From d32d232e05ae242c59744c541da07fbb6a923910 Mon Sep 17 00:00:00 2001
From: Roland van Laar <roland@rolandvanlaar.nl>
Date: Sun, 22 Aug 2021 00:31:10 +0200
Subject: [PATCH] DEVTOOLS: DUMPERCOMPANION: handle all filenames

Always punyencode filenames when the filename contains a char that
should be escaped. This makes it work better with windows.

Windows has a short list of chars that not allowed:
https://docs.microsoft.com/en-us/windows/win32/fileio/naming-a-file#naming-conventions
---
 devtools/dumper-companion.py | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/devtools/dumper-companion.py b/devtools/dumper-companion.py
index f73d078fd27..4bed0cc2891 100755
--- a/devtools/dumper-companion.py
+++ b/devtools/dumper-companion.py
@@ -163,6 +163,14 @@ def file_to_macbin(f: machfs.File, name: ByteString) -> bytes:
 
 
 def escape_string(s: str) -> str:
+    """
+    Escape strings
+
+    Escape the following:
+    - escape char: \x81
+    - unallowed filename chars: https://docs.microsoft.com/en-us/windows/win32/fileio/naming-a-file#naming-conventions
+    - control chars < 0x20
+    """
     new_name = ""
     for char in s:
         if char == "\x81":
@@ -174,6 +182,14 @@ def escape_string(s: str) -> str:
     return new_name
 
 
+def needs_punyencoding(orig: str) -> bool:
+    """
+    Filenames need punyencoding when it contains a char that should be
+    escaped.
+    """
+    return orig != escape_string(orig)
+
+
 def punyencode(orig: str) -> str:
     s = escape_string(orig)
     encoded = s.encode("punycode").decode("ascii")
@@ -219,8 +235,9 @@ def extract_volume(args: argparse.Namespace) -> int:
         for el in hpath:
             if japanese:
                 el = decode_macjapanese(el.encode("mac_roman"))
-            if punify:
+            if punify or needs_punyencoding(el):
                 el = punyencode(el)
+
             upath /= el
 
         if isinstance(obj, machfs.Folder):
@@ -454,6 +471,12 @@ def test_decode_name():
         assert punyencode(input) == expected
 
 
+def test_needs_punyencoding():
+    checks = [["Icon\r", True], ["ascii", False], ["バッドデイ(Power PC)", False]]
+    for input, expected in checks:
+        assert needs_punyencoding(input) == expected
+
+
 def test_escape_string():
     checks = [["\r", "\x81\x8d"], ["\x81", "\x81\x79"]]
     for input, expected in checks: