Transaction Configuration Tests¶

View source code

These tests cover WAL flush modes and their database-wide scope, read-your-writes, auto-transaction control, and combinations of these settings (plus error handling on a closed database).

Key Config Options¶

WAL flush: set_wal_flush("no" | "yes_nometadata" | "yes_full")
Read-your-writes: set_read_your_writes(bool)
Auto-transaction: set_auto_transaction(bool)

test_set_wal_flush_modes¶

Calls set_wal_flush() with each valid mode. The test has no assertion: it passes when none of the calls raises, and it does not read the mode back.

What it calls:

set_wal_flush("no") (default / maximum performance)
set_wal_flush("yes_nometadata")
set_wal_flush("yes_full") (maximum durability)
Resetting back to "no"

Pattern:

temp_db.set_wal_flush("no")
temp_db.set_wal_flush("yes_nometadata")
temp_db.set_wal_flush("yes_full")
temp_db.set_wal_flush("no")

test_set_wal_flush_is_database_wide¶

Pins that set_wal_flush() reaches the transactions of every thread on the database: a thread that never called it commits with the mode another thread set. If the scope ever narrows to the calling thread, the test fails and the durability section of the transactions guide must change with it. The process default is read from a fresh thread first rather than assumed to be NO, because a production-mode server anywhere in the process raises it to 1 for every database opened afterwards (test_server.py starts one, and CI runs it first).

Pattern:

default = flush_of_a_new_thread()          # the process default, NO unless something raised it
temp_db.set_wal_flush("yes_full")           # or "no" when the default is already YES_FULL
# this thread's transactions and a new thread's both carry YES_FULL

test_set_wal_flush_invalid_mode¶

Tests that invalid WAL flush modes raise ValueError.

What it tests:

An unknown mode ("invalid_mode") raises ValueError matching "Invalid WAL flush mode"
An uppercase mode ("YES_FULL") is rejected; the mode must be lowercase

Pattern:

with pytest.raises(ValueError, match="Invalid WAL flush mode"):
    temp_db.set_wal_flush("invalid_mode")

with pytest.raises(ValueError, match="Invalid WAL flush mode"):
    temp_db.set_wal_flush("YES_FULL")  # Must be lowercase

test_set_read_your_writes¶

Toggles read-your-writes. The test has no assertion: it passes when none of the calls raises, and it does not read the setting back.

What it calls:

Enabling read-your-writes (True, the default)
Disabling it (False, for better concurrency)
Re-enabling it

Pattern:

temp_db.set_read_your_writes(True)
temp_db.set_read_your_writes(False)
temp_db.set_read_your_writes(True)

test_set_auto_transaction¶

Toggles auto-transaction. The test has no assertion: it passes when none of the calls raises, and it does not read the setting back.

What it calls:

Enabling auto-transaction (True, the default)
Disabling it (False, for manual control)
Re-enabling it

Pattern:

temp_db.set_auto_transaction(True)
temp_db.set_auto_transaction(False)
temp_db.set_auto_transaction(True)

test_transaction_config_with_operations¶

Tests config methods alongside normal operations.

What it tests:

Configuring durability (set_wal_flush("yes_full"), set_read_your_writes(True)) then writing inside a transaction
Switching to performance mode (set_wal_flush("no"), set_read_your_writes(False)) and writing more
Verifying counts via SELECT count(*) queries (2, then 3)

Pattern:

temp_db.command("sql", "CREATE VERTEX TYPE ConfigTest")
temp_db.set_wal_flush("yes_full")
temp_db.set_read_your_writes(True)

with temp_db.transaction():
    temp_db.command("sql", "CREATE VERTEX ConfigTest SET name = 'test1'")
    temp_db.command("sql", "CREATE VERTEX ConfigTest SET name = 'test2'")

count = next(temp_db.query("sql", "SELECT count(*) as cnt FROM ConfigTest")).get("cnt")
assert count == 2

test_manual_transaction_mode¶

Tests manual transaction control with auto-transaction disabled.

What it tests:

With set_auto_transaction(False), writes inside an explicit db.transaction() context are visible to a later query (count == 2); the test does not try a write outside a transaction
Auto-transaction is re-enabled in a finally block

Pattern:

temp_db.command("sql", "CREATE VERTEX TYPE ManualTest")
temp_db.set_auto_transaction(False)
try:
    with temp_db.transaction():
        temp_db.command("sql", "CREATE VERTEX ManualTest SET name = 'manual1'")
        temp_db.command("sql", "CREATE VERTEX ManualTest SET name = 'manual2'")
finally:
    temp_db.set_auto_transaction(True)

test_wal_flush_with_bulk_operations¶

Tests WAL flush modes with chunked bulk inserts.

What it tests:

Bulk inserting with set_wal_flush("yes_full") in 200-record chunks up to 500
Switching to set_wal_flush("no") and continuing up to 1000
Verifying counts (500, then 1000)

Pattern:

temp_db.command("sql", "CREATE VERTEX TYPE BatchTest")
temp_db.set_wal_flush("yes_full")

chunk_size = 200
total = 500
for start in range(0, total, chunk_size):
    with temp_db.transaction():
        for i in range(start, min(start + chunk_size, total)):
            temp_db.command(
                "sql", "CREATE VERTEX BatchTest SET value = :value", {"value": i}
            )

test_config_methods_on_closed_database¶

Tests that config methods raise on a closed database.

What it tests:

set_wal_flush(...), set_read_your_writes(...), and set_auto_transaction(...) each raise ArcadeDBError matching "closed" after db.close()

Pattern:

db = create_database(temp_db_path)
db.close()

with pytest.raises(ArcadeDBError, match="closed"):
    db.set_wal_flush("no")

test_combined_config_changes¶

Tests changing multiple config settings together.

What it tests:

Starting with durability settings (yes_full, read-your-writes on, auto-transaction on)
Switching to performance settings (no, read-your-writes off) and writing more
Verifying the final count (2) and restoring durability settings

Pattern:

temp_db.set_wal_flush("yes_full")
temp_db.set_read_your_writes(True)
temp_db.set_auto_transaction(True)
temp_db.command("sql", "CREATE VERTEX TYPE Combined")

with temp_db.transaction():
    temp_db.command("sql", "CREATE VERTEX Combined SET value = 1")

temp_db.set_wal_flush("no")
temp_db.set_read_your_writes(False)

with temp_db.transaction():
    temp_db.command("sql", "CREATE VERTEX Combined SET value = 2")

Common Assertions¶

# Invalid WAL mode rejected
with pytest.raises(ValueError, match="Invalid WAL flush mode"):
    db.set_wal_flush("invalid_mode")

# Config methods raise on a closed database
with pytest.raises(ArcadeDBError, match="closed"):
    db.set_wal_flush("no")

# Records written under any flush mode are visible to a later query
count = next(db.query("sql", "SELECT count(*) as cnt FROM ConfigTest")).get("cnt")
assert count == 2

Configuration Methods¶

Method	Argument	Notes
`set_wal_flush(mode)`	`"no"`, `"yes_nometadata"`, `"yes_full"`	Lowercase only; invalid modes raise `ValueError`
`set_read_your_writes(flag)`	`bool`	Default `True`
`set_auto_transaction(flag)`	`bool`	Default `True`; when `False`, use explicit `db.transaction()`

Key Takeaways¶

WAL flush modes - "yes_full" for durability, "no" for performance
Lowercase modes only - Uppercase or unknown modes raise ValueError
Read-your-writes - Toggle for concurrency vs immediate visibility
Manual transactions - Disable auto-transaction and use db.transaction() contexts
Closed database - Config methods raise ArcadeDBError once closed

Transaction Configuration Tests¶

Key Config Options¶

test_set_wal_flush_modes¶

test_set_wal_flush_is_database_wide¶

test_set_wal_flush_invalid_mode¶

test_set_read_your_writes¶

test_set_auto_transaction¶

test_transaction_config_with_operations¶

test_manual_transaction_mode¶

test_wal_flush_with_bulk_operations¶

test_config_methods_on_closed_database¶

test_combined_config_changes¶

Common Assertions¶

Configuration Methods¶

Key Takeaways¶

See Also¶