Build URL from file path is working

details_data_processor.py

@@ -5,6 +5,7 @@ import json
 import re
 import numpy as np
 import requests
+from urllib.parse import quote
 
 class DetailsDataProcessor:
     # Download 
@@ -42,25 +43,15 @@ class DetailsDataProcessor:
         return df
 
     @staticmethod
-    def generate_url(file_path):
-        base_url = 'https://huggingface.co/datasets/open-llm-leaderboard/details/resolve/main/'
-
-
-        organization = '64bits'
-        model = 'LexPodLM-13B'
-        filename = '_2023-07-25T13%3A41%3A51.227672.json'
-        # extract organization, model, and filename from file_path instead of hardcoding
-        # filename = file_path.split('/')[-1]
-
-
-
-        other_chunk = 'details_harness%7ChendrycksTest-moral_scenarios%7C5'
-        constructed_url = base_url + organization + '/' + model + '/' + other_chunk + filename
-        return constructed_url
-
-
-
-
+    def build_url(file_path):
+        segments = file_path.split('/')
+        bits = segments[1]
+        model_name = segments[2]
+        timestamp = segments[3].split('_')[1]
+        
+        url = f'https://huggingface.co/datasets/open-llm-leaderboard/details/resolve/main/{bits}/{model_name}/details_harness%7ChendrycksTest-moral_scenarios%7C5_{quote(timestamp, safe="")}'
+        print(url)
+        return url
     
     def pipeline(self):
         dataframes = []

test_details_data_processing.py

@@ -20,13 +20,16 @@ class TestDetailsDataProcessor(unittest.TestCase):
         self.assertTrue(os.path.exists('test.html'))
         os.remove('test.html')
 
-    def test_generate_url(self):
-        results_file_path = "64bits/LexPodLM-13B/results_2023-07-25T13:41:51.227672.json"
-        expected_url = 'https://huggingface.co/datasets/open-llm-leaderboard/details/resolve/main/64bits/LexPodLM-13B/details_harness%7ChendrycksTest-moral_scenarios%7C5_2023-07-25T13%3A41%3A51.227672.json'
-
-
-        constructed_url = self.processor.generate_url(results_file_path)
-        self.assertEqual(expected_url, constructed_url)
+    def test_build_url(self):
+        test_cases = [
+            ('results/64bits/LexPodLM-13B/results_2023-07-25T13:41:51.227672.json',
+            'https://huggingface.co/datasets/open-llm-leaderboard/details/resolve/main/64bits/LexPodLM-13B/details_harness%7ChendrycksTest-moral_scenarios%7C5_2023-07-25T13%3A41%3A51.227672.json'),
+            ('results/AlpinDale/pygmalion-instruct/results_2023-08-17T11:20:15.687659.json',
+            'https://huggingface.co/datasets/open-llm-leaderboard/details/resolve/main/AlpinDale/pygmalion-instruct/details_harness%7ChendrycksTest-moral_scenarios%7C5_2023-08-17T11%3A20%3A15.687659.json')
+        ]
+    
+        for file_path, expected in test_cases:
+            assert self.processor.build_url(file_path) == expected, f"Test failed for file_path: {file_path}"
 
     def test_pipeline(self):
         df = self.processor.pipeline()