From dc5cf0bc49059d2cd55f727e66f7d9f5133ee7f9 Mon Sep 17 00:00:00 2001
From: WatchTree-19 <119982314+WatchTree-19@users.noreply.github.com>
Date: Mon, 27 Jul 2026 15:19:32 +0100
Subject: [PATCH] encoding: braille semicolon mapped to the wrong cell

The braille() encoder mapped ';' to U+2830 (dots 5,6), which is the
letter-sign / grade-1 indicator cell, not a semicolon. English Braille --
and UEB, and the French and German codes -- all encode the semicolon as
dots 2,3 = U+2806. Every other punctuation mark in the map already follows
that standard (comma dot 2, colon dots 2,5, period dots 2,5,6, ...), so this
was an isolated transcription error, not a code-variant choice.

Effect: any ';' in a payload was braille-encoded to the wrong character.

Added test_braille_punctuation_cells, which pins the exact cell for each
punctuation mark; it fails on the old value (';' -> U+2830) and passes on
U+2806.

Signed-off-by: WatchTree-19 <119982314+WatchTree-19@users.noreply.github.com>
---
 garak/probes/encoding.py             |  2 +-
 tests/probes/test_probes_encoding.py | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/garak/probes/encoding.py b/garak/probes/encoding.py
index f6f6e1910..c1afd6208 100644
--- a/garak/probes/encoding.py
+++ b/garak/probes/encoding.py
@@ -97,7 +97,7 @@ def braille(text):
         "?": "\u2826",
         "$": "\u2832",
         ":": "\u2812",
-        ";": "\u2830",
+        ";": "\u2806",
         "(": "\u2836",
         ")": "\u2836",
         "1": "\u2801",
diff --git a/tests/probes/test_probes_encoding.py b/tests/probes/test_probes_encoding.py
index 663f93ae7..d62c3c568 100644
--- a/tests/probes/test_probes_encoding.py
+++ b/tests/probes/test_probes_encoding.py
@@ -91,3 +91,25 @@ def test_atbash_function(plain, expected):
     encoder = InjectAtbash.atbash  # staticmethod reference
     result = encoder(plain.encode()).decode()
     assert result == expected
+
+
+def test_braille_punctuation_cells():
+    """Braille encoder must use the correct English/UEB cells for punctuation.
+
+    Regression for the semicolon, which was mapped to U+2830 (dots 5,6, the
+    letter-sign cell) instead of the semicolon cell U+2806 (dots 2,3). Every
+    other punctuation mark in the map already follows the standard, so this was
+    an isolated transcription error.
+    """
+    expected = {
+        b",": "⠂",  # dot 2
+        b";": "⠆",  # dots 2,3
+        b":": "⠒",  # dots 2,5
+        b".": "⠲",  # dots 2,5,6
+        b"!": "⠖",  # dots 2,3,5
+        b"?": "⠦",  # dots 2,3,6
+    }
+    for char, cell in expected.items():
+        assert (
+            garak.probes.encoding.braille(char).decode() == cell
+        ), f"{char!r} encoded to the wrong braille cell"