Bump opentelemetry to latest version

Author: sfc-gh-bnandibhatla

NOTICE.txt

@@ -6,7 +6,7 @@ The following notices are required by licensors of software used in the Snowflak
 --------------------------------------------------------------------------------
 
 This library includes software which is copied from or derived from the OpenTelemetry Python API and SDK.
-OpenTelemetry Python v1.26.0
+OpenTelemetry Python v1.33.1
 https://github.com/open-telemetry/opentelemetry-python
 
                                  Apache License

anaconda/meta.yaml

@@ -14,8 +14,8 @@ requirements:
         - setuptools >=40.0.0
     run:
         - python
-        - opentelemetry-api ==1.26.0
-        - opentelemetry-sdk ==1.26.0
+        - opentelemetry-api ==1.33.1
+        - opentelemetry-sdk ==1.33.1
 
 about:
     home: https://www.snowflake.com/

scripts/vendor_otlp_proto_common.sh

@@ -9,7 +9,7 @@
 # fixes needed in the OTLP exporter.
 
 # Pinned commit/branch/tag for the current version used in opentelemetry-proto python package.
-REPO_BRANCH_OR_COMMIT="v1.26.0"
+REPO_BRANCH_OR_COMMIT="v1.33.1"
 
 set -e
 

setup.py

@@ -27,8 +27,8 @@
     description=DESCRIPTION,
     long_description=LONG_DESCRIPTION,
     install_requires=[
-        "opentelemetry-api == 1.26.0",
-        "opentelemetry-sdk == 1.26.0",
+        "opentelemetry-api == 1.33.1",
+        "opentelemetry-sdk == 1.33.1",
     ],
     packages=find_namespace_packages(
         where='src'

src/snowflake/telemetry/_internal/exporter/otlp/proto/logs/__init__.py

@@ -84,7 +84,13 @@ class SnowflakeLoggingHandler(_logs.LoggingHandler):
     discarded by the original implementation.
     """
 
-    LOGGER_NAME_TEMP_ATTRIBUTE = "__snow.logging.temp.logger_name"
+    LOGGER_NAME_TEMP_ATTRIBUTE: typing.Final = "__snow.logging.temp.logger_name"
+    CODE_FILEPATH: typing.Final = "code.filepath"
+    CODE_FILE_PATH: typing.Final = "code.file.path"
+    CODE_FUNCTION: typing.Final = "code.function"
+    CODE_FUNCTION_NAME: typing.Final = "code.function.name"
+    CODE_LINENO: typing.Final = "code.lineno"
+    CODE_LINE_NUMBER: typing.Final = "code.line.number"
 
     def __init__(
             self,
@@ -101,6 +107,17 @@ def __init__(
     def _get_attributes(record: logging.LogRecord) -> types.Attributes:
         attributes = _logs.LoggingHandler._get_attributes(record) # pylint: disable=protected-access
 
+        # Preserving old naming conventions for code attributes that were changed as part of
+        # https://github.com/open-telemetry/opentelemetry-python/commit/1b1e8d80c764ad3aa76abfb56a7002ddea11fdb5 in
+        # order to avoid a behavior change for Snowflake customers.
+        if SnowflakeLoggingHandler.CODE_FILE_PATH in attributes:
+            attributes[SnowflakeLoggingHandler.CODE_FILEPATH] = attributes.pop(SnowflakeLoggingHandler.CODE_FILE_PATH)
+        if SnowflakeLoggingHandler.CODE_FUNCTION_NAME in attributes:
+            attributes[SnowflakeLoggingHandler.CODE_FUNCTION] = attributes.pop(
+                SnowflakeLoggingHandler.CODE_FUNCTION_NAME)
+        if SnowflakeLoggingHandler.CODE_LINE_NUMBER in attributes:
+            attributes[SnowflakeLoggingHandler.CODE_LINENO] = attributes.pop(SnowflakeLoggingHandler.CODE_LINE_NUMBER)
+
         # Temporarily storing logger's name in record's attributes.
         # This attribute will be removed by the logger.
         #

src/snowflake/telemetry/_internal/opentelemetry/exporter/otlp/proto/common/__init__.py

@@ -14,7 +14,7 @@
 #
 # This file has been modified from the original source code at
 #
-#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.26.0
+#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.33.1
 #
 # by Snowflake Inc.
 

src/snowflake/telemetry/_internal/opentelemetry/exporter/otlp/proto/common/_internal/__init__.py

@@ -14,42 +14,44 @@
 #
 # This file has been modified from the original source code at
 #
-#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.26.0
+#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.33.1
 #
 # by Snowflake Inc.
 
 
+from __future__ import annotations
+
 import logging
 from collections.abc import Sequence
 from itertools import count
 from typing import (
     Any,
-    Mapping,
-    Optional,
-    List,
     Callable,
-    TypeVar,
     Dict,
     Iterator,
+    List,
+    Mapping,
+    Optional,
+    TypeVar,
 )
 
-from opentelemetry.sdk.util.instrumentation import InstrumentationScope
+from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import AnyValue as PB2AnyValue
 from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import (
-    InstrumentationScope as PB2InstrumentationScope,
+    ArrayValue as PB2ArrayValue,
 )
-from snowflake.telemetry._internal.opentelemetry.proto.resource.v1.resource_marshaler import (
-    Resource as PB2Resource,
+from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import (
+    InstrumentationScope as PB2InstrumentationScope,
 )
-from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import AnyValue as PB2AnyValue
 from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import KeyValue as PB2KeyValue
 from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import (
     KeyValueList as PB2KeyValueList,
 )
-from snowflake.telemetry._internal.opentelemetry.proto.common.v1.common_marshaler import (
-    ArrayValue as PB2ArrayValue,
+from snowflake.telemetry._internal.opentelemetry.proto.resource.v1.resource_marshaler import (
+    Resource as PB2Resource,
 )
 from opentelemetry.sdk.trace import Resource
-from opentelemetry.util.types import Attributes
+from opentelemetry.sdk.util.instrumentation import InstrumentationScope
+from opentelemetry.util.types import _ExtendedAttributes
 
 _logger = logging.getLogger(__name__)
 
@@ -65,14 +67,19 @@ def _encode_instrumentation_scope(
     return PB2InstrumentationScope(
         name=instrumentation_scope.name,
         version=instrumentation_scope.version,
+        attributes=_encode_attributes(instrumentation_scope.attributes),
     )
 
 
 def _encode_resource(resource: Resource) -> PB2Resource:
     return PB2Resource(attributes=_encode_attributes(resource.attributes))
 
 
-def _encode_value(value: Any) -> PB2AnyValue:
+def _encode_value(
+    value: Any, allow_null: bool = False
+) -> Optional[PB2AnyValue]:
+    if allow_null is True and value is None:
+        return None
     if isinstance(value, bool):
         return PB2AnyValue(bool_value=value)
     if isinstance(value, str):
@@ -81,21 +88,49 @@ def _encode_value(value: Any) -> PB2AnyValue:
         return PB2AnyValue(int_value=value)
     if isinstance(value, float):
         return PB2AnyValue(double_value=value)
+    if isinstance(value, bytes):
+        return PB2AnyValue(bytes_value=value)
     if isinstance(value, Sequence):
         return PB2AnyValue(
-            array_value=PB2ArrayValue(values=[_encode_value(v) for v in value])
+            array_value=PB2ArrayValue(
+                values=_encode_array(value, allow_null=allow_null)
+            )
         )
     elif isinstance(value, Mapping):
         return PB2AnyValue(
             kvlist_value=PB2KeyValueList(
-                values=[_encode_key_value(str(k), v) for k, v in value.items()]
+                values=[
+                    _encode_key_value(str(k), v, allow_null=allow_null)
+                    for k, v in value.items()
+                ]
             )
         )
     raise Exception(f"Invalid type {type(value)} of value {value}")
 
 
-def _encode_key_value(key: str, value: Any) -> PB2KeyValue:
-    return PB2KeyValue(key=key, value=_encode_value(value))
+def _encode_key_value(
+    key: str, value: Any, allow_null: bool = False
+) -> PB2KeyValue:
+    return PB2KeyValue(
+        key=key, value=_encode_value(value, allow_null=allow_null)
+    )
+
+
+def _encode_array(
+    array: Sequence[Any], allow_null: bool = False
+) -> Sequence[PB2AnyValue]:
+    if not allow_null:
+        # Let the exception get raised by _encode_value()
+        return [_encode_value(v, allow_null=allow_null) for v in array]
+
+    return [
+        _encode_value(v, allow_null=allow_null)
+        if v is not None
+        # Use an empty AnyValue to represent None in an array. Behavior may change pending
+        # https://github.com/open-telemetry/opentelemetry-specification/issues/4392
+        else PB2AnyValue()
+        for v in array
+    ]
 
 
 def _encode_span_id(span_id: int) -> bytes:
@@ -107,14 +142,17 @@ def _encode_trace_id(trace_id: int) -> bytes:
 
 
 def _encode_attributes(
-    attributes: Attributes,
+    attributes: _ExtendedAttributes,
+    allow_null: bool = False,
 ) -> Optional[List[PB2KeyValue]]:
     if attributes:
         pb2_attributes = []
         for key, value in attributes.items():
             # pylint: disable=broad-exception-caught
             try:
-                pb2_attributes.append(_encode_key_value(key, value))
+                pb2_attributes.append(
+                    _encode_key_value(key, value, allow_null=allow_null)
+                )
             except Exception as error:
                 _logger.exception("Failed to encode key %s: %s", key, error)
     else:

src/snowflake/telemetry/_internal/opentelemetry/exporter/otlp/proto/common/_internal/_log_encoder/__init__.py

@@ -14,29 +14,28 @@
 #
 # This file has been modified from the original source code at
 #
-#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.26.0
+#     https://github.com/open-telemetry/opentelemetry-python/tree/v1.33.1
 #
 # by Snowflake Inc.
 from collections import defaultdict
-from typing import Sequence, List
+from typing import List, Sequence
 
 from snowflake.telemetry._internal.opentelemetry.exporter.otlp.proto.common._internal import (
+    _encode_attributes,
     _encode_instrumentation_scope,
     _encode_resource,
     _encode_span_id,
     _encode_trace_id,
     _encode_value,
-    _encode_attributes,
 )
 from snowflake.telemetry._internal.opentelemetry.proto.collector.logs.v1.logs_service_marshaler import (
     ExportLogsServiceRequest,
 )
+from snowflake.telemetry._internal.opentelemetry.proto.logs.v1.logs_marshaler import LogRecord as PB2LogRecord
 from snowflake.telemetry._internal.opentelemetry.proto.logs.v1.logs_marshaler import (
-    ScopeLogs,
     ResourceLogs,
+    ScopeLogs,
 )
-from snowflake.telemetry._internal.opentelemetry.proto.logs.v1.logs_marshaler import LogRecord as PB2LogRecord
-
 from opentelemetry.sdk._logs import LogData
 
 
@@ -55,15 +54,18 @@ def _encode_log(log_data: LogData) -> PB2LogRecord:
         if log_data.log_record.trace_id == 0
         else _encode_trace_id(log_data.log_record.trace_id)
     )
+    body = log_data.log_record.body
     return PB2LogRecord(
         time_unix_nano=log_data.log_record.timestamp,
         observed_time_unix_nano=log_data.log_record.observed_timestamp,
         span_id=span_id,
         trace_id=trace_id,
         flags=int(log_data.log_record.trace_flags),
-        body=_encode_value(log_data.log_record.body),
+        body=_encode_value(body, allow_null=True),
         severity_text=log_data.log_record.severity_text,
-        attributes=_encode_attributes(log_data.log_record.attributes),
+        attributes=_encode_attributes(
+            log_data.log_record.attributes, allow_null=True
+        ),
         dropped_attributes_count=log_data.log_record.dropped_attributes,
         severity_number=log_data.log_record.severity_number.value,
     )
@@ -88,6 +90,9 @@ def _encode_resource_logs(batch: Sequence[LogData]) -> List[ResourceLogs]:
                 ScopeLogs(
                     scope=(_encode_instrumentation_scope(sdk_instrumentation)),
                     log_records=pb2_logs,
+                    schema_url=sdk_instrumentation.schema_url
+                    if sdk_instrumentation
+                    else None,
                 )
             )
         pb2_resource_logs.append(