mirror of
https://github.com/storytold/lightcraft.git
synced 2026-10-29 20:48:33 -06:00
PERF-SYNC: a Synchronize Folder scan reads off the folder's listing whether a photo and its sidecar are there
The scan asked the disk about every library photo in the folder: is its file there, and has it a sidecar under any of four names. On a network share each question is a round trip, and the scan had just listed the folder. Measured on a hard disk over SMB, for unchanged folders of 1,600 to 2,400 photos: 4.6 to 7.1 s of questions after a listing of 4 to 5 s. - The walk hands each folder's listing over (`walk::files_in_noting`: every entry's name and kind, hidden ones too, and whether that is all of them). The scan walks the folder once, then checks the photos beside the probing of new files. - `sync::Listing` answers "is it there" / "is it a file" where that holds on any file system: a name the folder lists is there; a plain ASCII name is not there when the folder was listed to its end, all its names are ASCII and none is that name in another case. Everything else is asked of the disk as before (a link, a name Windows reads another way, a folder that was not walked, and every name not listed in a folder that holds a name outside ASCII). Now under 0.05 s for those folders. Two things differ on purpose: a file the folder lists is there though it can't be opened by its path (a folder that may be listed but not searched, an over-long path), where such photos were listed as missing; and the answers are the folder as the walk saw it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Nathaniel Ledford
co-authored by
Claude Opus 5.5
Claude Opus 5.5
parent
a4b53b28cc
commit
540a281ccd
@@ -0,0 +1,191 @@
|
||||
//! Synchronize Folder answers "is the photo's file there" and "has it a sidecar" from the
|
||||
//! listing of the folder its scan has just made, instead of asking the disk about every photo.
|
||||
//!
|
||||
//! On a network share each question is a round trip (about 5 ms a file on a hard disk over SMB,
|
||||
//! measured), and a photo without a sidecar took five of them; the listing has every name.
|
||||
//! The listing answers only what it can be sure of on any file system: where a name could match
|
||||
//! another way (case, letters outside ASCII, a link), the disk is asked as before.
|
||||
use super::*;
|
||||
use crate::walk::EntryKind::{File, Folder, Other};
|
||||
|
||||
fn listing(folders: &[(&str, &[(&str, crate::walk::EntryKind)], bool)]) -> Listing {
|
||||
let mut l = Listing::default();
|
||||
for (dir, names, complete) in folders {
|
||||
l.note(Path::new(dir), names.iter().map(|(n, k)| (std::ffi::OsString::from(n), *k)).collect(), *complete);
|
||||
}
|
||||
l
|
||||
}
|
||||
|
||||
/// Given the listing of a folder, when it is asked about a name, then it says "there" for a
|
||||
/// name it lists, "not there" for one nothing in the folder could be taken for, and nothing
|
||||
/// (ask the disk) wherever a file system could see it otherwise.
|
||||
#[test]
|
||||
fn the_listing_answers_only_what_it_is_sure_of() {
|
||||
let l = listing(&[
|
||||
(
|
||||
"/pics/trip",
|
||||
&[("IMG_1.CR3", File), ("IMG_1.xmp", File), ("img_2.jpg", File), ("sub", Folder), ("link.jpg", Other), (".hidden.xmp", File)],
|
||||
true,
|
||||
),
|
||||
("/pics/cut", &[("IMG_1.CR3", File)], false),
|
||||
("/pics/names", &[("IMG_1.CR3", File), ("Größe.JPG", File)], true),
|
||||
// (the capital sharp s, which a file system that folds case takes for `SS`)
|
||||
("/pics/sharp", &[("IMG_1.CR3", File), ("STRA\u{1e9e}E.jpg", File)], true),
|
||||
("/pics/devices", &[("aux.jpg", File), ("COM\u{b9}.jpg", File), ("IMG_1.CR3", File)], true),
|
||||
]);
|
||||
let (there, gone, ask) = (Some(true), Some(false), None);
|
||||
for (path, exists, is_file) in [
|
||||
("/pics/trip/IMG_1.CR3", there, there),
|
||||
("/pics/trip/IMG_1.xmp", there, there),
|
||||
("/pics/trip/.hidden.xmp", there, there),
|
||||
("/pics/trip/sub", there, gone),
|
||||
("/pics/trip/IMG_9.CR3", gone, gone),
|
||||
("/pics/trip/IMG_1.XMP", ask, ask),
|
||||
("/pics/trip/IMG_2.JPG", ask, ask),
|
||||
("/pics/trip/link.jpg", ask, ask),
|
||||
("/pics/trip/SUB", ask, ask),
|
||||
// a folder the walk did not list to its end, or not at all
|
||||
("/pics/cut/IMG_1.CR3", there, there),
|
||||
("/pics/cut/IMG_2.CR3", ask, ask),
|
||||
("/pics/other/IMG_1.CR3", ask, ask),
|
||||
// a folder that holds a name outside ASCII: what it lists is there, and nothing is surely not there (file
|
||||
// systems hold such names equal in more ways than one)
|
||||
("/pics/names/IMG_1.CR3", there, there),
|
||||
("/pics/names/Größe.JPG", there, there),
|
||||
("/pics/names/IMG_2.CR3", ask, ask),
|
||||
("/pics/names/größe.JPG", ask, ask),
|
||||
("/pics/trip/Größe.JPG", ask, ask),
|
||||
("/pics/sharp/strasse.jpg", ask, ask),
|
||||
("/pics/sharp/STRASSE.xmp", ask, ask),
|
||||
("/pics/sharp/IMG_2.CR3", ask, ask),
|
||||
// names Windows resolves another way: a short name, a stream, dots or spaces at the end, a device
|
||||
("/pics/trip/IMG_10~1.CR3", ask, ask),
|
||||
("/pics/trip/IMG_1.CR3::$DATA", ask, ask),
|
||||
("/pics/trip/IMG_1.CR3.", ask, ask),
|
||||
("/pics/trip/IMG_1.CR3 ", ask, ask),
|
||||
("/pics/devices/aux.jpg", ask, ask),
|
||||
("/pics/devices/NUL", ask, ask),
|
||||
("/pics/devices/nul.xmp", ask, ask),
|
||||
("/pics/devices/COM1 .jpg", ask, ask),
|
||||
("/pics/devices/COM\u{b9}.jpg", ask, ask),
|
||||
("/pics/trip/con", ask, ask),
|
||||
("/pics/trip/PRN.txt", ask, ask),
|
||||
("/pics/trip/com2.jpg", ask, ask),
|
||||
("/pics/trip/LPT9.xmp", ask, ask),
|
||||
("/pics/trip/lpt0", ask, ask),
|
||||
("/pics/trip/nul.tar.gz", ask, ask),
|
||||
("/pics/trip/CONIN$", ask, ask),
|
||||
("/pics/trip/com10.jpg", gone, gone),
|
||||
("/pics/trip/computer.jpg", gone, gone),
|
||||
("/pics/devices/IMG_1.CR3", there, there),
|
||||
("/pics/devices/auxiliary.jpg", ask, ask),
|
||||
("/pics/trip/auxiliary.jpg", gone, gone),
|
||||
// a path that does not end in its file's name is the disk's to resolve
|
||||
("/pics/trip/IMG_1.CR3/", ask, ask),
|
||||
("/pics/trip/IMG_1.CR3/.", ask, ask),
|
||||
("/pics/trip/", ask, ask),
|
||||
("", ask, ask),
|
||||
] {
|
||||
assert_eq!((l.exists(Path::new(path)), l.is_file(Path::new(path))), (exists, is_file), "{path:?}");
|
||||
}
|
||||
// a name that is no text: nothing in its folder is surely not there
|
||||
#[cfg(unix)]
|
||||
{
|
||||
use std::os::unix::ffi::OsStringExt;
|
||||
let mut l = Listing::default();
|
||||
l.note(
|
||||
Path::new("/pics/raw"),
|
||||
vec![(std::ffi::OsString::from("IMG_1.CR3"), File), (std::ffi::OsString::from_vec(vec![0x66, 0xff, 0x6f]), File)],
|
||||
true,
|
||||
);
|
||||
assert_eq!((l.exists(Path::new("/pics/raw/IMG_1.CR3")), l.exists(Path::new("/pics/raw/IMG_2.CR3"))), (there, ask));
|
||||
}
|
||||
}
|
||||
|
||||
fn photo(c: &mut Catalog, path: &Path) -> Arc<Photo> {
|
||||
let id = c.alloc_photo_id();
|
||||
let name = path.file_name().unwrap().to_string_lossy().to_string();
|
||||
c.apply(lightcraft_catalog::Op::AddPhoto {
|
||||
photo: Box::new(Photo::new(id, Source::File { path: path.to_string_lossy().to_string() }, &name, "JPEG", 6, 4, "2020-01-01T00:00:00")),
|
||||
})
|
||||
.unwrap();
|
||||
Arc::clone(c.photo(id).unwrap())
|
||||
}
|
||||
|
||||
/// Given a folder on disk whose photos are there or gone, with sidecars under every name a
|
||||
/// sidecar can have, when each photo is checked with the folder's listing, then the answer is
|
||||
/// the one the disk gives; and for photos with plain names the disk was asked nothing but the
|
||||
/// sidecars' dates.
|
||||
#[test]
|
||||
fn a_check_from_the_listing_is_the_check_from_the_disk() {
|
||||
let dir = std::env::temp_dir().join(format!("lc-listing-{}", std::process::id()));
|
||||
let _ = std::fs::remove_dir_all(&dir);
|
||||
std::fs::create_dir_all(dir.join("sub")).unwrap();
|
||||
let news = r#"<x:xmpmeta xmlns:x="adobe:ns:meta/"><rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"><rdf:Description xmlns:xmp="http://ns.adobe.com/xap/1.0/" xmp:Rating="5"/></rdf:RDF></x:xmpmeta>"#;
|
||||
for name in [
|
||||
"plain.jpg",
|
||||
"stem.jpg",
|
||||
"stem.xmp",
|
||||
"full.jpg",
|
||||
"full.jpg.xmp",
|
||||
"upper.jpg",
|
||||
"upper.XMP",
|
||||
"sub/deep.jpg",
|
||||
"sub/deep.xmp",
|
||||
"Größe.jpg",
|
||||
"Größe.xmp",
|
||||
"dir.jpg.xmp",
|
||||
] {
|
||||
std::fs::write(dir.join(name), if name.ends_with("mp") || name.ends_with("MP") { news.as_bytes() } else { b"jpeg" }).unwrap();
|
||||
}
|
||||
let mut c = Catalog::new();
|
||||
let photos: Vec<Arc<Photo>> =
|
||||
["plain.jpg", "stem.jpg", "full.jpg", "upper.jpg", "sub/deep.jpg", "Größe.jpg", "gone.jpg", "sub/gone.jpg", "dir.jpg"]
|
||||
.iter()
|
||||
.map(|n| photo(&mut c, &dir.join(n)))
|
||||
.collect();
|
||||
// (`dir.jpg` is gone, though a sidecar of its name is there)
|
||||
let mut l = Listing::default();
|
||||
let walked =
|
||||
crate::walk::files_in_noting(&dir, None, crate::walk::Limits::default(), |_| true, &mut |d, names, complete| l.note(d, names, complete));
|
||||
assert_eq!(walked.files.len(), 12);
|
||||
let nothing = Listing::default();
|
||||
for naming in [SidecarNaming::Stem, SidecarNaming::Full] {
|
||||
for p in &photos {
|
||||
let from_disk = check_photo(p, naming, &c, ¬hing);
|
||||
assert_eq!(check_photo(p, naming, &c, &l), from_disk, "{:?} {naming:?}", p.file_name);
|
||||
}
|
||||
}
|
||||
let expected = [Check::Nothing, Check::News, Check::News, Check::News, Check::News, Check::News, Check::Missing, Check::Missing, Check::Missing];
|
||||
assert_eq!(photos.iter().map(|p| check_photo(p, SidecarNaming::Stem, &c, &l)).collect::<Vec<_>>(), expected);
|
||||
|
||||
// what the disk was asked about the same photos, with the listing and without: fewer questions with it
|
||||
let asked = |l: &Listing| {
|
||||
for p in &photos {
|
||||
check_photo(p, SidecarNaming::Stem, &c, l);
|
||||
}
|
||||
l.asked.load(std::sync::atomic::Ordering::Relaxed)
|
||||
};
|
||||
let mut with = Listing::default();
|
||||
crate::walk::files_in_noting(&dir, None, crate::walk::Limits::default(), |_| true, &mut |d, names, complete| with.note(d, names, complete));
|
||||
let (with, without) = (asked(&with), asked(&Listing::default()));
|
||||
// (without: each photo's file, and the names of its sidecar up to the one that is there: 20 questions where
|
||||
// names are found in any case, 21 where they are not)
|
||||
assert!(without >= 20, "{without}");
|
||||
// (with: this folder holds `Größe.*`, so what it does not list is still asked about: the four names of the
|
||||
// sidecar `plain.jpg` has not, the two `full.jpg` has not, `upper.xmp`, and the two photos that are gone. What
|
||||
// it lists is not asked about, nor anything in `sub`, whose names are all ASCII)
|
||||
assert_eq!(with, 9, "asked {with} times with the listing, {without} without");
|
||||
|
||||
// a folder of plain names only: the disk is asked nothing at all
|
||||
let sub = dir.join("sub");
|
||||
let mut only_plain = Listing::default();
|
||||
crate::walk::files_in_noting(&sub, None, crate::walk::Limits::default(), |_| true, &mut |d, names, complete| only_plain.note(d, names, complete));
|
||||
let mut c2 = Catalog::new();
|
||||
for (name, expected) in [("deep.jpg", Check::News), ("gone.jpg", Check::Missing), ("gone.jpg.xmp", Check::Missing)] {
|
||||
let p = photo(&mut c2, &sub.join(name));
|
||||
assert_eq!(check_photo(&p, SidecarNaming::Stem, &c2, &only_plain), expected, "{name}");
|
||||
}
|
||||
assert_eq!(only_plain.asked.load(std::sync::atomic::Ordering::Relaxed), 0, "every answer came from the listing");
|
||||
let _ = std::fs::remove_dir_all(&dir);
|
||||
}
|
||||
Reference in New Issue
Block a user