DON'T MERGE YET: update: API calls to pass API key via auth headers (All routes including evaluation routes)

src/e2etests/judgment_client_test.py

@@ -377,7 +377,11 @@ def test_fetch_traces_by_time_period(self, client: JudgmentClient):
         for hours in [1, 3, 6, 12, 24, 72, 168]:
             response = requests.post(
                 f"{SERVER_URL}/traces/fetch_by_time_period/",
-                json={"hours": hours, "judgment_api_key": API_KEY}
+                headers={
+                    "Content-Type": "application/json",
+                    "Authorization": f"Bearer {API_KEY}"
+                },
+                json={"hours": hours}
             )
             assert response.status_code == 200
             data = response.json()
@@ -388,17 +392,24 @@ def test_fetch_traces_invalid_period(self, client: JudgmentClient):
         for hours in [0, 2, 4]:
             response = requests.post(
                 f"{SERVER_URL}/traces/fetch_by_time_period/",
-                json={"hours": hours, "judgment_api_key": API_KEY}
+                headers={
+                    "Content-Type": "application/json",
+                    "Authorization": f"Bearer {API_KEY}"
+                },
+                json={"hours": hours}
             )
             assert response.status_code == 400
 
     def test_fetch_traces_missing_api_key(self, client: JudgmentClient):
         """Test missing API key scenario."""
         response = requests.post(
             f"{SERVER_URL}/traces/fetch_by_time_period/",
+            headers={
+                "Content-Type": "application/json",
+            },
             json={"hours": 12}
         )
-        assert response.status_code == 422
+        assert response.status_code in [401, 403]
 
 @pytest.mark.skipif(not os.getenv("GOOGLE_APPLICATION_CREDENTIALS"),
                    reason="VertexAI credentials not configured")

src/e2etests/test_tracer.py

@@ -3,6 +3,7 @@
 import time
 import asyncio
 from typing import List
+import pytest
 
 # Third-party imports
 from openai import OpenAI
@@ -136,9 +137,17 @@ def trace_manager_client():
 async def test_token_counting(trace_manager_client):
     input = "Write a poem about Nissan R32 GTR"
 
+@pytest.fixture
+def test_input():
+    """Fixture providing default test input"""
+    return "What if these shoes don't fit?"
+
+@pytest.mark.asyncio
+async def test_evaluation_mixed(test_input):
     PROJECT_NAME = "TestingPoemBot"
+    print(f"Using test input: {test_input}")
     with judgment.trace("Use-claude-hehexd123", project_name=PROJECT_NAME, overwrite=True) as trace:
-        upper = await make_upper(input)
+        upper = await make_upper(test_input)
         result = await make_poem(upper)
         await answer_user_question("What if these shoes don't fit?")
 

src/judgeval/common/tracer.py

@@ -199,10 +199,11 @@ def fetch_trace(self, trace_id: str):
             JUDGMENT_TRACES_FETCH_API_URL,
             json={
                 "trace_id": trace_id,
-                "judgment_api_key": self.judgment_api_key,
+                # "judgment_api_key": self.judgment_api_key,
             },
             headers={
                 "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
             }
         )
 
@@ -225,6 +226,7 @@ def save_trace(self, trace_data: dict, empty_save: bool):
             json=trace_data,
             headers={
                 "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
             }
         )
 
@@ -248,6 +250,7 @@ def delete_trace(self, trace_id: str):
             },
             headers={
                 "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
             }
         )
 
@@ -263,11 +266,12 @@ def delete_traces(self, trace_ids: List[str]):
         response = requests.delete(
             JUDGMENT_TRACES_DELETE_API_URL,
             json={
-                "judgment_api_key": self.judgment_api_key,
+                # "judgment_api_key": self.judgment_api_key,
                 "trace_ids": trace_ids,
             },
             headers={
                 "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
             }
         )
 
@@ -576,6 +580,25 @@ def save(self, empty_save: bool = False, overwrite: bool = False) -> Tuple[str,
 
         self.trace_manager_client.save_trace(trace_data, empty_save)
 
+
+        # Save trace data by making POST request to API
+        response = requests.post(
+            JUDGMENT_TRACES_SAVE_API_URL,
+            json=trace_data,
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.tracer.api_key}"  # Bearer token format
+            }
+        )
+
+        if response.status_code == HTTPStatus.BAD_REQUEST:
+            raise ValueError(f"Failed to save trace data: Check your Trace name for conflicts, set overwrite=True to overwrite existing traces: {response.text}")
+        elif response.status_code != HTTPStatus.OK:
+            raise ValueError(f"Failed to save trace data: {response.text}")
+
+        if not empty_save and "ui_results_url" in response.json():
+            rprint(f"\n🔍 You can view your trace data here: [rgb(106,0,255)]{response.json()['ui_results_url']}[/]\n")
+
         return self.trace_id, trace_data
 
     def delete(self):

src/judgeval/data/datasets/eval_dataset_client.py

@@ -56,12 +56,16 @@ def push(self, dataset: EvalDataset, alias: str,overwrite: Optional[bool] = Fals
                     "ground_truths": [g.to_dict() for g in dataset.ground_truths],
                     "examples": [e.to_dict() for e in dataset.examples],
                     "overwrite": overwrite,
-                    "judgment_api_key": dataset.judgment_api_key
+                    # "judgment_api_key": dataset.judgment_api_key
                 }
             try:
                 response = requests.post(
                     JUDGMENT_DATASETS_PUSH_API_URL, 
-                    json=content
+                    json=content,
+                    headers={
+                        "Content-Type": "application/json",
+                        "Authorization": f"Bearer {self.judgment_api_key}"
+                    }
                 )
                 if response.status_code == 500:
                     error(f"Server error during push: {content.get('message')}")
@@ -115,13 +119,17 @@ def pull(self, alias: str) -> EvalDataset:
                 )
                 request_body = {
                     "alias": alias,
-                    "judgment_api_key": self.judgment_api_key
+                    # "judgment_api_key": self.judgment_api_key
                 }
 
                 try:
                     response = requests.post(
                         JUDGMENT_DATASETS_PULL_API_URL, 
-                        json=request_body
+                        json=request_body,
+                        headers={
+                            "Content-Type": "application/json",
+                            "Authorization": f"Bearer {self.judgment_api_key}"
+                        }
                     )
                     response.raise_for_status()
                 except requests.exceptions.RequestException as e:
@@ -169,13 +177,17 @@ def pull_all_user_dataset_stats(self) -> dict:
                     total=100,
                 )
                 request_body = {
-                    "judgment_api_key": self.judgment_api_key
+                    # "judgment_api_key": self.judgment_api_key
                 }
 
                 try:
                     response = requests.post(
                         JUDGMENT_DATASETS_PULL_ALL_API_URL, 
-                        json=request_body
+                        json=request_body,
+                        headers={
+                            "Content-Type": "application/json",
+                            "Authorization": f"Bearer {self.judgment_api_key}"
+                        }
                     )
                     response.raise_for_status()
                 except requests.exceptions.RequestException as e:

src/judgeval/judgment_client.py

@@ -182,6 +182,10 @@ def pull_eval(self, project_name: str, eval_run_name: str) -> List[Dict[str, Uni
                                                    eval_name=eval_run_name, 
                                                    judgment_api_key=self.judgment_api_key)
         eval_run = requests.post(JUDGMENT_EVAL_FETCH_API_URL,
+                                 headers={
+                                    "Content-Type": "application/json",
+                                    "Authorization": f"Bearer {self.judgment_api_key}"
+                                 },
                                  json=eval_run_request_body.model_dump())
         if eval_run.status_code != requests.codes.ok:
             raise ValueError(f"Error fetching eval results: {eval_run.json()}")
@@ -213,6 +217,7 @@ def delete_eval(self, project_name: str, eval_run_name: str) -> bool:
                         json=eval_run_request_body.model_dump(),
                         headers={
                             "Content-Type": "application/json",
+                            "Authorization": f"Bearer {self.judgment_api_key}"
                         })
         if response.status_code != requests.codes.ok:
             raise ValueError(f"Error deleting eval results: {response.json()}")
@@ -235,6 +240,7 @@ def delete_project_evals(self, project_name: str) -> bool:
                         },
                         headers={
                             "Content-Type": "application/json",
+                            "Authorization": f"Bearer {self.judgment_api_key}"
                         })
         if response.status_code != requests.codes.ok:
             raise ValueError(f"Error deleting eval results: {response.json()}")
@@ -246,7 +252,11 @@ def _validate_api_key(self):
         """
         response = requests.post(
             f"{ROOT_API}/validate_api_key/",
-            json={"api_key": self.judgment_api_key}
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}",
+            },
+            json={}  # Empty body now
         )
         if response.status_code == 200:
             return True, response.json()
@@ -268,12 +278,16 @@ def fetch_classifier_scorer(self, slug: str) -> ClassifierScorer:
         """
         request_body = {
             "slug": slug,
-            "judgment_api_key": self.judgment_api_key
+            # "judgment_api_key": self.judgment_api_key
         }
 
         response = requests.post(
             f"{ROOT_API}/fetch_scorer/",
-            json=request_body
+            json=request_body,
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
+            }
         )
 
         if response.status_code == 500:
@@ -306,13 +320,17 @@ def push_classifier_scorer(self, scorer: ClassifierScorer, slug: str = None) ->
             "name": scorer.name,
             "conversation": scorer.conversation,
             "options": scorer.options,
-            "judgment_api_key": self.judgment_api_key,
+            # "judgment_api_key": self.judgment_api_key, 
             "slug": slug
         }
 
         response = requests.post(
             f"{ROOT_API}/save_scorer/",
-            json=request_body
+            json=request_body,
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.judgment_api_key}"
+            }
         )
 
         if response.status_code == 500:

src/judgeval/run_evaluation.py

@@ -47,7 +47,12 @@ def execute_api_eval(evaluation_run: EvaluationRun) -> List[Dict]:
     try:
         # submit API request to execute evals
         payload = evaluation_run.model_dump(warnings=False)
-        response = requests.post(JUDGMENT_EVAL_API_URL, json=payload)
+        response = requests.post(
+            JUDGMENT_EVAL_API_URL, headers={
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {evaluation_run.judgment_api_key}"
+        }, 
+        json=payload)
         response_data = response.json()
     except Exception as e:
         error(f"Error: {e}")
@@ -151,6 +156,10 @@ def check_eval_run_name_exists(eval_name: str, project_name: str, judgment_api_k
     try:
         response = requests.post(
             f"{ROOT_API}/eval-run-name-exists/",
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {judgment_api_key}"
+            },
             json={
                 "eval_name": eval_name,
                 "project_name": project_name,
@@ -188,6 +197,10 @@ def log_evaluation_results(merged_results: List[ScoringResult], evaluation_run:
     try:
         res = requests.post(
             JUDGMENT_EVAL_LOG_API_URL,
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {evaluation_run.judgment_api_key}"
+            },
             json={
                 "results": [result.to_dict() for result in merged_results],
                 "judgment_api_key": evaluation_run.judgment_api_key,