QA2: Searchathon fixes

config/production.env

@@ -16,7 +16,6 @@ ENCRYPTED_NYPL_OAUTH_SECRET=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAA
 NYPL_CORE_VERSION=v2.21
 
 LOG_LEVEL=info
-NAME_QUERIES=true
 FEATURES=on-site-edd
 
 SEARCH_ITEMS_SIZE=3

config/qa.env

@@ -1,6 +1,6 @@
 # Greg built self-hosted qa domain:
 ENCRYPTED_ELASTICSEARCH_URI=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAAAJYwgZMGCSqGSIb3DQEHBqCBhTCBggIBADB9BgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDMIkDoQ9C/cCDCAq1wIBEIBQ+L3OgUGeOW9rs1CWkhpBjwM4LbbVRFIWedqew4UXIeSNMJ8cO9SNe4YGCUIoKwCDYt7W7ip3VtDRRRMVvz6QJw+Eg8ugTMVs2pbNFGNvaAQ=
-ENCRYPTED_RESOURCES_INDEX=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAAAHIwcAYJKoZIhvcNAQcGoGMwYQIBADBcBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDEQrZlNp930LqpZVywIBEIAvpxx39QAHfdY1+Lxv54TnDVW3M5cSHiyTbuhaJgDK/3y+5iHyw6Y0MmtGdgnl5jQ=
+ENCRYPTED_RESOURCES_INDEX=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAAAHIwcAYJKoZIhvcNAQcGoGMwYQIBADBcBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDGjIeRRI3cN5LwfcmgIBEIAvkB3K4sxagpqspNa3b3BnJQgbw6Ic0jDhuXzk8gmrCiGNMR90YjTZmyLj9aVUN3M=
 ENCRYPTED_ELASTICSEARCH_API_KEY=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAAAJ4wgZsGCSqGSIb3DQEHBqCBjTCBigIBADCBhAYJKoZIhvcNAQcBMB4GCWCGSAFlAwQBLjARBAx+kryf2KUmGdBYD9sCARCAV3ygz3eXIdq8JX/wpG9JRWlTNMRcpNE1qT0zNlN4t+ZvXEoedLQa/3p1YjgHw06GIAdA9xtkMV4eH9a1K8uCvjP8XxxNKekcMj59TlResnu9QF3r7pGXuQ==
 
 ENCRYPTED_SCSB_URL=AQECAHh7ea2tyZ6phZgT4B9BDKwguhlFtRC6hgt+7HbmeFsrsgAAAH8wfQYJKoZIhvcNAQcGoHAwbgIBADBpBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDBKllElmWYLxGOGopQIBEIA8JJyKde/8m8iCJGKR5D8HoTJhXHeyvw9eIDeuUNKiXLfJwoVz+PDAZSxkCQtM9O91zGhXbe3l6Bk1RlYJ

lib/api-request.js

@@ -2,6 +2,10 @@
 * This class wraps request params to ease interpretting the query
 **/
 
+// Build regex pattern for matching a phrase fully enclosed in quotes (or smart quotes):
+const QUOTE_CHARS = '"\u201C\u201D\u201E\u201F\u2033\u2036'
+const IN_QUOTES_PATTERN = new RegExp(`^[${QUOTE_CHARS}][^${QUOTE_CHARS}]+[${QUOTE_CHARS}]$`)
+
 class ApiRequest {
   static ADVANCED_SEARCH_PARAMS = ['title', 'subject', 'contributor']
   static IDENTIFIER_NUMBER_PARAMS = ['isbn', 'issn', 'lccn', 'oclc']
@@ -43,10 +47,10 @@ class ApiRequest {
   }
 
   /**
-  * Returns true if query is fully wrapped in quotes
+  * Returns true if query is fully wrapped in quotes (or smart quotes)
   **/
   queryIsFullyQuoted () {
-    return /^"[^"]+"$/.test(this.params.q)
+    return IN_QUOTES_PATTERN.test(this.params.q)
   }
 
   /**

lib/elasticsearch/config.js

@@ -62,6 +62,9 @@ const SEARCH_SCOPES = {
   series: {
     fields: ['seriesStatement.folded']
   },
+  journal_title: {
+    fields: null
+  },
   callnumber: {
     fields: null
   },

lib/elasticsearch/elastic-query-builder.js

@@ -22,6 +22,10 @@ class ElasticQueryBuilder {
       case 'title':
         this.buildTitleQuery()
         break
+      case 'journal_title':
+        this.buildTitleQuery()
+        this.requireIssuance('urn:biblevel:s')
+        break
       case 'subject':
         this.buildSubjectQuery()
         break
@@ -36,6 +40,9 @@ class ElasticQueryBuilder {
         this.buildAllQuery()
     }
 
+    // Add user filters:
+    this.applyFilters()
+
     // Apply global clauses:
     // Hide specific nypl-sources when configured to do so:
     this.applyHiddenNyplSources()
@@ -83,9 +90,6 @@ class ElasticQueryBuilder {
       }
     }
 
-    // Add user filters:
-    this.applyFilters()
-
     // Lastly, if any identifier-number params are present (lccn=, isbn=, etc),
     // add those clauses:
     if (this.request.hasIdentifierNumberParam()) {
@@ -187,6 +191,18 @@ class ElasticQueryBuilder {
     this.boostOnItemShelfmarkPrefix()
   }
 
+  /**
+  * Require a specific issuance (e.g. require 'urn:biblevel:s' for journal-
+  * title searches)
+  */
+  requireIssuance (issuance) {
+    this.query.addFilter({
+      term: {
+        'issuance.id': issuance
+      }
+    })
+  }
+
   /**
   * Require exact phrase-match on named fields
   */
@@ -213,8 +229,9 @@ class ElasticQueryBuilder {
           prefixMatch('identifierV2.value', q),
           termMatch('uri', q),
           termMatch('items.idBarcode', q),
-          termMatch('idIsbn_clean', q),
-          prefixMatch('items.shelfMark.raw', q)
+          termMatch('idIsbn.clean', q),
+          termMatch('idIssn.clean', q),
+          prefixMatch('items.shelfMark.keywordLowercased', q)
         ]
       }
     })
@@ -233,17 +250,9 @@ class ElasticQueryBuilder {
         }
       })
     } else if (this.request.params.issn) {
-      this.query.addMust({ term: { idIssn: this.request.params.issn } })
+      this.query.addMust({ term: { 'idIssn.clean': this.request.params.issn } })
     } else if (this.request.params.isbn) {
-      this.query.addMust({
-        bool: {
-          should: [
-            { term: { idIsbn: this.request.params.isbn } },
-            { term: { idIsbn_clean: this.request.params.isbn } }
-          ],
-          minimum_should_match: 1
-        }
-      })
+      this.query.addMust({ term: { 'idIsbn.clean': this.request.params.isbn } })
     } else if (this.request.params.oclc) {
       this.query.addMust({ term: { idOclc: this.request.params.oclc } })
     }
@@ -258,8 +267,8 @@ class ElasticQueryBuilder {
     this.query.addMust({
       bool: {
         should: [
-          prefixMatch('shelfMark.raw', q),
-          prefixMatch('items.shelfMark.raw', q)
+          prefixMatch('shelfMark.keywordLowercased', q),
+          prefixMatch('items.shelfMark.keywordLowercased', q)
         ]
       }
     })
@@ -350,14 +359,18 @@ class ElasticQueryBuilder {
   * Boost bibs with items having exact callnumber match
   **/
   boostOnItemShelfmarkExact (boost = 50) {
-    this.query.addShould(termMatch('items.shelfMark.raw', this.request.querySansQuotes(), 50))
+    this.query.addShould(termMatch('items.shelfMark.keywordLowercased', this.request.querySansQuotes(), boost))
+    // Specially boost case-sensitive match:
+    this.query.addShould(termMatch('items.shelfMark.raw', this.request.querySansQuotes(), boost * 2))
   }
 
   /**
   * Boost bibs with items having prefix callnumber match
   **/
   boostOnItemShelfmarkPrefix (boost = 10) {
-    this.query.addShould(prefixMatch('items.shelfMark.raw', boost))
+    this.query.addShould(prefixMatch('items.shelfMark.keywordLowercased', this.request.querySansQuotes(), boost))
+    // Specially boost case-sensitive match:
+    this.query.addShould(prefixMatch('items.shelfMark.raw', this.request.querySansQuotes(), boost * 2))
   }
 
   /**
@@ -367,20 +380,27 @@ class ElasticQueryBuilder {
     const q = this.request.querySansQuotes()
 
     this.query.addShoulds([
-      prefixMatch('identifierV2.value', q, 10),
-      termMatch('uri', q, 10),
-      prefixMatch('idIsbn', q, 10),
+      prefixMatch('identifierV2.value', q, 5),
+      termMatch('uri', q, 5),
+
+      // Tiered matches on isbn:
+      prefixMatch('idIsbn.clean', q, 10),
+      termMatch('idIsbn.clean', q, 20),
+      termMatch('idIsbn', q, 20),
+
+      // Tiered matches on issn:
+      prefixMatch('idIssn.clean', q, 10),
+      termMatch('idIssn.clean', q, 20),
+      termMatch('idIssn', q, 20),
+
       prefixMatch('idLccn', q, 10),
       prefixMatch('idOclc', q, 10),
-      prefixMatch('shelfMark.raw', q, 10),
+      prefixMatch('shelfMark.keywordLowercased', q, 10),
+      prefixMatch('items.shelfMark.keywordLowercased', q, 10),
+
       // I would like this to be a prefix, but doesn't seem to work:
-      termMatch('items.identifierV2.value', q),
-      prefixMatch('items.shelfMark.raw', q)
+      termMatch('items.identifierV2.value', q)
     ])
-
-    if (/\d{6,}/.test(this.request.querySansQuotes())) {
-      this.query.addShould(prefixMatch('idIsbn_clean', q))
-    }
   }
 
   /**

lib/elasticsearch/utils.js

@@ -94,7 +94,7 @@ const termMatch = (field, value, boost = 1) => {
 * Given a field name and a value, returns a plainobject representing a ES
 * `match_phrase` clause that can be inserted into a ES query
 */
-const phraseMatch = (field, value, boost = 1) => {
+const phraseMatch = (field, value, boost = 0) => {
   return {
     match_phrase: {
       [field]: namedQuery({

lib/jsonld_serializers.js

@@ -107,6 +107,11 @@ class JsonLdItemSerializer extends JsonLdSerializer {
       }
     })
 
+    // If updatedAt (int) found, add updatedAtDate (Date string)
+    if (stmts.updatedAt && /^\d+$/.test(stmts.updatedAt)) {
+      stmts.updatedAtDate = (new Date(stmts.updatedAt)).toISOString()
+    }
+
     return stmts
   }
 
@@ -228,8 +233,6 @@ class ResourceSerializer extends JsonLdItemSerializer {
       })
     }
 
-    stmts.suppressed = this.body.suppressed === true
-
     if (this.body.items) {
       stmts.items = this.body.items
         // Amend items to include source identifier (e.g. urn:SierraNypl:1234, urn:RecapCul:4567)

lib/load-config.js

@@ -24,10 +24,12 @@ module.exports.loadConfig = async () => {
 
   // ECS task definition is using NODE_ENV, so we'll support that too for now
   const envTag = (process.env.ENV || process.env.NODE_ENV)
-  const env = envTag &&
-    ['local', 'qa', 'qa-new-domain', 'production'].includes(envTag)
+  let env = envTag &&
+    ['local', 'qa', 'qa2', 'production'].includes(envTag)
     ? envTag
     : 'production'
+  // ECS task definition for qa2 has NODE_ENV=qa2, so translate that to qa:
+  env = env === 'qa2' ? 'qa' : env
   const envPath = `config/${env}.env`
 
   // Load env vars

lib/resources.js

@@ -62,7 +62,17 @@ const SORT_FIELDS = {
 }
 
 // The following fields can be excluded from ES responses because we don't pass them to client:
-const EXCLUDE_FIELDS = ['uris', '*_packed', '*_sort', 'items.*_packed', 'contentsTitle']
+const EXCLUDE_FIELDS = [
+  'uris',
+  '*_packed',
+  '*_sort',
+  'items.*_packed',
+  'contentsTitle',
+  'suppressed',
+  // Hide contributor and creator transformed fields:
+  '*WithoutDates',
+  '*Normalized'
+]
 
 // Configure controller-wide parameter parsing:
 const parseSearchParams = function (params) {
@@ -111,6 +121,12 @@ module.exports = function (app, _private = null) {
       include_item_aggregations: { type: 'boolean', default: true }
     })
 
+    // Validate uri:
+    const { id, nyplSource } = NyplSourceMapper.instance().splitIdentifier(params.uri)
+    if (!id || !nyplSource) {
+      throw new errors.InvalidParameterError(`Invalid bnum: ${params.uri}`)
+    }
+
     // If we need to return itemAggregations or filter on item_status,
     // then we need to pre-retrieve SCSB item statuses to incorporate them into
     // aggregations and filters.
@@ -211,7 +227,7 @@ module.exports = function (app, _private = null) {
       .then((resp) => {
         // need to check that the query actually found an entry
         if (!resp.data) {
-          throw new errors.NotFoundError('Record not found')
+          throw new errors.NotFoundError(`Record not found: bibs/${nyplSource}/${id}`)
         } else {
           return resp.data
         }

package.json

@@ -29,7 +29,8 @@
     "sinon": "^17.0.1",
     "standard": "^17.0.1",
     "supertest": "^7.0.0",
-    "table": "^6.8.2"
+    "table": "^6.8.2",
+    "yaml": "^2.6.0"
   },
   "scripts": {
     "test": "./node_modules/.bin/standard --env mocha && NODE_ENV=test ./node_modules/.bin/mocha test --exit",
@@ -58,7 +59,7 @@
       "/logs-to-tsv/*"
     ]
   },
-  "version": "1.1.3",
+  "version": "1.1.4",
   "repository": {
     "type": "git",
     "url": "https://github.com/NYPL/discovery-api.git"

swagger.v1.1.x.json

@@ -85,6 +85,7 @@
             "enum": [
               "all",
               "title",
+              "journal_title",
               "contributor",
               "subject",
               "series",
@@ -287,6 +288,7 @@
             "enum": [
               "all",
               "title",
+              "journal_title",
               "contributor",
               "subject",
               "series",

test/api-request.test.js

@@ -31,6 +31,10 @@ describe('ApiRequest', function () {
     request = ApiRequest.fromParams({ q: '"toast"' })
     expect(request.queryIsFullyQuoted()).to.eq(true)
 
+    // Test smart-quotes, the great scorge
+    request = ApiRequest.fromParams({ q: '“toast“' })
+    expect(request.queryIsFullyQuoted()).to.eq(true)
+
     request = ApiRequest.fromParams({ q: '"toast" and jam' })
     expect(request.queryIsFullyQuoted()).to.eq(false)
   })