From 089faff3a3571e3ecb5be2c61d403a7f778f2b60 Mon Sep 17 00:00:00 2001
From: Mark Eaton <MarkEEaton@users.noreply.github.com>
Date: Wed, 17 Sep 2025 19:08:44 -0400
Subject: [PATCH] feat: add script to pretty-print JSON Lines (.jsonl) files
 with indentation option

Co-authored-by: aider (o3) <aider@aider.chat>
---
 scripts/jsonl_pretty_print.py | 64 +++++++++++++++++++++++++++++++++++
 1 file changed, 64 insertions(+)
 create mode 100644 scripts/jsonl_pretty_print.py

diff --git a/scripts/jsonl_pretty_print.py b/scripts/jsonl_pretty_print.py
new file mode 100644
index 0000000..d8332ed
--- /dev/null
+++ b/scripts/jsonl_pretty_print.py
@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+"""
+Pretty-print a JSON Lines (.jsonl) file.
+
+Usage
+-----
+python scripts/jsonl_pretty_print.py path/to/file.jsonl [--indent 2]
+
+The script reads each line, parses it as JSON, and writes a
+human-readable representation to stdout.  Redirect or pipe as needed.
+
+Example
+-------
+python scripts/jsonl_pretty_print.py data.jsonl --indent 4 > pretty.json
+"""
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+from pathlib import Path
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Pretty-print a .jsonl file.")
+    parser.add_argument(
+        "jsonl_path",
+        type=Path,
+        help="Path to the input .jsonl file",
+    )
+    parser.add_argument(
+        "--indent",
+        type=int,
+        default=2,
+        help="Number of spaces to indent (default: 2)",
+    )
+    return parser.parse_args()
+
+
+def pretty_print_jsonl(path: Path, indent: int) -> None:
+    if not path.exists():
+        sys.exit(f"Error: {path} does not exist")
+
+    with path.open("r", encoding="utf-8") as fp:
+        for idx, line in enumerate(fp, start=1):
+            line = line.strip()
+            if not line:
+                continue  # skip empty lines
+            try:
+                obj = json.loads(line)
+            except json.JSONDecodeError as exc:
+                sys.exit(f"JSON decode error on line {idx}: {exc}")  # fail fast
+            # Dump the pretty JSON followed by a newline separator
+            print(json.dumps(obj, indent=indent, ensure_ascii=False, sort_keys=True))
+            print()  # blank line between records for readability
+
+
+def main() -> None:
+    args = parse_args()
+    pretty_print_jsonl(args.jsonl_path, args.indent)
+
+
+if __name__ == "__main__":
+    main()