From 4ba624efae2d63057c2bd5be23841be5017bd457 Mon Sep 17 00:00:00 2001 From: sterni Date: Tue, 11 Jul 2023 17:54:20 +0200 Subject: fix(tvix/eval): use byte, not codepoint index for slicing in escape This fixes a subtle issue which would occasionally lead to a crash (e.g. when evaluating (pkgs.systemd.outPath with --trace-runtime): With each character in the string that has a multi byte representation in UTF-8, the actual byte position and what tvix thought it was would get out of sync. This could either lead to * Tvix swallowing characters or jumbling characters if multi byte characters would cause the tracked index to become out of sync with the byte position before the first character to be escaped, or * Tvix crashing if (in the same situation) the out of sync index would be within a UTF-8 byte sequence. Luckily, std's `char_indices()` iterator implements exactly what `nix_escape_char()`'s original author had in mind with `.chars().enumerate()`. Using `i + 1` for continuing is safe, since all characters that need (in fact, can) to be escaped in Nix are represented as a single byte in UTF-8. Change-Id: I1c836f70cde3d72db1c644e9112852f0d824715e Reviewed-on: https://cl.tvl.fyi/c/depot/+/8952 Reviewed-by: tazjin Tested-by: BuildkiteCI Autosubmit: sterni --- tvix/eval/src/value/string.rs | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'tvix/eval/src/value') diff --git a/tvix/eval/src/value/string.rs b/tvix/eval/src/value/string.rs index 7144ca360d15..2649e00f0830 100644 --- a/tvix/eval/src/value/string.rs +++ b/tvix/eval/src/value/string.rs @@ -193,7 +193,7 @@ fn is_valid_nix_identifier(s: &str) -> bool { /// /// Note that this does not add the outer pair of surrounding quotes. fn nix_escape_string(input: &str) -> Cow { - let mut iter = input.chars().enumerate().peekable(); + let mut iter = input.char_indices().peekable(); while let Some((i, c)) = iter.next() { if let Some(esc) = nix_escape_char(c, iter.peek().map(|(_, c)| c)) { @@ -201,6 +201,10 @@ fn nix_escape_string(input: &str) -> Cow { escaped.push_str(&input[..i]); escaped.push_str(esc); + // In theory we calculate how many bytes it takes to represent `esc` + // in UTF-8 and use that for the offset. It is, however, safe to + // assume that to be 1, as all characters that can be escaped in a + // Nix string are ASCII. let mut inner_iter = input[i + 1..].chars().peekable(); while let Some(c) = inner_iter.next() { match nix_escape_char(c, inner_iter.peek()) { -- cgit 1.4.1