Add one name dedupe function and use it for parsed names and metadata.json

server/utils/generators/abmetadataGenerator.js

@@ -1,5 +1,6 @@
 const Logger = require('../../Logger')
 const parseSeriesString = require('../parsers/parseSeriesString')
+const { dedupeNames } = require('../parsers/parseNameString')
 
 const mediaTypeKeys = {
   book: {
@@ -140,8 +141,7 @@ function validateMetadataValue(key, value, expectedType) {
       return undefined
     }
 
-    const cleanedArray = value.filter((t) => typeof t === 'string')
-    return [...new Set(cleanedArray.map((t) => t.trim()).filter((t) => t))]
+    return dedupeNames(value)
   }
 
   Logger.warn(`[abmetadataGenerator] Unknown expected type "${expectedType}" for key "${key}"`)

server/utils/parsers/parseNameString.js

@@ -35,6 +35,28 @@ module.exports.nameToLastFirst = (firstLast) => {
   return `${nameObj.last_name}, ${nameObj.first_name}`
 }
 
+/**
+ * Trim names, drop blanks, and drop exact repeats. The first spelling is kept.
+ * Callers that build bookAuthors rows should use this instead of deduping per metadata source.
+ *
+ * @param {string[]} names
+ * @returns {string[]}
+ */
+function dedupeNames(names) {
+  if (!Array.isArray(names)) return []
+  const seen = new Set()
+  const deduped = []
+  for (const name of names) {
+    if (typeof name !== 'string') continue
+    const trimmed = name.trim()
+    if (!trimmed || seen.has(trimmed)) continue
+    seen.add(trimmed)
+    deduped.push(trimmed)
+  }
+  return deduped
+}
+module.exports.dedupeNames = dedupeNames
+
 /**
  * Parses a name string into an array of names
  *
@@ -62,7 +84,7 @@ module.exports.parse = (nameString) => {
   // If names are in Chinese，Japanese and Korean languages, return as is.
   if (/[\u4e00-\u9fff\u3040-\u30ff\u31f0-\u31ff]/.test(splitNames[0])) {
     return {
-      names: splitNames
+      names: dedupeNames(splitNames)
     }
   }
 
@@ -102,7 +124,7 @@ module.exports.parse = (nameString) => {
   names = names.filter((n) => n.first_name || n.last_name)
 
   // Set name strings and remove duplicates
-  const namesArray = [...new Set(names.map((a) => (a.first_name ? `${a.first_name} ${a.last_name}` : a.last_name)))]
+  const namesArray = dedupeNames(names.map((a) => (a.first_name ? `${a.first_name} ${a.last_name}` : a.last_name)))
 
   return {
     names: namesArray // Array of first last

test/server/utils/parsers/parseNameString.test.js

@@ -1,6 +1,6 @@
 const chai = require('chai')
 const expect = chai.expect
-const { parse, nameToLastFirst } = require('../../../../server/utils/parsers/parseNameString')
+const { parse, nameToLastFirst, dedupeNames } = require('../../../../server/utils/parsers/parseNameString')
 
 describe('parseNameString', () => {
   describe('parse', () => {
@@ -80,6 +80,20 @@ describe('parseNameString', () => {
     })
   })
 
+  describe('dedupeNames', () => {
+    it('trims, drops blanks, and keeps the first spelling of an exact repeat', () => {
+      expect(dedupeNames(['  Stephen King ', 'Stephen King', '', '   ', 'stephen king'])).to.deep.equal(['Stephen King', 'stephen king'])
+    })
+
+    it('preserves order and ignores non-strings', () => {
+      expect(dedupeNames(['Amy Yao', null, 'Amy Yao', 3, 'Raymond Chandler'])).to.deep.equal(['Amy Yao', 'Raymond Chandler'])
+    })
+
+    it('returns an empty array for non-arrays', () => {
+      expect(dedupeNames(null)).to.deep.equal([])
+    })
+  })
+
   describe('nameToLastFirst', () => {
     it('converts First Last to Last, First format', () => {
       const result = nameToLastFirst('John Smith')