From 6216074caa2043bf1bc6a09ab7c72609e8dec0b3 Mon Sep 17 00:00:00 2001
From: kwahlin <kalle.wahlin@kb.se>
Date: Tue, 25 Apr 2023 14:08:22 +0200
Subject: [PATCH] Always produce report with split clusters

---
 .../datatool/scripts/mergeworks/Html.groovy   | 30 +++++++++++++------
 .../scripts/mergeworks/WorkToolJob.groovy     | 20 ++++++++++---
 .../scripts/mergeworks/compare/Id.groovy      |  2 +-
 3 files changed, 38 insertions(+), 14 deletions(-)

diff --git a/whelktool/src/main/groovy/datatool/scripts/mergeworks/Html.groovy b/whelktool/src/main/groovy/datatool/scripts/mergeworks/Html.groovy
index 8e7d045b54..4b57e963bf 100644
--- a/whelktool/src/main/groovy/datatool/scripts/mergeworks/Html.groovy
+++ b/whelktool/src/main/groovy/datatool/scripts/mergeworks/Html.groovy
@@ -52,27 +52,39 @@ class Html {
     }
 
     static String hubTable(List<Collection<Doc>> docs) {
-        def mergedWorks = docs*.first()
-        def ids = docs.collect { group ->
-            group.drop(1).collectEntries { doc ->
+        def mergedWorks = []
+        def derivedFromIds = []
+
+        docs.each {
+            def work = it.head()
+            def derivedFrom = it.tail()
+            mergedWorks.add(work)
+            derivedFromIds.add(derivedFrom.collectEntries { doc ->
                 [doc.doc.shortId, doc.view.link()]
-            }
+            })
         }
-        def clusterId = clusterId(ids*.keySet().flatten())
+
+        def clusterId = clusterId(derivedFromIds*.keySet().flatten())
 
         String header = """
             <tr>
                 <th><a id="${clusterId}"><a href="#${clusterId}">${clusterId}</th>
-                ${mergedWorks.collect { "<th></th>" }.join('\n')}
+                ${docs.collect {
+            def work = it.head()
+            def derivedFrom = it.tail()
+            derivedFrom.size() > 1 || it instanceof UpdatedWork
+                    ? "<th><a id=\"${work.doc.shortId}\" href=\"${work.view.link()}\">${work.doc.shortId}</a></th>"
+                    : "<th></th>" }
+                .join('\n')}
             </tr>
            """.stripIndent()
 
         String derivedFrom =
                 """
                     <tr class="info">
-                        <td>_derivedFrom</td>
-                        ${ids.collect { "<td>${it.collect { id, link -> "<a id=\"$id\" href=\"$link\">$id</a>" }.join('<br>')}</td>" }.join('\n')}
-                        </tr> 
+                        <td>_instances</td>
+                        ${derivedFromIds.collect { "<td>${it.collect { id, link -> "<a id=\"$id\" href=\"$link\">$id</a>" }.join('<br>')}</td>" }.join('\n')}
+                        </tr>
                 """.stripIndent()
 
         def statuses = WorkComparator.compare(mergedWorks)
diff --git a/whelktool/src/main/groovy/datatool/scripts/mergeworks/WorkToolJob.groovy b/whelktool/src/main/groovy/datatool/scripts/mergeworks/WorkToolJob.groovy
index b75346a92e..c81f578129 100644
--- a/whelktool/src/main/groovy/datatool/scripts/mergeworks/WorkToolJob.groovy
+++ b/whelktool/src/main/groovy/datatool/scripts/mergeworks/WorkToolJob.groovy
@@ -132,14 +132,19 @@ class WorkToolJob {
 
     void merge() {
         def s = statistics.printOnShutdown()
+        def multiWorkClusters = Collections.synchronizedList([])
 
         run({ cluster ->
             return {
                 def titles = titleClusters(cluster)
                 def works = mergedWorks(titles)
 
-                def needsStore = { it instanceof UpdatedWork || it.derivedFrom.size() > 1 }
-                def storedWorks = works.findAll(needsStore).each { store(it) }
+                if (works.size() > 1) {
+                    multiWorkClusters.add(works.collect { [new Doc(whelk, it.doc)] + it.derivedFrom })
+                }
+
+                def multiInstanceWorks = works.findAll { it.derivedFrom.size() > 1 || it instanceof UpdatedWork }
+                def storedWorks = multiInstanceWorks.each { store(it) }
 
                 String report = htmlReport(titles, storedWorks)
 
@@ -155,6 +160,14 @@ class WorkToolJob {
                 }
             }
         })
+
+        new File(reportDir, "multi-work-clusters.html").with {f ->
+            f.append(Html.START)
+            multiWorkClusters.each {
+                f.append(Html.hubTable(it) + Html.HORIZONTAL_RULE)
+            }
+            f.append(Html.END)
+        }
     }
 
     void revert() {
@@ -226,8 +239,7 @@ class WorkToolJob {
                         LegacyIntegrationTools.determineLegacyCollection(work.doc, whelk.getJsonld()), false)) {
                     throw new WhelkRuntimeException("Could not store new work: ${work.doc.shortId}")
                 }
-            }
-            else if (work instanceof UpdatedWork) {
+            } else if (work instanceof UpdatedWork) {
                 whelk.storeAtomicUpdate(work.doc, !loud, false, changedIn, generationProcess, work.checksum)
             }
 
diff --git a/whelktool/src/main/groovy/datatool/scripts/mergeworks/compare/Id.groovy b/whelktool/src/main/groovy/datatool/scripts/mergeworks/compare/Id.groovy
index 555e661ac4..fc3305148b 100644
--- a/whelktool/src/main/groovy/datatool/scripts/mergeworks/compare/Id.groovy
+++ b/whelktool/src/main/groovy/datatool/scripts/mergeworks/compare/Id.groovy
@@ -7,7 +7,7 @@ class Id implements ValuePicker {
 
     @Override
     boolean isCompatible(Object a, Object b) {
-        return !a || !b
+        return true
     }
 
     @Override