Merge pull request #3179 from pulibrary/contains-title-index

Rewrite contains_title_index in Rust

Author: christinach

lib/bibdata_rs/src/marc/ruby_bindings.rs

@@ -102,7 +102,7 @@ fn solr_fields(ruby: &Ruby, record_string: String) -> Result<RHash, magnus::Erro
         .ok()
         .and_then(|date| date.maybe_to_string());
 
-    let hash = ruby.hash_new_capa(35);
+    let hash = ruby.hash_new_capa(36);
     hash.aset("aat_s", ruby.ary_from_iter(genre::aat_s(&record)))?;
     hash.aset("action_notes_1display", action_notes_1display)?;
     hash.aset("access_restrictions_note_display", access_notes(&record))?;
@@ -128,6 +128,7 @@ fn solr_fields(ruby: &Ruby, record_string: String) -> Result<RHash, magnus::Erro
         ruby.ary_from_iter(cjk::subjects_cjk(&record)),
     )?;
     hash.aset("cjk_title", ruby.ary_from_iter(cjk::cjk_titles(&record)))?;
+    hash.aset("contains_title_index", ruby.ary_from_iter(title::contains_titles_index(&record)))?;
     hash.aset(
         "fast_subject_display",
         ruby.ary_from_iter(subject::fast_subjects(&record)),

lib/bibdata_rs/src/marc/string_normalize.rs

@@ -56,6 +56,13 @@ pub fn upcase_first(string: &str) -> Cow<'_, str> {
     }
 }
 
+pub fn maybe_not_empty<S>(s: S) -> Option<S>
+where
+    S: AsRef<str>,
+{
+    if s.as_ref().is_empty() { None } else { Some(s) }
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;

lib/bibdata_rs/src/marc/title.rs

@@ -1,7 +1,27 @@
-use crate::marc::extract_values::ExtractValues;
+use crate::marc::{
+    extract_values::ExtractValues,
+    string_normalize::maybe_not_empty,
+    trim_punctuation,
+    variable_length_field::{SubfieldIterator, latin_or_non_latin_tag_included_in},
+};
 use itertools::Itertools;
 use marctk::Record;
 
+pub fn contains_titles_index(record: &Record) -> impl Iterator<Item = String> {
+    record.extract_field_values_by(
+        latin_or_non_latin_tag_included_in(&["700", "710", "711"]),
+        |field| {
+            maybe_not_empty(trim_punctuation(
+                &field
+                    .subfields()
+                    .iter()
+                    .subfields_after("t")
+                    .join(" "),
+            ))
+        },
+    )
+}
+
 pub fn latin_script_title(record: &Record) -> Option<String> {
     record
         .extract_field_values_by(
@@ -19,3 +39,18 @@ pub fn latin_script_title(record: &Record) -> Option<String> {
         )
         .next()
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn it_can_find_contains_titles_index() {
+        let record = Record::from_breaker(r#"=700 12$6880-12$aDawwānī, Muḥammad ibn Asʻad, $d 1426 or 1427-1512 or 1513. $t Zawrāʼ.
+=880 12$6700-12$aدواني، محمد بن اسعد, $d 1426 or 1427-1512 or 1513. $t زوراء."#).unwrap();
+        let mut contains_titles = contains_titles_index(&record);
+        assert_eq!(contains_titles.next(), Some(String::from("Zawrāʼ")));
+        assert_eq!(contains_titles.next(), Some(String::from("زوراء")));
+        assert_eq!(contains_titles.next(), None);
+    }
+}

lib/bibdata_rs/src/marc/variable_length_field.rs

@@ -3,6 +3,7 @@ use marctk::{Field, Subfield};
 
 pub trait SubfieldIterator<'a>: Iterator<Item = &'a Subfield> {
     fn content(self) -> impl Iterator<Item = &'a str>;
+    fn join(self, delimiter: &'a str) -> String;
     fn filter_by_code(self, codes: &'a [&str]) -> impl Iterator<Item = &'a Subfield>;
     fn subfields_before(self, stop_before: &'a str) -> impl Iterator<Item = &'a Subfield>;
     fn subfields_after(self, start_at: &'a str) -> impl Iterator<Item = &'a Subfield>;
@@ -14,6 +15,9 @@ where
     fn content(self) -> impl Iterator<Item = &'a str> {
         self.map(|subfield| subfield.content())
     }
+    fn join(self, delimiter: &'a str) -> String {
+        self.content().join(delimiter)
+    }
     fn filter_by_code(self, codes: &'a [&str]) -> impl Iterator<Item = &'a Subfield> {
         self.filter(move |subfield| codes.contains(&subfield.code()))
     }

marc_to_solr/lib/traject_config.rb

@@ -221,8 +221,8 @@
 
 to_field 'content_title_index', extract_marc('505t')
 
-to_field 'contains_title_index' do |record, accumulator|
-  accumulator.replace(everything_after_t(record, '700:710:711'))
+to_field 'contains_title_index' do |_record, accumulator, context|
+  accumulator.replace(context.clipboard[:solr_fields]['contains_title_index'])
 end
 
 to_field 'linked_title_index', extract_marc('765st:767st:770st:772st:773st:774st:775st:776st:777st:780st:785st:786st:787st')