Decode URL-encoded headers in environment vars (open-telemetry#2312)

mattoberle · web-flow · commit d8bebdd5f7c3 · 2022-01-04T13:47:26.000-05:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -6,6 +6,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased](https://github.com/open-telemetry/opentelemetry-python/compare/v1.8.0-0.27b0...HEAD)
 
+- Decode URL-encoded headers in environment variables
+  ([#2312](https://github.com/open-telemetry/opentelemetry-python/pull/2312))
+
 ## [1.8.0-0.27b0](https://github.com/open-telemetry/opentelemetry-python/releases/tag/v1.8.0-0.27b0) - 2021-12-17
 
 - Adds Aggregation and instruments as part of Metrics SDK
diff --git a/exporter/opentelemetry-exporter-otlp-proto-grpc/src/opentelemetry/exporter/otlp/proto/grpc/exporter.py b/exporter/opentelemetry-exporter-otlp-proto-grpc/src/opentelemetry/exporter/otlp/proto/grpc/exporter.py
@@ -19,7 +19,7 @@
 from collections.abc import Sequence
 from os import environ
 from time import sleep
-from typing import Any, Callable, Dict, Generic, List, Optional
+from typing import Any, Callable, Dict, Generic, List, Optional, Tuple, Union
 from typing import Sequence as TypingSequence
 from typing import TypeVar
 from urllib.parse import urlparse
@@ -204,7 +204,9 @@ def __init__(
         endpoint: Optional[str] = None,
         insecure: Optional[bool] = None,
         credentials: Optional[ChannelCredentials] = None,
-        headers: Optional[Sequence] = None,
+        headers: Optional[
+            Union[TypingSequence[Tuple[str, str]], Dict[str, str], str]
+        ] = None,
         timeout: Optional[int] = None,
         compression: Optional[Compression] = None,
     ):
@@ -229,6 +231,8 @@ def __init__(
         if isinstance(self._headers, str):
             temp_headers = parse_headers(self._headers)
             self._headers = tuple(temp_headers.items())
+        elif isinstance(self._headers, dict):
+            self._headers = tuple(self._headers.items())
 
         self._timeout = timeout or int(
             environ.get(OTEL_EXPORTER_OTLP_TIMEOUT, 10)
diff --git a/exporter/opentelemetry-exporter-otlp-proto-grpc/tests/test_otlp_trace_exporter.py b/exporter/opentelemetry-exporter-otlp-proto-grpc/tests/test_otlp_trace_exporter.py
@@ -281,6 +281,13 @@ def test_otlp_headers_from_env(self, mock_ssl_channel, mock_secure):
         self.assertEqual(
             exporter._headers, (("key3", "value3"), ("key4", "value4"))
         )
+        exporter = OTLPSpanExporter(
+            headers={"key5": "value5", "key6": "value6"}
+        )
+        # pylint: disable=protected-access
+        self.assertEqual(
+            exporter._headers, (("key5", "value5"), ("key6", "value6"))
+        )
 
     # pylint: disable=no-self-use
     @patch("opentelemetry.exporter.otlp.proto.grpc.exporter.insecure_channel")
diff --git a/opentelemetry-api/src/opentelemetry/util/re.py b/opentelemetry-api/src/opentelemetry/util/re.py
@@ -15,6 +15,7 @@
 import logging
 from re import compile, split
 from typing import Mapping
+from urllib.parse import unquote
 
 _logger = logging.getLogger(__name__)
 
@@ -51,8 +52,8 @@ def parse_headers(s: str) -> Mapping[str, str]:
             continue
         # value may contain any number of `=`
         name, value = match.string.split("=", 1)
-        name = name.strip().lower()
-        value = value.strip()
+        name = unquote(name).strip().lower()
+        value = unquote(value).strip()
         headers[name] = value
 
     return headers
diff --git a/opentelemetry-api/tests/util/test_re.py b/opentelemetry-api/tests/util/test_re.py
@@ -34,6 +34,12 @@ def test_parse_headers(self):
             # different header values
             ("name=", [("name", "")], False),
             ("name===value=", [("name", "==value=")], False),
+            # url-encoded headers
+            ("key=value%20with%20space", [("key", "value with space")], False),
+            ("key%21=value", [("key!", "value")], False),
+            ("%20key%20=%20value%20", [("key", "value")], False),
+            # header name case normalization
+            ("Key=Value", [("key", "Value")], False),
             # mix of valid and invalid headers
             (
                 "name1=value1,invalidName, name2 =   value2   , name3=value3==",
