From d6ef04ba3e528acc219a61b815b45b1d3076de1f Mon Sep 17 00:00:00 2001
From: Piero Toffanin <pt@masseranolabs.com>
Date: Sun, 22 May 2022 16:17:41 -0400
Subject: [PATCH] Extract JSONL from suggestions.db script

---
 suggestions-to-jsonl.py | 46 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)
 create mode 100755 suggestions-to-jsonl.py

diff --git a/suggestions-to-jsonl.py b/suggestions-to-jsonl.py
new file mode 100755
index 0000000..7878d52
--- /dev/null
+++ b/suggestions-to-jsonl.py
@@ -0,0 +1,46 @@
+#!/usr/bin/env python
+import argparse
+import time
+import sqlite3
+import json
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Program to generate JSONL files from a LibreTranslate's suggestions.db")
+    parser.add_argument(
+        "--db",
+        type=str,
+        nargs=1,
+        help="Path to suggestions.db file",
+        default='suggestions.db'
+    )
+    parser.add_argument(
+        "--clear",
+        action='store_true',
+        help="Clear suggestions.db after generation",
+        default=False
+    )
+    args = parser.parse_args()
+
+    output_file = str(int(time.time())) + ".jsonl"
+
+    con = sqlite3.connect(args.db, check_same_thread=False)
+    cur = con.cursor()
+    
+    with open(output_file, 'w', encoding="utf-8") as f:
+        for row in cur.execute('SELECT q, s, source, target FROM suggestions WHERE source != "auto" ORDER BY source'):
+            q, s, source, target = row
+            obj = {
+                'q': q,
+                's': s, 
+                'source': source,
+                'target': target
+            }
+            json.dump(obj, f, ensure_ascii=False)
+            f.write('\n')
+    
+    print("Wrote %s" % output_file)
+
+    if args.clear:
+        cur.execute("DELETE FROM suggestions")
+        con.commit()
+        print("Cleared " + args.db)
\ No newline at end of file