#176 - Option to index CamelCaseWords

src/search/omnisearch.ts

@@ -26,7 +26,11 @@ const tokenize = (text: string): string[] => {
     return tokens.flatMap(word =>
       chsRegex.test(word) ? chsSegmenter.cut(word) : [word]
     )
-  } else return tokens.flatMap(splitCamelCase)
+  } else {
+    if (settings.splitCamelCase)
+      return [...tokens, ...tokens.flatMap(splitCamelCase)]
+    return tokens
+  }
 }
 
 export class Omnisearch {

src/settings.ts

@@ -46,6 +46,7 @@ export interface OmnisearchSettings extends WeightingSettings {
   /** If a query returns 0 result, try again with more relax conditions */
   simpleSearch: boolean
   highlight: boolean
+  splitCamelCase: boolean
 }
 
 /**
@@ -204,6 +205,25 @@ export class SettingsTab extends PluginSettingTab {
         })
       )
 
+    // Split CamelCaseWords
+    const camelCaseDesc = new DocumentFragment()
+    camelCaseDesc.createSpan({}, span => {
+      span.innerHTML = `Enable this if you want to be able to search for CamelCaseWords as separate words.<br/>        
+        ⚠️ <span style="color: var(--text-accent)">Changing this setting will clear the cache.</span><br>
+        <strong style="color: var(--text-accent)">Needs a restart to fully take effect.</strong>
+        `
+    })
+    new Setting(containerEl)
+      .setName('Split CamelCaseWords')
+      .setDesc(camelCaseDesc)
+      .addToggle(toggle =>
+        toggle.setValue(settings.splitCamelCase).onChange(async v => {
+          await database.clearCache()
+          settings.splitCamelCase = v
+          await saveSettings(this.plugin)
+        })
+      )
+
     // Simpler search
     new Setting(containerEl)
       .setName('Simpler search')
@@ -379,6 +399,7 @@ export const DEFAULT_SETTINGS: OmnisearchSettings = {
   indexedFileTypes: [] as string[],
   PDFIndexing: false,
   imagesIndexing: false,
+  splitCamelCase: false,
 
   ribbonIcon: true,
   showExcerpt: true,

src/tools/utils.ts

@@ -91,14 +91,17 @@ export function getAllIndices(text: string, regex: RegExp): SearchMatch[] {
  */
 export function stringsToRegex(strings: string[]): RegExp {
   if (!strings.length) return /^$/g
-  // Default word split is not applied if the user uses the cm-chs-patch plugin
   const joined =
     '(' +
-    (getChsSegmenter() ? '' : `^|${SPACE_OR_PUNCTUATION.source}`) +
+    // Default word split is not applied if the user uses the cm-chs-patch plugin
+    (getChsSegmenter()
+      ? ''
+      : // Split on start of line, spaces, punctuation, or capital letters (for camelCase)
+      settings.splitCamelCase
+      ? `^|${SPACE_OR_PUNCTUATION.source}|[A-Z]`
+      : `^|${SPACE_OR_PUNCTUATION.source}`) +
     ')' +
-    '(' +
+    `(${strings.map(s => escapeRegex(s)).join('|')})`
-    strings.map(s => escapeRegex(s)).join('|') +
-    ')'
 
   const reg = new RegExp(`${joined}`, 'giu')
   return reg