Handle pre-epoch file timestamps consistently and record Rust audit

This commit is contained in:
karl committed 2026-10-13 17:02:16 +02:00
1 parent 59b2fc39b2
commit 14adbbc616
7 files changed
+262 -46

No files matched your search

+3 -10
View File
@@ -21,7 +21,6 @@ use std::{
fs,
path::{Path, PathBuf},
sync::atomic::{AtomicBool, Ordering},
time::UNIX_EPOCH,
};
use walkdir::WalkDir;
@@ -657,12 +656,7 @@ impl SearchIndex {
report.discovered += 1;
let metadata = entry.metadata()?;
let size = i64::try_from(metadata.len()).context("file exceeds supported size")?;
let modified = metadata
.modified()?
.duration_since(UNIX_EPOCH)
.context("file timestamp predates Unix epoch")?;
let mtime = i64::try_from(modified.as_nanos())
.context("file timestamp exceeds supported range")?;
let mtime = searchi_media::unix_time_ns(metadata.modified()?)?;
let extension = entry
.path()
.extension()
@@ -1257,7 +1251,7 @@ impl SearchIndex {
pub fn file_signature(&self, path: &str) -> Result<Option<(u64, i64)>> {
self.connection
.query_row(
"SELECT size,mtime FROM files WHERE path=?1 AND size>=0 AND mtime>=0",
"SELECT size,mtime FROM files WHERE path=?1 AND size>=0",
[path],
|row| Ok((row.get(0)?, row.get(1)?)),
)
@@ -1294,9 +1288,8 @@ impl SearchIndex {
Ok(value) if value.is_file() => value,
_ => return Ok(None),
};
let mtime = metadata.modified()?.duration_since(UNIX_EPOCH)?;
let size = i64::try_from(metadata.len())?;
let modified = i64::try_from(mtime.as_nanos())?;
let modified = searchi_media::unix_time_ns(metadata.modified()?)?;
let Some(expected_hash) = self.extracted_source_hash(path)? else {
return Ok(None);
};
@@ -90,3 +90,40 @@ fn oversized_text_and_unsupported_files_remain_filename_only() -> Result<()> {
assert_eq!(index.search("large.txt", 10)?.len(), 1);
Ok(())
}
#[test]
fn pre_epoch_source_remains_searchable_and_verifiable() -> Result<()> {
let directory = tempfile::tempdir()?;
let path = directory.path().join("archived-notes.txt");
let content = "Historical supplier cancellation notice\n";
fs::write(&path, content)?;
fs::File::options().write(true).open(&path)?.set_times(
FileTimes::new().set_modified(std::time::UNIX_EPOCH - std::time::Duration::from_secs(1)),
)?;
let mut index = SearchIndex::open(":memory:")?;
let report = index.index_folder(directory.path())?;
assert_eq!(report.indexed, 1);
assert_eq!(report.failures, 0);
assert_eq!(index.search("archived-notes.txt", 10)?.len(), 1);
assert_eq!(index.search("cancellation", 10)?.len(), 1);
let passage = index.passage_batch(0, 1)?.remove(0);
assert_eq!(
index.file_signature(&passage.path)?,
Some((content.len() as u64, -1_000_000_000))
);
assert_eq!(
index.extracted_source_hash(&passage.path)?,
Some(format!("{:x}", Sha256::digest(content.as_bytes())))
);
assert_eq!(
index
.source_passage(&passage.path, &passage.source, 8192)?
.unwrap()
.text,
content
);
let rescan = index.index_folder(directory.path())?;
assert_eq!(rescan.unchanged, 1);
assert_eq!(rescan.indexed, 0);
Ok(())
}