From 5591a16cd73fa4a4c820e996686156056fc44565 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Mon, 3 Nov 2025 13:22:39 +0100
Subject: [PATCH] [formatter] add 'Lb' format specifier - 'L' for bytes

---
 docs/formatting.md      | 10 ++++++++++
 gallery_dl/formatter.py | 23 +++++++++++++++--------
 test/test_formatter.py  | 13 +++++++++++++
 3 files changed, 38 insertions(+), 8 deletions(-)

diff --git a/docs/formatting.md b/docs/formatting.md
index b81069cf..7adfee88 100644
--- a/docs/formatting.md
+++ b/docs/formatting.md
@@ -293,6 +293,16 @@ Format specifiers can be used for advanced formatting by using the options provi
     <td><code>{foo:L3/long/}</code></td>
     <td><code>long</code></td>
 </tr>
+<tr>
+    <td rowspan="2"><code>Lb&lt;maxlen&gt;/&lt;ext&gt;/</code></td>
+    <td rowspan="2">Same as <code>L</code>, but applies to the <a href="https://docs.python.org/3/library/stdtypes.html#bytes"><code>bytes()</code></a> representation of a string in <a href="https://docs.python.org/3/library/sys.html#sys.getfilesystemencoding">filesystem encoding</a></td>
+    <td><code>{foo_ja:Lb15/長い/}</code></td>
+    <td><code>フー・バー</code></td>
+</tr>
+<tr>
+    <td><code>{foo_ja:Lb8/長い/}</code></td>
+    <td><code>長い</code></td>
+</tr>
 <tr>
     <td rowspan="2"><code>X&lt;maxlen&gt;/&lt;ext&gt;/</code></td>
     <td rowspan="2">Limit output to <code>&lt;maxlen&gt;</code> characters. Cut output and add <code>&lt;ext&gt;</code> to its end if its length exceeds <code>&lt;maxlen&gt;</code></td>
diff --git a/gallery_dl/formatter.py b/gallery_dl/formatter.py
index 04f5cc8d..0f4f35ee 100644
--- a/gallery_dl/formatter.py
+++ b/gallery_dl/formatter.py
@@ -413,12 +413,20 @@ def _parse_conversion(format_spec, default):
 
 def _parse_maxlen(format_spec, default):
     maxlen, replacement, format_spec = format_spec.split(_SEPARATOR, 2)
-    maxlen = text.parse_int(maxlen[1:])
     fmt = _build_format_func(format_spec, default)
 
-    def mlen(obj):
-        obj = fmt(obj)
-        return obj if len(obj) <= maxlen else replacement
+    if maxlen[1] == "b":
+        maxlen = text.parse_int(maxlen[2:])
+
+        def mlen(obj):
+            obj = fmt(obj)
+            return obj if len(obj.encode(_ENCODING)) <= maxlen else replacement
+    else:
+        maxlen = text.parse_int(maxlen[1:])
+
+        def mlen(obj):
+            obj = fmt(obj)
+            return obj if len(obj) <= maxlen else replacement
     return mlen
 
 
@@ -501,13 +509,12 @@ def _parse_sort(format_spec, default):
     fmt = _build_format_func(format_spec, default)
 
     if "d" in args or "r" in args:
-        def sort_desc(obj):
+        def sort(obj):
             return fmt(sorted(obj, reverse=True))
-        return sort_desc
     else:
-        def sort_asc(obj):
+        def sort(obj):
             return fmt(sorted(obj))
-        return sort_asc
+    return sort
 
 
 def _parse_limit(format_spec, default):
diff --git a/test/test_formatter.py b/test/test_formatter.py
index b4603c06..0ad9461b 100644
--- a/test/test_formatter.py
+++ b/test/test_formatter.py
@@ -248,6 +248,19 @@ class TestFormatter(unittest.TestCase):
         self._run_test("{a:L50/foo/>51}", "foo")
         self._run_test("{a:Lab/foo/}", "foo")
 
+    def test_specifier_maxlen_bytes(self):
+        v = self.kwdict["a"]
+        self._run_test("{a:Lb5/foo/}" , "foo")
+        self._run_test("{a:Lb50/foo/}", v)
+        self._run_test("{a:Lb50/foo/>50}", " " * 39 + v)
+        self._run_test("{a:Lb50/foo/>51}", "foo")
+        self._run_test("{a:Lbab/foo/}", "foo")
+
+        v = self.kwdict["j"]
+        self._run_test("{j:Lb5/foo/}" , "foo")
+        self._run_test("{j:Lb50/foo/}", v)
+        self._run_test("{j:Lbab/foo/}", "foo")
+
     def test_specifier_join(self):
         self._run_test("{l:J}"       , "abc")
         self._run_test("{l:J,}"      , "a,b,c")