From b165722c58d265a1d1ebd20b2c2097f80ac7f1a6 Mon Sep 17 00:00:00 2001
From: Midgard <midgard@zeus.ugent.be>
Date: Mon, 9 Dec 2024 03:34:17 +0100
Subject: [PATCH] Add purging script

---
 README   | 12 +++++++++++-
 purge.py | 55 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+), 1 deletion(-)
 create mode 100755 purge.py

diff --git a/README b/README
index a2db790..4314c21 100644
--- a/README
+++ b/README
@@ -1,2 +1,12 @@
-Do something like
+Find and purge unused strings
+=============================
+
+Dependencies
+------------
+- python
+- ripgrep (only for find.py)
+
+Usage
+-----
 ./find.py ~/dev/mapcomplete > found.txt
+./purge.py ~/dev/mapcomplete found.txt
diff --git a/purge.py b/purge.py
new file mode 100755
index 0000000..b04bf94
--- /dev/null
+++ b/purge.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+
+import os
+import sys
+import json
+
+
+
+if len(sys.argv) != 3 or sys.argv[1].startswith("-"):
+	print(f"Usage: {sys.argv[0]} <path to mapcomplete repo> <path to found.txt>", file=sys.stderr)
+	sys.exit(1)
+
+path_mc_repo = sys.argv[1]
+path_found_txt = sys.argv[2]
+
+if not os.path.isdir(path_mc_repo):
+	print("First argument (mapcomplete repo) is not an existing directory", file=sys.stderr)
+	sys.exit(1)
+
+if not os.path.isfile(path_found_txt):
+	print("Second argument (found.txt) is not an existing file", file=sys.stderr)
+	sys.exit(1)
+
+
+with open(path_found_txt, "r", encoding="utf-8") as fh:
+	to_discard = set(map(str.strip, fh))
+
+
+def walk_and_discard(obj, prefix):
+	result = {}
+	for k, v in obj.items():
+		fqkey = f"{prefix}{k}"
+		if isinstance(v, str):
+			if fqkey not in to_discard:
+				result[k] = v
+		elif hasattr(v, "items"):
+			result[k] = walk_and_discard(v, f"{prefix}{k}.")
+	return result
+
+
+path_langs = os.path.join(path_mc_repo, "langs")
+json_files = [
+	f for f in os.listdir(path_langs)
+	if os.path.isfile(os.path.join(path_langs, f)) and f.endswith(".json")
+]
+for file in json_files:
+	path = os.path.join(path_langs, file)
+	print(f"Processing {path}", file=sys.stderr)
+	with open(path, "r", encoding="utf-8") as fh:
+		obj = json.load(fh)
+
+	new_obj = walk_and_discard(obj, "")
+
+	with open(path, "w", encoding="utf-8") as fh:
+		json.dump(new_obj, fh, indent=4, ensure_ascii=False)