Fix name parser to not use "last, first" format when not using comma separators. Adds unit tests #3940

server/utils/parsers/parseNameString.js

@@ -35,11 +35,18 @@ module.exports.nameToLastFirst = (firstLast) => {
   return `${nameObj.last_name}, ${nameObj.first_name}`
 }
 
-// Handle any name string
+/**
+ * Parses a name string into an array of names
+ *
+ * @param {string} nameString - The name string to parse
+ * @returns {{ names: string[] }} Array of names
+ */
 module.exports.parse = (nameString) => {
   if (!nameString) return null
 
-  var splitNames = []
+  let splitNames = []
+  const isCommaSeparated = nameString.includes(',')
+
   // Example &LF: Friedman, Milton & Friedman, Rose
   if (nameString.includes('&')) {
     nameString.split('&').forEach((asa) => (splitNames = splitNames.concat(asa.split(','))))
@@ -59,17 +66,18 @@ module.exports.parse = (nameString) => {
     }
   }
 
-  var names = []
+  let names = []
 
   // 1 name FIRST LAST
   if (splitNames.length === 1) {
     names.push(parseName(nameString))
   } else {
-    var firstChunkIsALastName = checkIsALastName(splitNames[0])
-    var isEvenNum = splitNames.length % 2 === 0
+    // Determines whether this is formatted as last, first or first last (only if using comma separator)
+    // Example: "Smith; James Jones" -> ["Smith", "James Jones"]
+    let firstChunkIsALastName = !isCommaSeparated ? false : checkIsALastName(splitNames[0])
+    let isEvenNum = splitNames.length % 2 === 0
 
     if (!isEvenNum && firstChunkIsALastName) {
-      // console.error('Multi-name LAST,FIRST entry has a straggler (could be roman numerals or a suffix), ignore it')
       splitNames = splitNames.slice(0, splitNames.length - 1)
     }
 

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

@@ -0,0 +1,99 @@
+const chai = require('chai')
+const expect = chai.expect
+const { parse, nameToLastFirst } = require('../../../../server/utils/parsers/parseNameString')
+
+describe('parseNameString', () => {
+  describe('parse', () => {
+    it('returns null if nameString is empty', () => {
+      const result = parse('')
+      expect(result).to.be.null
+    })
+
+    it('parses single name in First Last format', () => {
+      const result = parse('John Smith')
+      expect(result.names).to.deep.equal(['John Smith'])
+    })
+
+    it('parses single name in Last, First format', () => {
+      const result = parse('Smith, John')
+      expect(result.names).to.deep.equal(['John Smith'])
+    })
+
+    it('parses multiple names separated by &', () => {
+      const result = parse('John Smith & Jane Doe')
+      expect(result.names).to.deep.equal(['John Smith', 'Jane Doe'])
+    })
+
+    it('parses multiple names separated by "and"', () => {
+      const result = parse('John Smith and Jane Doe')
+      expect(result.names).to.deep.equal(['John Smith', 'Jane Doe'])
+    })
+
+    it('parses multiple names separated by comma and "and"', () => {
+      const result = parse('John Smith, Jane Doe and John Doe')
+      expect(result.names).to.deep.equal(['John Smith', 'Jane Doe', 'John Doe'])
+    })
+
+    it('parses multiple names separated by semicolon', () => {
+      const result = parse('John Smith; Jane Doe')
+      expect(result.names).to.deep.equal(['John Smith', 'Jane Doe'])
+    })
+
+    it('parses multiple names in Last, First format', () => {
+      const result = parse('Smith, John, Doe, Jane')
+      expect(result.names).to.deep.equal(['John Smith', 'Jane Doe'])
+    })
+
+    it('parses multiple names with single word name', () => {
+      const result = parse('John Smith, Jones, James Doe, Ludwig von Mises')
+      expect(result.names).to.deep.equal(['John Smith', 'Jones', 'James Doe', 'Ludwig von Mises'])
+    })
+
+    it('parses multiple names with single word name listed first (semicolon separator)', () => {
+      const result = parse('Jones; John Smith; James Doe; Ludwig von Mises')
+      expect(result.names).to.deep.equal(['Jones', 'John Smith', 'James Doe', 'Ludwig von Mises'])
+    })
+
+    it('handles names with suffixes', () => {
+      const result = parse('Smith, John Jr.')
+      expect(result.names).to.deep.equal(['John Jr. Smith'])
+    })
+
+    it('handles compound last names', () => {
+      const result = parse('von Mises, Ludwig')
+      expect(result.names).to.deep.equal(['Ludwig von Mises'])
+    })
+
+    it('handles Chinese/Japanese/Korean names', () => {
+      const result = parse('张三, 李四')
+      expect(result.names).to.deep.equal(['张三', '李四'])
+    })
+
+    it('removes duplicate names', () => {
+      const result = parse('John Smith & John Smith')
+      expect(result.names).to.deep.equal(['John Smith'])
+    })
+
+    it('filters out empty names', () => {
+      const result = parse('John Smith,')
+      expect(result.names).to.deep.equal(['John Smith'])
+    })
+  })
+
+  describe('nameToLastFirst', () => {
+    it('converts First Last to Last, First format', () => {
+      const result = nameToLastFirst('John Smith')
+      expect(result).to.equal('Smith, John')
+    })
+
+    it('returns last name only when no first name', () => {
+      const result = nameToLastFirst('Smith')
+      expect(result).to.equal('Smith')
+    })
+
+    it('handles names with middle names', () => {
+      const result = nameToLastFirst('John Middle Smith')
+      expect(result).to.equal('Smith, John Middle')
+    })
+  })
+})