pearmut 1.0.1__py3-none-any.whl → 1.0.2__py3-none-any.whl

pearmut/app.py

@@ -4,7 +4,7 @@ from typing import Any
 
 from fastapi import FastAPI, Query
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import JSONResponse, Response
+from fastapi.responses import FileResponse, JSONResponse, Response
 from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel
 
@@ -17,6 +17,7 @@ from .results_export import (
 )
 from .utils import (
     ROOT,
+    TOKEN_MAIN,
     check_validation_threshold,
     load_progress_data,
     save_db_payload,
@@ -192,7 +193,11 @@ async def _dashboard_data(request: DashboardDataRequest):
         progress_new[user_id] = entry
 
     return JSONResponse(
-        content={"data": progress_new, "validation_threshold": validation_threshold},
+        content={
+            "data": progress_new,
+            "validation_threshold": validation_threshold,
+            "assignment": assignment,
+        },
         status_code=200,
     )
 
@@ -288,7 +293,7 @@ class PurgeCampaignRequest(BaseModel):
 @app.post("/purge-campaign")
 async def _purge_campaign(request: PurgeCampaignRequest):
     global progress_data, tasks_data
-    
+
     campaign_id = request.campaign_id
     token = request.token
 
@@ -298,57 +303,69 @@ async def _purge_campaign(request: PurgeCampaignRequest):
         return JSONResponse(content="Invalid token", status_code=400)
 
     # Unlink assets if they exist
-    destination = tasks_data[campaign_id].get("info", {}).get("assets", {}).get("destination")
+    destination = (
+        tasks_data[campaign_id].get("info", {}).get("assets", {}).get("destination")
+    )
     if destination:
         symlink_path = f"{ROOT}/data/{destination}".rstrip("/")
         if os.path.islink(symlink_path):
             os.remove(symlink_path)
-    
+
     # Remove task file
     task_file = f"{ROOT}/data/tasks/{campaign_id}.json"
     if os.path.exists(task_file):
         os.remove(task_file)
-    
+
     # Remove output file
     output_file = f"{ROOT}/data/outputs/{campaign_id}.jsonl"
     if os.path.exists(output_file):
         os.remove(output_file)
-    
+
     # Remove from in-memory data structures
     del tasks_data[campaign_id]
     del progress_data[campaign_id]
-    
+
     # Save updated progress data
     save_progress_data(progress_data)
-    
+
     return JSONResponse(content="ok", status_code=200)
 
 
 class AddCampaignRequest(BaseModel):
     campaign_data: dict[str, Any]
+    token_main: str
 
 
 @app.post("/add-campaign")
 async def _add_campaign(request: AddCampaignRequest):
     global progress_data, tasks_data
-    
+
     from .cli import _add_single_campaign
-    
+
+    if request.token_main != TOKEN_MAIN:
+        return JSONResponse(
+            content={"error": "Invalid main token. Use the latest one."},
+            status_code=400,
+        )
+
     try:
         server = f"{os.environ.get('PEARMUT_SERVER_URL', 'http://localhost:8001')}"
         _add_single_campaign(request.campaign_data, overwrite=False, server=server)
-        
-        campaign_id = request.campaign_data['campaign_id']
+
+        campaign_id = request.campaign_data["campaign_id"]
         with open(f"{ROOT}/data/tasks/{campaign_id}.json", "r") as f:
             tasks_data[campaign_id] = json.load(f)
-        
+
         progress_data = load_progress_data(warn=None)
-        
-        return JSONResponse(content={
-            "status": "ok",
-            "campaign_id": campaign_id,
-            "token": tasks_data[campaign_id]["token"]
-        }, status_code=200)
+
+        return JSONResponse(
+            content={
+                "status": "ok",
+                "campaign_id": campaign_id,
+                "token": tasks_data[campaign_id]["token"],
+            },
+            status_code=200,
+        )
     except Exception as e:
         return JSONResponse(content={"error": str(e)}, status_code=400)
 
@@ -418,6 +435,17 @@ if not os.path.exists(static_dir + "index.html"):
         "Static directory not found. Please build the frontend first."
     )
 
+# Serve HTML files directly without redirect
+@app.get("/annotate")
+async def serve_annotate():
+    return FileResponse(static_dir + "annotate.html")
+
+
+@app.get("/dashboard")
+async def serve_dashboard():
+    return FileResponse(static_dir + "dashboard.html")
+
+
 # Mount user assets from data/assets/
 assets_dir = f"{ROOT}/data/assets"
 os.makedirs(assets_dir, exist_ok=True)

pearmut/assignment.py

@@ -1,5 +1,4 @@
 import collections
-import copy
 import random
 import statistics
 from typing import Any
@@ -148,7 +147,7 @@ def get_i_item_taskbased(
             | {
                 k: v
                 for k, v in data_all[campaign_id]["info"].items()
-                if k in {"protocol", "sliders"}
+                if k in {"protocol", "sliders", "textfield", "show_model_names"}
             },
             "payload": data_all[campaign_id]["data"][user_id][item_i],
         }
@@ -195,7 +194,7 @@ def get_i_item_singlestream(
             | {
                 k: v
                 for k, v in data_all[campaign_id]["info"].items()
-                if k in {"protocol", "sliders"}
+                if k in {"protocol", "sliders", "textfield", "show_model_names"}
             },
             "payload": data_all[campaign_id]["data"][item_i],
         }
@@ -242,7 +241,7 @@ def get_next_item_taskbased(
             | {
                 k: v
                 for k, v in data_all[campaign_id]["info"].items()
-                if k in {"protocol", "sliders"}
+                if k in {"protocol", "sliders", "textfield", "show_model_names"}
             },
             "payload": data_all[campaign_id]["data"][user_id][item_i],
         }
@@ -298,7 +297,7 @@ def get_next_item_singlestream(
             | {
                 k: v
                 for k, v in data_all[campaign_id]["info"].items()
-                if k in {"protocol", "sliders"}
+                if k in {"protocol", "sliders", "textfield", "show_model_names"}
             },
             "payload": data_all[campaign_id]["data"][item_i],
         }
@@ -459,7 +458,7 @@ def get_next_item_dynamic(
             | {
                 k: v
                 for k, v in campaign_data["info"].items()
-                if k in {"protocol", "sliders"}
+                if k in {"protocol", "sliders", "textfield", "show_model_names"}
             },
             "payload": pruned_item,
         },
@@ -504,9 +503,15 @@ def reset_task(
     """
     Reset the task progress for the user in the specified campaign.
     Saves a reset marker to mask existing annotations.
+    
+    Note: Dynamic assignment does not support user-level deletion.
     """
     assignment = tasks_data[campaign_id]["info"]["assignment"]
-    if assignment == "task-based":
+    if assignment == "dynamic":
+        return JSONResponse(
+            content="User-level deletion is not supported for dynamic assignments", status_code=400
+        )
+    elif assignment == "task-based":
         # Save reset marker for this user to mask existing annotations
         num_items = len(tasks_data[campaign_id]["data"][user_id])
         for item_i in range(num_items):
@@ -533,30 +538,6 @@ def reset_task(
             for item_i in user_items:
                 progress_data[campaign_id][uid]["progress"][item_i] = False
         
-        # Reset only the specified user's time
-        _reset_user_time(progress_data, campaign_id, user_id)
-        return JSONResponse(content="ok", status_code=200)
-    elif assignment == "dynamic":
-        # Find all items that this user has annotated
-        user_items = _get_user_annotated_items(campaign_id, user_id)
-        
-        # Save reset markers only for items this user has touched
-        for item_i in user_items:
-            save_db_payload(
-                campaign_id,
-                {"user_id": user_id, "item_i": item_i, "annotation": RESET_MARKER},
-            )
-
-        progress_data_user = copy.deepcopy(progress_data[campaign_id][user_id]["progress"])
-        
-        # Reset only the touched items in all users' progress (shared pool, use lists to track models)
-        for uid in progress_data[campaign_id]:
-            for item_i in user_items:
-                progress_data[campaign_id][uid]["progress"][item_i] = [
-                    x for x in progress_data[campaign_id][uid]["progress"][item_i]
-                    if x not in progress_data_user[item_i]
-                ]
-        
         # Reset only the specified user's time
         _reset_user_time(progress_data, campaign_id, user_id)
         return JSONResponse(content="ok", status_code=200)

pearmut/cli.py

@@ -3,20 +3,44 @@ Command-line interface for managing and running the Pearmut server.
 """
 
 import argparse
+import atexit
+import fcntl
 import hashlib
 import json
 import os
 import urllib.parse
 
-import psutil
-
-from .utils import ROOT, load_progress_data, save_progress_data
+from .utils import ROOT, TOKEN_MAIN, load_progress_data, save_progress_data
 
 os.makedirs(f"{ROOT}/data/tasks", exist_ok=True)
 load_progress_data(warn=None)
 
 
+
 def _run(args_unknown):
+    # Acquire lock before starting server
+    lock_file = f"{ROOT}/data/.lock"
+    try:
+        lock_fd = open(lock_file, "a+")
+        fcntl.flock(lock_fd.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
+        lock_fd.seek(0)
+        lock_fd.truncate()
+        lock_fd.write(str(os.getpid()))
+        lock_fd.flush()
+    except BlockingIOError:
+        try:
+            with open(lock_file, "r") as f:
+                pid = f.read().strip()
+            print("You can't run multiple instances of Pearmut in the same directory.")
+            if pid:
+                print(f"Another instance (PID {pid}) is holding the lock.")
+        except (FileNotFoundError, PermissionError, OSError):
+            print("You can't run multiple instances of Pearmut in the same directory.")
+        exit(1)
+
+    # Register cleanup to remove lock file on exit
+    atexit.register(lambda: os.path.exists(lock_file) and os.remove(lock_file))
+
     import uvicorn
 
     from .app import app, tasks_data
@@ -33,21 +57,22 @@ def _run(args_unknown):
     args = args.parse_args(args_unknown)
 
     # print access dashboard URL for all campaigns
-    if tasks_data:
-        dashboard_url = (
-            args.server
-            + "/dashboard.html?"
-            + "&".join(
-                [
-                    f"campaign_id={urllib.parse.quote_plus(campaign_id)}&token={campaign_data["token"]}"
-                    for campaign_id, campaign_data in tasks_data.items()
-                ]
-            )
-        )
-        print(
-            "\033[92mNow serving Pearmut, use the following URL to access the everything-dashboard:\033[0m"
+    dashboard_url = (
+        args.server
+        + "/dashboard?"
+        + f"token_main={TOKEN_MAIN}"
+        + "".join(
+            [
+                f"&campaign_id={urllib.parse.quote_plus(campaign_id)}&token={campaign_data["token"]}"
+                for campaign_id, campaign_data in tasks_data.items()
+            ]
         )
-        print("🍐", dashboard_url + "\n", flush=True)
+    )
+    print(
+        "\033[92mNow serving Pearmut, use the following URL to access the everything-dashboard:\033[0m"
+    )
+    print("🍐", dashboard_url + "\n", flush=True)
+
 
     # disable startup message
     uvicorn.config.LOGGING_CONFIG["loggers"]["uvicorn.error"]["level"] = "WARNING"
@@ -90,7 +115,7 @@ def _validate_item_structure(items):
         if "ref" in item and not isinstance(item["ref"], str):
             raise ValueError("Item 'ref' must be a string")
 
-        # Validate tgt is a dictionary (basic template with model names)
+        # Validate tgt is a dictionary (annotate template with model names)
         if isinstance(item["tgt"], str):
             # String not allowed - suggest using dictionary (don't include user input to prevent injection)
             raise ValueError(
@@ -238,7 +263,7 @@ def _add_single_campaign(campaign_data, overwrite, server):
     if "assignment" not in campaign_data["info"]:
         raise ValueError("Campaign 'info' must contain 'assignment' field.")
 
-    # Template defaults to "basic" if not specified
+    # Template defaults to "annotate" if not specified
     assignment = campaign_data["info"]["assignment"]
     # use random words for identifying users
     rng = random.Random()
@@ -431,7 +456,7 @@ def _add_single_campaign(campaign_data, overwrite, server):
             "time_end": None,
             "time": 0,
             "url": (
-                f"{campaign_data['info'].get("template", "basic")}.html"
+                f"{campaign_data['info'].get('template', 'annotate')}"
                 f"?campaign_id={urllib.parse.quote_plus(campaign_data['campaign_id'])}"
                 f"&user_id={user_id}"
             ),
@@ -578,13 +603,6 @@ def main():
     )
     args, args_unknown = args.parse_known_args()
 
-    # enforce that only one pearmut process is running
-    for p in psutil.process_iter():
-        if "pearmut" == p.name() and p.pid != os.getpid():
-            print("Exit all running pearmut processes before running more commands.")
-            print(p)
-            exit(1)
-
     if args.command == "run":
         _run(args_unknown)
     elif args.command == "add":

pearmut/results_export.py

@@ -39,7 +39,7 @@ def compute_model_scores(campaign_id):
     # Compute model scores from annotations
     model_scores = collections.defaultdict(dict)
 
-    # Iterate through all tasks to find items with 'models' field (basic template)
+    # Iterate through all tasks to find items with 'models' field (annotate template)
     log = get_db_log(campaign_id)
     for entry in log:
         if "item" not in entry or "annotation" not in entry: