ASnapshot represents a read-onlytransaction: when multiple read operations are performed via a Snapshot,the results are consistent as of a particular point in time.

Beginning a Snapshot

To begin using a snapshot using the default “bound” (which is “strong”),meaning all reads are performed at a timestamp where all previously-committedtransactions are visible:

with database.snapshot() as snapshot:    ...

You can also specify a weaker bound, which can either be to perform allreads as of a given timestamp:

import datetimeTIMESTAMP = datetime.datetime.utcnow().replace(tzinfo=datetime.timezone.utc)with database.snapshot(read_timestamp=TIMESTAMP) as snapshot:    ...

or as of a given duration in the past:

import datetimeDURATION = datetime.timedelta(seconds=5)with database.snapshot(exact_staleness=DURATION) as snapshot:    ...

Single Use and Multiple Use Snapshots

In the context of read only transactions,read andexecute_sqlmethods can be used multiple times if you specifymulti_use=Truein the constructor of the snapshot. However,multi_use=True isincompatible with eithermax_staleness and/ormin_read_timestamp.

Otherwisemulti_use defaults toFalse and the snapshot cannot bereused.

with database.snapshot(multi_use=True) as snapshot:    ...

begin() can only be used on asnapshot withmulti_use=True. In which case it is also necessaryto call if you need to have multiple pending operations.

Read Table Data

To read data for selected rows from a table in the database, callread() which will returnall rows specified inkeyset, or fail if the result set is too large,

with database.snapshot() as snapshot:    result = snapshot.read(        table='table-name', columns=['first_name', 'last_name', 'age'],        keyset=spanner.KeySet([['phred@example.com'], ['bharney@example.com']]))    for row in result:        print(row)

NOTE: Perform all iterations within the context of thewith database.snapshot()block.

Execute a SQL Select Statement

To read data from tables in the database using a query, callexecute_sql()which will return all rows matching the query, or fail if theresult set is too large,

with database.snapshot() as snapshot:    QUERY = (        'SELECT e.first_name, e.last_name, p.telephone '        'FROM employees as e, phones as p '        'WHERE p.employee_id == e.employee_id')    result = snapshot.execute_sql(QUERY)    for row in result:        print(row)

NOTE: Perform all iteration within the context of thewith database.snapshot()block.

Next Step

Next, learn aboutRead-write Transactions.