Files
drobboxordinacebackup/recovery.py
2026-02-12 07:54:22 +01:00

145 lines
4.2 KiB
Python

"""
Recovery script: reconstruct directory tree from a specific run.
Usage: python recovery.py
(interactive — shows last 10 runs, asks which one to recover)
For a given run_id, finds all files that existed at that point
(first_seen_run <= run_id AND last_seen_run >= run_id)
and copies them from backup storage to output_dir preserving
the original directory structure.
"""
import os
import sys
import pyzipper
from indexer.config import BACKUP_PATH, BACKUP_PASSWORD
from indexer.db import get_connection
from indexer.backup import blob_path
DEFAULT_OUTPUT_DIR = r"U:\recovery"
def show_last_runs(n: int = 10):
"""Show last N completed runs and return the list."""
conn = get_connection()
cur = conn.cursor()
cur.execute(
"""SELECT id, started_at, finished_at, status,
files_total, files_new, files_modified, files_deleted
FROM runs
ORDER BY id DESC
LIMIT %s""",
(n,)
)
rows = cur.fetchall()
conn.close()
if not rows:
print("No runs found in DB.")
return []
print(f"\n{'='*80}")
print(f"{'Run':>5} {'Started':>19} {'Status':>10} {'Total':>7} {'New':>5} {'Mod':>5} {'Del':>5}")
print(f"{'-'*80}")
for row in reversed(rows):
run_id, started, finished, status, total, new, mod, deleted = row
started_str = started.strftime("%Y-%m-%d %H:%M:%S") if started else "?"
print(f"{run_id:>5} {started_str:>19} {status:>10} {total or 0:>7} {new or 0:>5} {mod or 0:>5} {deleted or 0:>5}")
print(f"{'='*80}")
return [r[0] for r in rows]
def recover(run_id: int, output_dir: str):
conn = get_connection()
cur = conn.cursor()
cur.execute(
"""SELECT relative_path, content_hash
FROM files
WHERE first_seen_run <= %s AND last_seen_run >= %s""",
(run_id, run_id)
)
rows = cur.fetchall()
conn.close()
if not rows:
print(f"No files found for run #{run_id}.")
return
print(f"\nRecovering {len(rows)} files from run #{run_id} to {output_dir}")
recovered = 0
missing = 0
password = BACKUP_PASSWORD.encode("utf-8")
for i, (relative_path, content_hash) in enumerate(rows, 1):
source = blob_path(BACKUP_PATH, content_hash)
target = os.path.join(output_dir, relative_path.replace("/", os.sep))
if not os.path.exists(source):
print(f" MISSING zip: {content_hash.hex()} for {relative_path}")
missing += 1
continue
os.makedirs(os.path.dirname(target), exist_ok=True)
try:
with pyzipper.AESZipFile(source, "r") as zf:
zf.setpassword(password)
names = zf.namelist()
if not names:
print(f" WARN: empty zip: {source}")
missing += 1
continue
data = zf.read(names[0])
with open(target, "wb") as f:
f.write(data)
recovered += 1
except Exception as e:
print(f" ERROR extracting {source} for {relative_path}: {e}")
missing += 1
continue
if i % 1000 == 0:
print(f" [{i}/{len(rows)}] recovered={recovered} missing={missing}")
print(f"\n{'='*60}")
print(f"Recovery complete.")
print(f" Run : #{run_id}")
print(f" Output dir : {output_dir}")
print(f" Recovered : {recovered}")
print(f" Missing/err : {missing}")
print(f"{'='*60}")
if __name__ == "__main__":
run_ids = show_last_runs(10)
if not run_ids:
sys.exit(1)
print()
choice = input("Enter run ID to recover (or 'q' to quit): ").strip()
if choice.lower() == "q":
print("Aborted.")
sys.exit(0)
try:
run_id = int(choice)
except ValueError:
print(f"Invalid run ID: {choice}")
sys.exit(1)
output_dir = DEFAULT_OUTPUT_DIR
print(f"\nOutput directory: {output_dir}")
if os.path.exists(output_dir) and os.listdir(output_dir):
confirm = input("Directory is not empty. Continue? (y/n): ").strip().lower()
if confirm != "y":
print("Aborted.")
sys.exit(0)
os.makedirs(output_dir, exist_ok=True)
recover(run_id, output_dir)