes NIL on invalid UTF-8 bytes. |
str |
a → string|nil |
Convert any atom (Bytes/Int/Float/Symbol/String) to Expr.String. |
float |
a → float|nil |
Convert Int, Bytes (exactly 8 bytes, IEEE 754 little-endian), String, or Symbol to Expr.Float. |
int |
a → int|nil |
Convert Bytes (little-endian signed), String, Symbol, or Float to Expr.Int. |
bytes |
a → bytes|nil |
Convert Int (variable-length signed), Float (8-byte IEEE 754 little-endian), String, or Symbol (UTF-8) to Expr.Bytes. |
ref |
hash → expr|nil |
Resolve a 32‑byte hash to its stored expression via content‑addressable lookup (node.get_expr). For Link values, returns a thunk‑wrapped pair (' head_hash ' tail_hash) for lazy traversal of subtrees. Pushes NIL on miss or invalid hash. |
load |
hash → full_expr|nil |
Deep‑resolve a 32‑byte hash recursively through the entire sub‑tree (node.get_expr_full). Cost is 2× the resolved expression size. Pushes NIL on any unresolved child. |
Actor Model
The machine supports concurrent actors communicating via named mailboxes.
| Operator | Stack effect | Description |
|---|---|---|
spawn |
body name → name|nil |
Spawn a new actor thread running body in a child environment. name must be a Symbol. Returns name on success, NIL if the name is already taken or threading is disabled. |
send |
target msg → — |
Send msg to the mailbox of actor target. Returns nothing. Drops silently if the mailbox doesn't exist. |
receive |
target → msg|nil |
Block until a message arrives in the mailbox of actor target. Returns NIL if the mailbox doesn't exist. |
Actors run on daemon threads with their own environment (parented to the spawner's environment). The Machine constructor accepts a mode parameter ("dynamic" or "deterministic", default "dynamic"). In deterministic mode spawn, send, receive, and eval push NIL — they either require concurrency or runtime code-as-data, both of which break determinism.
The def operator is write-once per environment: if a name already exists in the target environment's own bindings, def is a no-op (pushes NIL). This prevents accidental overwrites and simplifies future ZK-proof generation.
Quickstart Example
from astreum.machine.main import Machine
from astreum.machine import Env, Expr, tokenize, parse, Meter
from astreum.node import Node
node = Node()
machine = Machine(node)
# Call an fn inline: (3 5 (quote ($0 $1)) (quote ($0 $1 +)) fn)
# Then add 2 to the result
src = "((3 5 (quote ($0 $1)) (quote ($0 $1 +)) fn) 2 +)"
tokens = tokenize(src)
expr, _ = parse(tokens)
env = Env()
stack = machine.run(expr, env)
# First value on stack is the result
result = stack[0]
print(result.value) # 10
Handling errors
tokenize and parse raise ParseError (from astreum.machine.parser) on malformed input:
from astreum.machine import tokenize, parse, ParseError
try:
tokens = tokenize("(1 2")
expr, _ = parse(tokens)
except ParseError as e:
print("Parse failed:", e)
Logging
Every Node instance wires up structured logging automatically:
- Set
config["logging_enabled"] = Falseto skip logging setup entirely. This bypasses log directory creation, file rotation, console mirroring, and the background listener thread. - Logs land in per-instance files named
node.csvunder%LOCALAPPDATA%\Astreum\lib-py\logs/<instance_id>on Windows and$XDG_STATE_HOME(or~/.local/state)/Astreum/lib-py/logs/<instance_id>on other platforms. The<instance_id>is the first 16 hex characters of a BLAKE3 hash of the caller's file path, so running the node from different entry points keeps their logs isolated. - Files rotate at midnight UTC with gzip compression (
node-YYYY-MM-DD.csv.gz) and retain 7 days by default. Override viaconfig["logging_retention_days"]. - Each event is a single CSV row with columns
ts,level,msg,module, andfunc. - Set
config["verbose"] = Trueto mirror logs to stdout in a human-friendly format like[2025-04-13-42-59] [info] Starting Astreum Node. - The very first entry emitted is the banner
Starting Astreum Node, signalling that the logging pipeline is live before other subsystems spin up.
Testing
python3 -m venv venv
source venv/bin/activate
pip install -e .
for all tests
python3 -m unittest discover -s tests
for individual tests
|| Test | Method | Pass |
|| --- | --- | --- |
|| pytest tests/node/test_current_validator.py | python3 -m unittest tests.node.test_current_validator | ✅ |
|| pytest tests/node/test_node_connection.py | python3 -m unittest tests.node.test_node_connection | ✅ |
|| pytest tests/node/test_node_init.py | python3 -m unittest tests.node.test_node_init | ✅ |
|| pytest tests/node/test_node_validation.py | python3 -m unittest tests.node.test_node_validation | |
|| pytest tests/node/eval.py | — | ✅ |
|| pytest tests/node/machine/eval.py | python3 -m unittest tests.node.machine.eval | ✅ |
|| pytest tests/node/machine/parser.py | python3 -m unittest tests.node.machine.parser | ✅ |
|| pytest tests/node/machine/integer.py | python3 -m unittest tests.node.machine.integer | ✅ |
|| pytest tests/node/machine/stack.py | python3 -m unittest tests.node.machine.stack | ✅ |
|| pytest tests/node/machine/shifts.py | python3 -m unittest tests.node.machine.shifts | ✅ |
|| pytest tests/block/expr.py | — | ✅ |
|| pytest tests/block/nonce.py | — | ✅ |
|| pytest tests/communication/test_message_port.py | python3 -m unittest tests.communication.test_message_port | ✅ |
|| pytest tests/communication/test_integration_port_handling.py | — | ✅ |
|| pytest tests/consensus/genesis.py | python3 -m unittest tests.consensus.genesis | ✅ |
|| pytest tests/consensus/test_treasury_record.py | — | ✅ |
|| pytest tests/consensus/transaction/test_apply.py | — | ✅ |
|| pytest tests/storage/indexing.py | python3 -m unittest tests.storage.indexing | ✅ |
|| pytest tests/storage/cold.py | python3 -m unittest tests.storage.cold | ✅ |
|| pytest tests/models/test_patricia.py | python3 -m unittest tests.models.test_patricia | ✅ |
|| pytest tests/crypto/bloom_filter.py | python3 -m unittest tests.crypto.bloom_filter | ✅ |
|| pytest tests/crypto/bloom_tree.py | python3 -m unittest tests.crypto.bloom_tree | ✅ |
|| pytest tests/utils/test_logging.py | — | ✅ |
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file astreum-0.4.43.tar.gz.
File metadata
- Download URL: astreum-0.4.43.tar.gz
- Upload date:
- Size: 134.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
986b918b85f72ec48927d188b7a5a8cee3ccef141f709028a70ed302b0056b7d
|
|
| MD5 |
3417b178dd19fdf03a42ca25b5895d4e
|
|
| BLAKE2b-256 |
101b5d8497dd8a4ca1dce3da054968aa095ed3c9e666dd3d46f74155dec6b274
|
File details
Details for the file astreum-0.4.43-py3-none-any.whl.
File metadata
- Download URL: astreum-0.4.43-py3-none-any.whl
- Upload date:
- Size: 195.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
84bf300cb51597321b72de358019762a6815c37e5722518252a9dcd8798ab6f0
|
|
| MD5 |
abecd2fc20ef9f1e3e89b64a3d779917
|
|
| BLAKE2b-256 |
cb772600ce6c2215ce8960725ea97d3df97b9af596f0e234e8bc9fe98242a096
|