pull from lighteval everytime

app.py

@@ -15,22 +15,19 @@ This file stays outside the lighteval src tree, per request.
 """
 
 import ast
-import json
 import os
 import re
 from collections import Counter
-from dataclasses import asdict, dataclass
+from dataclasses import dataclass
 
 import gradio as gr
 
 
-REPO_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
+REPO_ROOT = "."
 TASK_DIRS = [
-    os.path.join(REPO_ROOT, "src", "lighteval", "tasks", "tasks"),
-    os.path.join(REPO_ROOT, "src", "lighteval", "tasks", "multilingual", "tasks"),
+    os.path.join(REPO_ROOT, "lighteval", "src", "lighteval", "tasks", "tasks"),
+    os.path.join(REPO_ROOT, "lighteval", "src", "lighteval", "tasks", "multilingual", "tasks"),
 ]
-# place cache inside repo root to avoid dirname('') issue
-CACHE_PATH = "tasks_index.json"
 
 
 star_benchmarks = [
@@ -112,7 +109,9 @@ def split_list_field(value: str) -> list[str]:
 
 def discover_task_files() -> list[str]:
     files: list[str] = []
+    print(f"Discovering task files in: {TASK_DIRS}")
     for base in TASK_DIRS:
+        print(f"Discovering task files in: {base}")
         if not os.path.isdir(base):
             continue
         # Top-level python files in the directory
@@ -167,42 +166,11 @@ def index_tasks() -> tuple[list[TaskDoc], list[str], list[str]]:
     return docs, languages_sorted, tags_sorted
 
 
-def save_index(path: str, tasks: list[TaskDoc], langs: list[str], tags: list[str]) -> None:
-    data = {
-        "tasks": [asdict(t) for t in tasks],
-        "languages": list(langs),
-        "tags": list(tags),
-    }
-    os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
-    with open(path, "w", encoding="utf-8") as f:
-        json.dump(data, f, ensure_ascii=False, indent=2)
+def build_index() -> tuple[list[TaskDoc], list[str], list[str]]:
+    return index_tasks()
 
 
-def load_index(path: str) -> tuple[list[TaskDoc], list[str], list[str]] | None:
-    if not os.path.exists(path):
-        return None
-    with open(path, "r", encoding="utf-8") as f:
-        data = json.load(f)
-    tasks = [TaskDoc(**t) for t in data.get("tasks", [])]
-    langs = list(data.get("languages", []))
-    tags = list(data.get("tags", []))
-    return tasks, langs, tags
-
-
-def build_and_cache_index() -> tuple[list[TaskDoc], list[str], list[str]]:
-    tasks, langs, tags = index_tasks()
-    save_index(CACHE_PATH, tasks, langs, tags)
-    return tasks, langs, tags
-
-
-_loaded = load_index(CACHE_PATH)
-if _loaded is None:
-    print("Building and caching index...")
-    ALL_TASKS, ALL_LANGS, ALL_TAGS = build_and_cache_index()
-else:
-    print("Loading index from cache...")
-    ALL_TASKS, ALL_LANGS, ALL_TAGS = _loaded
-print(f"Loaded {len(ALL_TASKS)} tasks from cache")
+ALL_TASKS, ALL_LANGS, ALL_TAGS = build_index()
 TOP_LANGS = ALL_LANGS[:8]  # show more by default
 
 
@@ -273,6 +241,7 @@ def render_cards(tasks: list[TaskDoc]) -> str:
         fallback_name = parts[-2] if base_no_ext == "main" and len(parts) >= 2 else base_no_ext
         task_name = (t.name or fallback_name).replace("_", " ").title()
         mod_path = t.module.replace("\\", "/")
+        mod_path = mod_path.split("/", 1)[1]
         source_html = f'<a href="https://github.com/huggingface/lighteval/blob/main/{mod_path}" target="_blank" rel="noopener">source</a>'
         paper_html = f'<a href="{t.paper}" target="_blank" rel="noopener">paper</a>' if t.paper else ""
         tags_html = " ".join([f'<span class=\"chip\" title=\"tag: {tag}\">{tag}</span>' for tag in t.tags]) if t.tags else ""
@@ -446,5 +415,18 @@ with gr.Blocks(title="Lighteval Tasks Explorer", css=None) as demo:
 
 
 if __name__ == "__main__":
+    from git import Repo  # pip install gitpython
+
+    git_url = "https://github.com/huggingface/lighteval.git"
+    repo_dir = "./lighteval"
+    
+    if os.path.exists(repo_dir) and os.path.isdir(os.path.join(repo_dir, ".git")):
+        print(f"Pulling latest changes from {git_url}...")
+        repo = Repo(repo_dir)
+        repo.remotes.origin.pull()
+    else:
+        print(f"Cloning {git_url} to {repo_dir}...")
+        Repo.clone_from(git_url, repo_dir)
+    
     # Run with `python benchmark_finder/app.py`
     demo.launch()