0.2.2 — a subtitle begins with a capital

title_case counted small words from the start of the whole string, so
every subtitle that opened with one was set in lower case: 'Aesthetic in
Kant: a Critique', 'Born Fighting: the Scots-Irish'. A colon starts a
title of its own and the first word of one is capitalised however small
it is.

Found by book-pipeline, which had been proposing a title page's capitals
verbatim into a catalogue and needed this to stop.

Cargo.lock

@@ -101,7 +101,7 @@ checksum = "f2032f911046de80f0a198e0901378627c33f59ea0ac00e363d481118bd70a53"
 
 [[package]]
 name = "avplan"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -126,7 +126,7 @@ checksum = "ac07cdecf99051d9a5238b80f35af32cdeba5b336e55d957b318b50137e18da5"
 
 [[package]]
 name = "bibclean"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -143,7 +143,7 @@ dependencies = [
 
 [[package]]
 name = "bibfetch"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -196,7 +196,7 @@ dependencies = [
 
 [[package]]
 name = "bookmeta"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -210,7 +210,7 @@ dependencies = [
 
 [[package]]
 name = "booksniff"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -250,7 +250,7 @@ checksum = "fc652a48c352aef3ea3aed32080501cf3ef6ed5da78602a020c991775b0aff04"
 
 [[package]]
 name = "catena"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "bibclean",
  "bibfetch",
@@ -277,7 +277,7 @@ dependencies = [
 
 [[package]]
 name = "catena-cli"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "clap",
  "clap_mangen",
@@ -312,7 +312,7 @@ checksum = "4e7648175b45a9a48536d676f68d918270699102aa8dab5496df06904c914600"
 
 [[package]]
 name = "chunkmd"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -334,7 +334,7 @@ dependencies = [
 
 [[package]]
 name = "citefind"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "chunkmd",
@@ -399,7 +399,7 @@ dependencies = [
 
 [[package]]
 name = "colophon"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "bibclean",
  "catena-cli",
@@ -448,7 +448,7 @@ dependencies = [
 
 [[package]]
 name = "corpus-core"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "schemars",
  "serde",
@@ -492,7 +492,7 @@ dependencies = [
 
 [[package]]
 name = "cuekit"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "corpus-core",
  "regex",
@@ -504,7 +504,7 @@ dependencies = [
 
 [[package]]
 name = "cuenorm"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -514,7 +514,7 @@ dependencies = [
 
 [[package]]
 name = "cuepack"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -525,7 +525,7 @@ dependencies = [
 
 [[package]]
 name = "cueseg"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -612,7 +612,7 @@ dependencies = [
 
 [[package]]
 name = "emb"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -876,7 +876,7 @@ dependencies = [
 
 [[package]]
 name = "idnorm"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "bibclean",
  "catena-cli",
@@ -942,7 +942,7 @@ dependencies = [
 
 [[package]]
 name = "langid"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1022,7 +1022,7 @@ dependencies = [
 
 [[package]]
 name = "mdscrub"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1034,7 +1034,7 @@ dependencies = [
 
 [[package]]
 name = "mdsplit"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1113,7 +1113,7 @@ checksum = "384b8ab6d37215f3c5301a95a4accb5d64aa607f1fcb26a11b5303878451b4fe"
 
 [[package]]
 name = "pdfgutter"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1126,7 +1126,7 @@ dependencies = [
 
 [[package]]
 name = "pdftriage"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1370,7 +1370,7 @@ checksum = "323c417e1d9665a65b263ec744ba09030cfb277e9daa0b018a4ab62e57bc8189"
 
 [[package]]
 name = "rrf"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1517,7 +1517,7 @@ dependencies = [
 
 [[package]]
 name = "seriesof"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",
@@ -1634,7 +1634,7 @@ checksum = "adb6935a6f5c20170eeceb1a3835a49e12e19d792f6dd344ccc76a985ca5a6ca"
 
 [[package]]
 name = "textkit"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "regex",
  "tiktoken-rs",
@@ -1724,7 +1724,7 @@ checksum = "fd3ca314f692efd6c868f8408f53fe444634a845f96c028b97d35f6a1f79f0ee"
 
 [[package]]
 name = "tocfind"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "catena-cli",
  "clap",

Cargo.toml

@@ -3,39 +3,39 @@ resolver = "3"
 members = ["crates/*"]
 
 [workspace.package]
-version = "0.2.1"
+version = "0.2.2"
 edition = "2024"
 rust-version = "1.85"
 license = "MPL-2.0"
 repository = "https://git.pumpin.online/funkpower/catena"
 
 [workspace.dependencies]
-catena-cli = { path = "crates/catena-cli", version = "0.2.1" }
+catena-cli = { path = "crates/catena-cli", version = "0.2.2" }
-corpus-core = { path = "crates/corpus-core", version = "0.2.1" }
+corpus-core = { path = "crates/corpus-core", version = "0.2.2" }
-cuekit = { path = "crates/cuekit", version = "0.2.1" }
+cuekit = { path = "crates/cuekit", version = "0.2.2" }
-textkit = { path = "crates/textkit", version = "0.2.1" }
+textkit = { path = "crates/textkit", version = "0.2.2" }
-chunkmd = { path = "crates/chunkmd", version = "0.2.1" }
+chunkmd = { path = "crates/chunkmd", version = "0.2.2" }
-mdscrub = { path = "crates/mdscrub", version = "0.2.1" }
+mdscrub = { path = "crates/mdscrub", version = "0.2.2" }
-mdsplit = { path = "crates/mdsplit", version = "0.2.1" }
+mdsplit = { path = "crates/mdsplit", version = "0.2.2" }
-pdftriage = { path = "crates/pdftriage", version = "0.2.1" }
+pdftriage = { path = "crates/pdftriage", version = "0.2.2" }
-cuenorm = { path = "crates/cuenorm", version = "0.2.1" }
+cuenorm = { path = "crates/cuenorm", version = "0.2.2" }
-cueseg = { path = "crates/cueseg", version = "0.2.1" }
+cueseg = { path = "crates/cueseg", version = "0.2.2" }
-rrf = { path = "crates/rrf", version = "0.2.1" }
+rrf = { path = "crates/rrf", version = "0.2.2" }
-cuepack = { path = "crates/cuepack", version = "0.2.1" }
+cuepack = { path = "crates/cuepack", version = "0.2.2" }
-bibclean = { path = "crates/bibclean", version = "0.2.1" }
+bibclean = { path = "crates/bibclean", version = "0.2.2" }
-avplan = { path = "crates/avplan", version = "0.2.1" }
+avplan = { path = "crates/avplan", version = "0.2.2" }
-pdfgutter = { path = "crates/pdfgutter", version = "0.2.1" }
+pdfgutter = { path = "crates/pdfgutter", version = "0.2.2" }
-catena = { path = "crates/bindings", version = "0.2.1" }
+catena = { path = "crates/bindings", version = "0.2.2" }
-seriesof = { path = "crates/seriesof", version = "0.2.1" }
+seriesof = { path = "crates/seriesof", version = "0.2.2" }
-idnorm = { path = "crates/idnorm", version = "0.2.1" }
+idnorm = { path = "crates/idnorm", version = "0.2.2" }
-bibfetch = { path = "crates/bibfetch", version = "0.2.1" }
+bibfetch = { path = "crates/bibfetch", version = "0.2.2" }
-bookmeta = { path = "crates/bookmeta", version = "0.2.1" }
+bookmeta = { path = "crates/bookmeta", version = "0.2.2" }
-langid = { path = "crates/langid", version = "0.2.1" }
+langid = { path = "crates/langid", version = "0.2.2" }
-tocfind = { path = "crates/tocfind", version = "0.2.1" }
+tocfind = { path = "crates/tocfind", version = "0.2.2" }
-booksniff = { path = "crates/booksniff", version = "0.2.1" }
+booksniff = { path = "crates/booksniff", version = "0.2.2" }
-citefind = { path = "crates/citefind", version = "0.2.1" }
+citefind = { path = "crates/citefind", version = "0.2.2" }
-colophon = { path = "crates/colophon", version = "0.2.1" }
+colophon = { path = "crates/colophon", version = "0.2.2" }
-emb = { path = "crates/emb", version = "0.2.1" }
+emb = { path = "crates/emb", version = "0.2.2" }
 
 clap = { version = "4", features = ["derive"] }
 clap_mangen = "0.3"

crates/bibclean/src/fix.rs

@@ -67,7 +67,19 @@ pub fn title_case(title: &str) -> String {
                 .to_lowercase()
                 .trim_end_matches(['.', ',', ':', ';', '!', '?', ')', ']', '"', '\''])
                 .to_string();
-            if SMALL_WORDS.binary_search(&core.as_str()).is_ok() && at > 0 && at < last {
+            // A small word stays small in the middle of a title, but not
+            // where a title actually starts. A colon starts a subtitle, and
+            // "Aesthetic in Kant: a Critique" is the giveaway that the rule
+            // was only ever counting from the beginning of the whole string.
+            let opens_a_subtitle = at > 0
+                && words[at - 1]
+                    .trim_end_matches(['"', '\'', ')', ']'])
+                    .ends_with([':', ';', '?', '!', '.']);
+            if SMALL_WORDS.binary_search(&core.as_str()).is_ok()
+                && at > 0
+                && at < last
+                && !opens_a_subtitle
+            {
                 return word.to_lowercase();
             }
             let lowered = word.to_lowercase();
@@ -218,3 +230,35 @@ mod tests {
         );
     }
 }
+
+#[cfg(test)]
+mod subtitle_case_tests {
+    use super::title_case;
+
+    #[test]
+    fn a_subtitle_starts_with_a_capital() {
+        // The rule counted small words from the start of the string, so every
+        // subtitle beginning "A", "The" or "Of" was set lower case.
+        assert_eq!(
+            title_case("AESTHETIC IN KANT: A CRITIQUE"),
+            "Aesthetic in Kant: A Critique"
+        );
+        assert_eq!(
+            title_case("BORN FIGHTING: THE SCOTS-IRISH"),
+            "Born Fighting: The Scots-Irish"
+        );
+        assert_eq!(
+            title_case("A HISTORY: AN INTRODUCTION TO IT"),
+            "A History: An Introduction to It"
+        );
+    }
+
+    #[test]
+    fn a_small_word_in_the_middle_is_still_small() {
+        assert_eq!(
+            title_case("THE RISE AND FALL OF THE THIRD REICH"),
+            "The Rise and Fall of the Third Reich"
+        );
+        assert_eq!(title_case("WAR OF THE WORLDS"), "War of the Worlds");
+    }
+}

crates/bibclean/src/lib.rs

@@ -260,9 +260,14 @@ mod tests {
     #[test]
     fn an_underscored_title_is_cleaned_before_it_is_cased() {
         let cleaned = clean(&book("THE FALL_ AND THE RISE_txt", "Arendt, Hannah"));
+        // "And" is capitalised because the colon starts a subtitle, and the
+        // first word of one is capitalised however small it is. This asserted
+        // "and" until the rule was found to be counting small words from the
+        // start of the whole string, which set every subtitle opening "A",
+        // "The" or "An" in lower case.
         assert_eq!(
             cleaned.book.title.as_deref(),
-            Some("The Fall: and the Rise")
+            Some("The Fall: And the Rise")
         );
         assert_eq!(cleaned.changes.len(), 2);
     }