From d27d106ed11ee9b951660b93f42f49041313ebbc Mon Sep 17 00:00:00 2001
From: benoit74 <benoit74@users.noreply.github.com>
Date: Fri, 13 Sep 2024 13:03:42 +0000
Subject: [PATCH] Utility script to create a monthly report of tasks ran on a
 Zimfarm instance

---
 .../maint-scripts/monthly_report_tasks.py     | 78 +++++++++++++++++++
 1 file changed, 78 insertions(+)
 create mode 100644 dispatcher/backend/maint-scripts/monthly_report_tasks.py

diff --git a/dispatcher/backend/maint-scripts/monthly_report_tasks.py b/dispatcher/backend/maint-scripts/monthly_report_tasks.py
new file mode 100644
index 00000000..00db1b3b
--- /dev/null
+++ b/dispatcher/backend/maint-scripts/monthly_report_tasks.py
@@ -0,0 +1,78 @@
+# Creates a CSV report of tasks completed (successfully or not)
+# during last month (for 1st to last day of the month)
+
+import datetime
+import logging
+import os
+from pathlib import Path
+
+import requests
+from dateutil.relativedelta import relativedelta
+
+logging.basicConfig(
+    level=logging.DEBUG, format="[%(asctime)s: %(levelname)s] %(message)s"
+)
+logger = logging.getLogger(__name__)
+
+# url of the zimfarm API to request
+url = os.getenv("ZF_URI", "https://api.farm.zimit.kiwix.org/v1")
+
+# prefix of csv file to create: {prefix}{year}-{month}.csv
+file_prefix = os.getenv("FILE_PREFIX", "zimfarm_tasks_")
+
+# list of offliners to include
+offliners = os.getenv("OFFLINERS", "zimit").split(",")
+
+
+def main():
+    """Creates recipes for TED by topics"""
+
+    now = datetime.datetime.now(datetime.UTC)
+    start_of_this_month = now.replace(day=1, hour=0, minute=0, second=0, microsecond=0)
+    start_of_last_month = start_of_this_month - relativedelta(months=1)
+
+    nb_per_page = 100
+    current_page = 0
+
+    last_task_found = False
+    with open(
+        Path(
+            f"{file_prefix}{start_of_last_month.year}-{start_of_last_month.month}.csv"
+        ),
+        "w",
+    ) as fh:
+        fh.write("ID,URL,Status,Requested,Started,Completed\n")
+        while not last_task_found:
+            response = requests.get(
+                f"{url}/tasks/?skip={current_page*nb_per_page}&limit={nb_per_page}"
+                "&status=failed&status=canceled&status=succeeded"
+            )
+            response.raise_for_status()
+            items = response.json()["items"]
+            for item in items:
+                item_last_modification = datetime.datetime.fromisoformat(
+                    item["updated_at"]
+                )
+                if item_last_modification >= start_of_this_month:
+                    continue  # too recent
+                if item_last_modification < start_of_last_month:
+                    last_task_found = True  # we found the most ancient task to report
+                    break
+
+                response = requests.get(f'{url}/tasks/{item["_id"]}')
+                response.raise_for_status()
+                task = response.json()
+                offliner = task["config"]["task_name"]
+                if offliners != "all" and offliner not in offliners:
+                    continue
+                fh.write(
+                    f'{task["_id"]},{task["config"]["flags"]["url"]},{task["status"]},'
+                    f'{task["timestamp"]["requested"]},'
+                    f'{task["timestamp"].get("started", "-")},{task["updated_at"]}\n'
+                )
+            current_page += 1
+            fh.flush()
+
+
+if __name__ == "__main__":
+    main()