Dylan search function (#648)

* Add: oneword token search based on the search patter we use

* removed: line reference comments

* add: sutta finder variable at the top for one word query

* create: onewordtoken function and TDD - first failed now passes

* trying to get item titles from results array of objects

* add: fallback for oneword search results back to results on no results

* add: findoneword function in completed form with TDD

* add: new normalizetitle function and better results handling

* update: findoneword...() with matching against joinedTitles made from store items with new normalized titles

* add: new test cases to normalizeSuttaTitles

* add: remove the from beginning, better nikaya index handling and update: remove all text after : for scale

* add: several test cases to match cases suggested on PR review, to show my feature works

* Add: extra test cases and new normalizeSuttaTitles() handling

* remove: un-used tokenResults variable

* update: test case names to match test cases properly

* update: test case description for testing parse lal sutra

Author: DylanAustin-TheDreamer

assets/js/search_functions.js

@@ -1,6 +1,27 @@
 // Parameters
 var BMAX = 250; // Max blurb size in characters
 var RMAX = 100; // Max number of results to display
+var joinedTitles = []
+
+function normalizeSuttaTitles (obj) {
+  var joinedTitleDatabase = []
+
+  for (var i in obj){
+    const item = obj[i];
+    if (!item || item.type !== "content" || item.category !== "canon") continue;
+    const title = item.title || "";
+      const titleJoin = title.normalize("NFD").replace(/[\u0300-\u036f]/g, "").replace(/^\s*(?:DN|MN|SN|AN|KN|LAL|DA|MA|SA|EA|SNP|DHP|ITI|THAG|THIG|UD|NIDD|CV|BV|AP|JA|PV|VV|KP|PTS)\s*\d+(?:\.\d+)?\s*[:.-]?\s*/i, "").replace(/\s*[:\-–]\s*.*$/, "").toLowerCase().replace(/[^a-z0-9]/g, "");
+      const removedTheOnJoin = titleJoin.replace(/^\s*(?:the)\s*/i, "");
+      if(removedTheOnJoin.includes('sutta') || removedTheOnJoin.includes('sutra') || removedTheOnJoin.includes('gatha')) {
+        joinedTitleDatabase.push({
+          ref: i,
+          title: removedTheOnJoin,
+          matchData: { metadata: {} }
+      });
+    }
+  }
+  return joinedTitleDatabase;
+}
 
 function getPositions(result, field) {
     var positions = [];
@@ -169,10 +190,27 @@ function displaySearchResults(results) {
     }
 }
 
+function findOneWordSuttaTitleMatches(query, joinedTitles) {
+  var tokenResults = [];
+  const normalizedQuery = query.normalize("NFD").replace(/[\u0300-\u036f]/g, "").toLowerCase().replace(/[^a-z0-9]/g, "");
+  for (var i in joinedTitles){
+    const item = joinedTitles[i]
+    if(item.title === normalizedQuery){
+      tokenResults.push({
+        ref: item.ref,
+        score: 1,
+        matchData: { metadata: {} }
+      });
+    }
+  }
+  return tokenResults;
+}
+
 function handleSearchMessage(data, searchFn) {
   var results = [];
   var warning = "";
   var words = data.q.trim().split(" ");
+
   for (var i = 0; i < words.length; i++) {
     const s = words[i].trim();
     if (!s.startsWith("+") && !s.startsWith("-") && s.length > 1 && lunr.stopWordFilter(s)) {
@@ -213,10 +251,11 @@ function handleSearchMessage(data, searchFn) {
       });
     });
   }
+  finalResults = results.length ? results : findOneWordSuttaTitleMatches(data.q.trim(), joinedTitles);
   return {
     "warninghtml": warning,
-    "html": displaySearchResults(results),
-    "count": results ? results.length : 0,
+    "html": displaySearchResults(finalResults),
+    "count": finalResults ? finalResults.length : 0,
     "q": data.q,
     "filterquery": data.filterquery,
     "qt": data.qt

assets/js/search_index.js

@@ -69,6 +69,8 @@ var idx = lunr(function () {
     }
 });
 
+joinedTitles = normalizeSuttaTitles(store);
+
 self.onmessage = function(e) {
   self.postMessage(handleSearchMessage(e.data, idx.search.bind(idx)));
 }

assets/js/tests/search-worker.test.js

@@ -69,14 +69,16 @@ vm.runInContext(
   'this.resultMatched = resultMatched;\n' +
   'this.addMatchHighlights = addMatchHighlights;\n' +
   'this.getBlurbForResult = getBlurbForResult;\n' +
+  'this.normalizeSuttaTitles = normalizeSuttaTitles;\n' +
+  'this.findOneWordSuttaTitleMatches = findOneWordSuttaTitleMatches;\n' +
   'this.handleSearchMessage = handleSearchMessage;\n' +
   'this.displaySearchResults = displaySearchResults;\n',
   sandbox
 );
 
 const {
   categoryName, getPositions, resultMatched,
-  addMatchHighlights, getBlurbForResult, handleSearchMessage
+  addMatchHighlights, getBlurbForResult, oneWordToken, normalizeSuttaTitles, findOneWordSuttaTitleMatches, handleSearchMessage
 } = sandbox;
 
 // ── categoryName ────────────────────────────────────────────────────
@@ -296,6 +298,128 @@ describe('getBlurbForResult', () => {
   });
 });
 
+// ── normalizeSuttaTitles ─────────────────────────────────────────────
+describe('normalizeSuttaTitles', () => {
+
+  it('returns an array of database objects with a new normalized title', () => {
+    const mockStore = {
+      id1: {
+        title: 'MN 35 Cūḷa Saccaka Sutta: The Shorter Discourse With Saccaka',
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 1);
+    assert.equal(result[0].ref, 'id1');
+    assert.equal(result[0].title, 'culasaccakasutta');
+  });
+
+  it('remove words after sutta. Also handles sutra', () => {
+    const mockStore = {
+      id1: {
+        title: 'MA 128 Upasaka Sutra: Discourse on the White-Clad Disciple',
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 1);
+    assert.equal(result[0].ref, 'id1');
+    assert.equal(result[0].title, 'upasakasutra');
+  });
+
+  it('can parse a lal sutra', () => {
+    const mockStore = {
+      id1: {
+        title: 'Lal 26 Dharmacakrapravartana Sūtra: The Discourse that Set the Dharma-Wheel Rolling',
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 1);
+    assert.equal(result[0].ref, 'id1');
+    assert.equal(result[0].title, 'dharmacakrapravartanasutra');
+  });
+
+  it('can parse Therigathas', () => {
+    const mockStore = {
+      id1: {
+        title: "Thig 3.8 Somā Therīgāthā: Somā's Verses",
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const mockStore2 = {
+      id1: {
+        title: "Thag 1.7 Bhalliya Theragāthā: Bhalliya's Verse",
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result2 = normalizeSuttaTitles(mockStore2);
+    assert.equal(result2.length, 1);
+    assert.equal(result2[0].ref, 'id1');
+    assert.equal(result2[0].title, 'bhalliyatheragatha');
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 1);
+    assert.equal(result[0].ref, 'id1');
+    assert.equal(result[0].title, 'somatherigatha');
+  });
+
+  it('handles "the" and removes it from a string if it appears at the beginning', () => {
+    const mockStore = {
+      id1: {
+        title: 'DN 22 The Mahāsatipaṭṭhāna Sutta: The Long Discourse about the Ways of Attending to Mindfulness',
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 1);
+    assert.equal(result[0].ref, 'id1');
+    assert.equal(result[0].title, 'mahasatipatthanasutta');
+  });
+
+  it('filters out objects with types that are not equal to content', () => {
+    const mockStore = {
+      id1: {
+        title: 'MN 3 Dhammadāyāda Sutta: Heirs in the Teaching',
+        type: 'content',
+        category: 'av'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 0);
+  });
+
+  it('filters out objects where titles do not include sutta, sutra, or gatha', () => {
+    const mockStore = {
+      id1: {
+        title: 'The Hairy Spider Climbed Up The Waterspout: A day in the life Part 3 ',
+        type: 'content',
+        category: 'canon'
+      }
+    };
+    const result = normalizeSuttaTitles(mockStore);
+    assert.equal(result.length, 0);
+  });
+
+
+});
+
+// ── findOneWordSuttaTitleMatches ─────────────────────────────────────────────
+describe('findOneWordSuttaTitleMatches', () => {
+  it('returns matched item when query matches title exactly', () => {
+    const mockStore = {
+      'id1': { title: 'culasaccakasutta', type: 'content', category: 'canon' }
+    };
+    const result = findOneWordSuttaTitleMatches('culasaccakasutta', mockStore);
+    assert.equal(toLocal(result).length, 1);
+  });
+});
+
 // ── handleSearchMessage ─────────────────────────────────────────────
 
 describe('handleSearchMessage', () => {