about summary refs log tree commit diff
path: root/tvix/eval/src/tests/tvix_tests
diff options
context:
space:
mode:
Diffstat (limited to 'tvix/eval/src/tests/tvix_tests')
-rw-r--r--tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.exp1
-rw-r--r--tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.nix6
2 files changed, 7 insertions, 0 deletions
diff --git a/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.exp b/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.exp
new file mode 100644
index 000000000000..d889063f9ab1
--- /dev/null
+++ b/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.exp
@@ -0,0 +1 @@
+"๐Ÿ’ญ(\":thonking:\")"
diff --git a/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.nix b/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.nix
new file mode 100644
index 000000000000..49f4b6273106
--- /dev/null
+++ b/tvix/eval/src/tests/tvix_tests/eval-okay-escape-string-correct-char-boundaries.nix
@@ -0,0 +1,6 @@
+# Regression test for a bug where tvix would crash in nix_escape_string
+# because it counted the string position by unicode code point count,
+# but then used it as a byte index for slicing. Consequently, it would
+# try slicing ๐Ÿ’ญ in half, thinking the first element to be escaped was
+# at byte index 2 (i.e. the quote).
+"๐Ÿ’ญ(\":thonking:\")"