Determine this is the right repository
Summary of the issue
When google.api_core.exceptions.Aborted is raised inside run_in_transaction with an empty errors attribute (e.g., manually instantiated Aborted("...")), run_in_transaction fails with an uncaught IndexError: list index out of range because it directly accesses exc.errors[0] without checking if exc.errors is empty.
While instantiating Aborted manually without errors may not be the standard path, the SDK should defensively handle empty errors (e.g., falling back to default_retry_delay) instead of crashing with an IndexError.
API client name and version
google-cloud-spanner==3.69.0
Reproduction steps: code
from google.api_core.exceptions import Aborted
from google.cloud import spanner
spanner_client = spanner.Client()
instance = spanner_client.instance("your-instance-id")
database = instance.database("your-database-id")
def callback(transaction):
# Raising Aborted without providing the errors payload
raise Aborted("VersionMismatch: CAS failed")
database.run_in_transaction(callback)
Reproduction steps: supporting files
No response
Reproduction steps: actual results
IndexError: list index out of range
(Occurs inside delay_seconds = _get_retry_delay(exc.errors[0], ...))
Reproduction steps: expected results
The SDK handles empty exc.errors gracefully (e.g., uses default_retry_delay) and executes the transaction retry loop as intended without raising an IndexError.
OS & version + platform
Windows 11
Python environment
Python 3.14.6
Python dependencies
Package Version
certifi 2026.7.22
cffi 2.1.0
charset-normalizer 3.4.9
cryptography 49.0.0
google-api-core 2.33.0
google-auth 2.56.2
google-cloud-core 2.6.0
google-cloud-monitoring 2.31.0
google-cloud-spanner 3.69.0
google-cloud-storage 3.13.0
google-crc32c 1.8.0
google-resumable-media 2.10.0
googleapis-common-protos 1.75.0
grpc-google-iam-v1 0.14.4
grpc-interceptor 0.15.4
grpcio 1.82.1
grpcio-status 1.82.1
idna 3.18
mmh3 5.2.1
opentelemetry-api 1.44.0
opentelemetry-resourcedetector-gcp 1.12.0a0
opentelemetry-sdk 1.44.0
opentelemetry-semantic-conventions 0.65b0
pip 26.1.2
proto-plus 1.28.2
protobuf 7.35.1
pyasn1 0.6.4
pyasn1_modules 0.4.2
pycparser 3.0
requests 2.34.2
sqlparse 0.5.5
typing_extensions 4.16.0
urllib3 2.7.0
Additional context
Defensive check like if exc.errors: before accessing exc.errors[0] would prevent IndexError and make error handling in run_in_transaction much more robust.
Determine this is the right repository
Summary of the issue
When
google.api_core.exceptions.Abortedis raised insiderun_in_transactionwith an emptyerrorsattribute (e.g., manually instantiatedAborted("...")),run_in_transactionfails with an uncaughtIndexError: list index out of rangebecause it directly accessesexc.errors[0]without checking ifexc.errorsis empty.While instantiating
Abortedmanually withouterrorsmay not be the standard path, the SDK should defensively handle emptyerrors(e.g., falling back todefault_retry_delay) instead of crashing with anIndexError.API client name and version
google-cloud-spanner==3.69.0
Reproduction steps: code
from google.api_core.exceptions import Aborted
from google.cloud import spanner
spanner_client = spanner.Client()
instance = spanner_client.instance("your-instance-id")
database = instance.database("your-database-id")
def callback(transaction):
# Raising Aborted without providing the
errorspayloadraise Aborted("VersionMismatch: CAS failed")
database.run_in_transaction(callback)
Reproduction steps: supporting files
No response
Reproduction steps: actual results
IndexError: list index out of range
(Occurs inside
delay_seconds = _get_retry_delay(exc.errors[0], ...))Reproduction steps: expected results
The SDK handles empty
exc.errorsgracefully (e.g., usesdefault_retry_delay) and executes the transaction retry loop as intended without raising anIndexError.OS & version + platform
Windows 11
Python environment
Python 3.14.6
Python dependencies
Package Version
certifi 2026.7.22
cffi 2.1.0
charset-normalizer 3.4.9
cryptography 49.0.0
google-api-core 2.33.0
google-auth 2.56.2
google-cloud-core 2.6.0
google-cloud-monitoring 2.31.0
google-cloud-spanner 3.69.0
google-cloud-storage 3.13.0
google-crc32c 1.8.0
google-resumable-media 2.10.0
googleapis-common-protos 1.75.0
grpc-google-iam-v1 0.14.4
grpc-interceptor 0.15.4
grpcio 1.82.1
grpcio-status 1.82.1
idna 3.18
mmh3 5.2.1
opentelemetry-api 1.44.0
opentelemetry-resourcedetector-gcp 1.12.0a0
opentelemetry-sdk 1.44.0
opentelemetry-semantic-conventions 0.65b0
pip 26.1.2
proto-plus 1.28.2
protobuf 7.35.1
pyasn1 0.6.4
pyasn1_modules 0.4.2
pycparser 3.0
requests 2.34.2
sqlparse 0.5.5
typing_extensions 4.16.0
urllib3 2.7.0
Additional context
Defensive check like
if exc.errors:before accessingexc.errors[0]would preventIndexErrorand make error handling inrun_in_transactionmuch more robust.