From 0b48f4a46d24248e4c012ef37d25b5963c27a78c Mon Sep 17 00:00:00 2001 From: nsfisis Date: Sun, 23 Aug 2026 12:43:10 +0900 Subject: fix(fs): carry file_get_contents results as bytes file_get_contents() and file_get_contents_with_max_length() return Vec instead of a from_utf8_lossy'd String. Call sites whose consumer takes a &str still convert lossily and are marked TODO(bytes). file_get_contents_with_max_length() now reads at most the requested number of bytes instead of the whole file. Co-Authored-By: Claude Opus 5 (1M context) --- crates/shirabe-php-shim/src/string.rs | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'crates/shirabe-php-shim/src/string.rs') diff --git a/crates/shirabe-php-shim/src/string.rs b/crates/shirabe-php-shim/src/string.rs index 024a3943..3a61a59d 100644 --- a/crates/shirabe-php-shim/src/string.rs +++ b/crates/shirabe-php-shim/src/string.rs @@ -299,11 +299,8 @@ pub fn trim(s: &str, chars: Option<&str>) -> String { } // Byte-based, matching PHP's substr. A negative start/length counts from the end. -// The result is reinterpreted as UTF-8 (lossily), which only matters when a slice -// boundary falls inside a multibyte sequence. -pub fn substr(s: &str, start: i64, length: Option) -> String { - let bytes = s.as_bytes(); - let len = bytes.len() as i64; +pub fn substr_bytes(s: &[u8], start: i64, length: Option) -> Vec { + let len = s.len() as i64; let start = if start < 0 { (len + start).max(0) } else { @@ -314,7 +311,13 @@ pub fn substr(s: &str, start: i64, length: Option) -> String { Some(l) if l < 0 => (len + l).max(start), Some(l) => (start + l).min(len), }; - String::from_utf8_lossy(&bytes[start as usize..end as usize]).into_owned() + s[start as usize..end as usize].to_vec() +} + +// The result is reinterpreted as UTF-8 (lossily), which only matters when a slice +// boundary falls inside a multibyte sequence. +pub fn substr(s: &str, start: i64, length: Option) -> String { + String::from_utf8_lossy(&substr_bytes(s.as_bytes(), start, length)).into_owned() } pub fn implode(glue: &str, pieces: &[String]) -> String { -- cgit v1.3.1-4-g156e