font-store: index PostScript names shared by many files in linear time (#538)

Index::assemble searched every earlier position listed under a PostScript
name before adding the next, so an index that gives one name to many files
took time quadratic in their number. Files are visited in order, so a file
that lists a name twice can only be the last position pushed: comparing
with the end is enough.

find_postscript compared each hit's family name with every name collected
so far; it now keeps the lowercased names in a set.

crates/font-store/src/index.rs

@@ -1,7 +1,7 @@
 //! Font indexes (aiongg/craft-font-index, format 1.x): the families of every source in one
 //! [`Index`], and exact lookup.
 
-use std::collections::{BTreeMap, HashMap};
+use std::collections::{BTreeMap, HashMap, HashSet};
 
 use serde::Deserialize;
 
@@ -519,8 +519,11 @@ impl Index {
             by_name.entry(fam.family.to_lowercase()).or_default().push(fi);
             for (xi, file) in fam.files.iter().enumerate() {
                 for ps in &file.postscript {
+                    // Files are visited in order, so a file that lists a name twice can only
+                    // have been the last one pushed: checking the end keeps this linear
+                    // however many files share the name.
                     let hits = by_postscript.entry(ps.clone()).or_default();
-                    if !hits.contains(&(fi, xi)) {
+                    if hits.last() != Some(&(fi, xi)) {
                         hits.push((fi, xi));
                     }
                 }
@@ -610,9 +613,10 @@ impl Index {
     pub fn find_postscript(&self, name: &str) -> Result<Face<'_>, Unavailable> {
         let hits = self.by_postscript.get(name).map(Vec::as_slice).unwrap_or_default();
         let mut names: Vec<&str> = Vec::new();
+        let mut seen: HashSet<String> = HashSet::new();
         for &(fi, _) in hits {
             if let Some(f) = self.families.get(fi)
-                && !names.iter().any(|n| n.to_lowercase() == f.family.to_lowercase())
+                && seen.insert(f.family.to_lowercase())
             {
                 names.push(&f.family);
             }

crates/font-store/src/tests.rs

@@ -206,6 +206,25 @@ fn postscript_lookups_match_exactly() {
     assert!(index.family("Shared Two").unwrap().face_by_postscript("Shared-Regular").is_some());
 }
 
+#[test]
+fn many_files_sharing_a_postscript_name_index_in_linear_time() {
+    let mut v = fixture().index;
+    let bytes = font("Many", "Regular", 400);
+    for f in 0..40 {
+        let files: Vec<serde_json::Value> = (0..2500)
+            .map(|i| file_entry(&format!("ofl/many{f}/Many-{i}.ttf"), &bytes, "normal", 400, json!([]), &["Many-Regular", "Many-Regular"]))
+            .collect();
+        let family = json!({"family": format!("Many {f}"), "license": "OFL", "license_path": null, "files": files});
+        v["families"].as_array_mut().unwrap().push(family);
+    }
+    let index = index_of(&v).unwrap();
+    let Err(Unavailable::Ambiguous { families, .. }) = index.find_postscript("Many-Regular") else { panic!("not ambiguous") };
+    assert_eq!(families.len(), 40);
+    assert_eq!(families[0], "Many 0");
+    let first = index.family("Many 7").unwrap();
+    assert_eq!(first.face_by_postscript("Many-Regular").unwrap().file_name(), "Many-0.ttf");
+}
+
 #[test]
 fn styles_parse_and_urls_pin_the_commit() {
     let index = index_of(&fixture().index).unwrap();