From 56213ebc0363d296928897e330642b3de1b6ff3a Mon Sep 17 00:00:00 2001
From: sa6anw <joakim@joasve.se>
Date: Fri, 18 Jul 2025 20:04:25 +0000
Subject: [PATCH] i18n

---
 clean_language_blocks.py | 65 +++++++++++++++++++++++-----------------
 1 file changed, 37 insertions(+), 28 deletions(-)

diff --git a/clean_language_blocks.py b/clean_language_blocks.py
index 838eee4..3b3ce9f 100644
--- a/clean_language_blocks.py
+++ b/clean_language_blocks.py
@@ -3,47 +3,56 @@ import re
 import shutil
 
 SOURCE_DIR = "docs"
-ENGLISH_DIR = os.path.join(SOURCE_DIR, "en")
+ENGLISH_DIR = "docs/en"
 
-# Kopiera alla filer från docs/ till docs/en/
 if os.path.exists(ENGLISH_DIR):
     shutil.rmtree(ENGLISH_DIR)
 shutil.copytree(SOURCE_DIR, ENGLISH_DIR)
 
+# Hantera block i formatet:
+# :sv
+# text...
+# :::
+def remove_other_language_blocks(content, keep_lang):
+    block_pattern = re.compile(r":(sv|en)\n(.*?)\n:::", re.DOTALL)
+
+    result = []
+    pos = 0
+    for match in block_pattern.finditer(content):
+        start, end = match.span()
+        lang, block = match.groups()
+
+        # Behåll neutral text mellan block
+        result.append(content[pos:start])
+        if lang == keep_lang:
+            result.append(block.strip() + "\n\n")
+        pos = end
+    result.append(content[pos:])
+    return ''.join(result)
+
+# Inline-taggar, t.ex. [:sv]text
+def remove_other_language_inline(content, keep_lang):
+    def replacer(match):
+        lang, text = match.groups()
+        return text if lang == keep_lang else ""
+    return re.sub(r"\[:(sv|en)](.*?)(?=(\[:|$))", replacer, content)
+
 def clean_file(path, keep_lang):
     with open(path, "r", encoding="utf-8") as f:
         content = f.read()
 
-    # Ta bort språkblock som inte ska behållas
-    if keep_lang == "sv":
-        content = re.sub(r":::\s*en\s*\n(.*?)\n:::", "", content, flags=re.DOTALL)
-    elif keep_lang == "en":
-        content = re.sub(r":::\s*sv\s*\n(.*?)\n:::", "", content, flags=re.DOTALL)
-
-    # Ta bort språkblock-taggar för det språk vi behåller
-    content = re.sub(r":::\s*(sv|en)", "", content)
-    content = re.sub(r":::", "", content)
-
-    # Hantera inline-översättningar: [:sv]Text[:], [:en]Text[:]
-    def inline_replacer(match):
-        lang = match.group(1)
-        text = match.group(2)
-        return text if lang == keep_lang else ""
-
-    # Använd re.DOTALL för att hantera radbrytningar i inline-taggar
-    content = re.sub(r"\[:(sv|en)](.*?)\[:\]", inline_replacer, content, flags=re.DOTALL)
+    content = remove_other_language_blocks(content, keep_lang)
+    content = remove_other_language_inline(content, keep_lang)
 
     with open(path, "w", encoding="utf-8") as f:
         f.write(content.strip() + "\n")
 
-# Rensa svenska filer i docs/
+# Rensa svenska filer
 for filename in os.listdir(SOURCE_DIR):
-    path = os.path.join(SOURCE_DIR, filename)
-    if os.path.isfile(path) and filename.endswith(".md"):
-        clean_file(path, keep_lang="sv")
+    if filename.endswith(".md"):
+        clean_file(os.path.join(SOURCE_DIR, filename), keep_lang="sv")
 
-# Rensa engelska filer i docs/en/
+# Rensa engelska filer
 for filename in os.listdir(ENGLISH_DIR):
-    path = os.path.join(ENGLISH_DIR, filename)
-    if os.path.isfile(path) and filename.endswith(".md"):
-        clean_file(path, keep_lang="en")
+    if filename.endswith(".md"):
+        clean_file(os.path.join(ENGLISH_DIR, filename), keep_lang="en")