Merge remote-tracking branch 'upstream/main' into SOLR-17577

apache · Dec 2, 2024 · 06f6e0c · 06f6e0c
2 parents b7432f8 + cebdb2d
commit 06f6e0c
Show file tree

Hide file tree

Showing 5 changed files with 37 additions and 12 deletions.
diff --git a/solr/CHANGES.txt b/solr/CHANGES.txt
@@ -227,6 +227,8 @@ Bug Fixes
 * SOLR-16976: Remove log4j-jul jar and use slf4j bridge for JUL to prevent exception from being logged when remote JMX
   is enabled (Shawn Heisey, Stephen Zhou, Eric Pugh, Christine Poerschke, David Smiley)
 
+* SOLR-17575: Fixed broken backwards compatibility with the legacy "langid.whitelist" config in Solr Langid. (Jan Høydahl, Alexander Zagniotov)
+
 Dependency Upgrades
 ---------------------
 (No changes)

diff --git a/solr/bin/solr.cmd b/solr/bin/solr.cmd
@@ -1174,11 +1174,9 @@ for %%a in (%*) do (
       if "!arg:~0,1!" equ "-" set "option=!arg!"
    ) else (
       set "option!option!=%%a"
-      if "!option!" equ "-s" set "SOLR_HOME=%%a"
       if "!option!" equ "--solr-home" set "SOLR_HOME=%%a"
-      if "!option!" equ "-d" set "SOLR_SERVER_DIR=%%a"
-      if "!option!" equ "--server-dir" set "SOLR_SERVER_DIR=%%a"
-      if not "!option!" equ "-s" if not "!option!" equ "--solr-home" if not "!option!" equ "-d" if not "!option!" equ "--server-dir" (
+      if "!option!" equ "--server-dir" set "SOLR_SERVER_DIR=%%a"    
+      if not "!option!" equ "--solr-home" if not "!option!" equ "--server-dir" (
         set "AUTH_PARAMS=!AUTH_PARAMS! !option! %%a"
       )
       set "option="

diff --git a/...s/langid/src/java/org/apache/solr/update/processor/LanguageIdentifierUpdateProcessor.java b/...s/langid/src/java/org/apache/solr/update/processor/LanguageIdentifierUpdateProcessor.java
@@ -111,20 +111,19 @@ private void initParams(SolrParams params) {
       overwrite = params.getBool(OVERWRITE, false);
       langAllowlist = new HashSet<>();
       threshold = params.getDouble(THRESHOLD, DOCID_THRESHOLD_DEFAULT);
-      String legacyAllowList = params.get(LANG_WHITELIST, "");
-      if (legacyAllowList.length() > 0) {
+      final String legacyAllowList = params.get(LANG_WHITELIST, "").trim();
+      if (!legacyAllowList.isEmpty()) {
         // nowarn compile time string concatenation
         log.warn(
             LANG_WHITELIST
                 + " parameter is deprecated; use "
                 + LANG_ALLOWLIST
                 + " instead."); // nowarn
       }
-      if (params.get(LANG_ALLOWLIST, legacyAllowList).length() > 0) {
-        for (String lang : params.get(LANG_ALLOWLIST, "").split(",")) {
-          langAllowlist.add(lang);
-        }
-      }
+      Arrays.stream(params.get(LANG_ALLOWLIST, legacyAllowList).split(","))
+          .map(String::trim)
+          .filter(lang -> !lang.isEmpty())
+          .forEach(langAllowlist::add);
 
       // Mapping params (field centric)
       enableMapping = params.getBool(MAP_ENABLE, false);

diff --git a/...st/org/apache/solr/update/processor/LanguageIdentifierUpdateProcessorFactoryTestCase.java b/...st/org/apache/solr/update/processor/LanguageIdentifierUpdateProcessorFactoryTestCase.java
@@ -18,6 +18,7 @@
 
 import java.util.ArrayList;
 import java.util.List;
+import java.util.Set;
 import org.apache.solr.SolrTestCaseJ4;
 import org.apache.solr.common.SolrInputDocument;
 import org.apache.solr.common.params.ModifiableSolrParams;
@@ -464,6 +465,31 @@ public void testMapIndividual() throws Exception {
     assertTrue(mappedIndividual.containsKey("text2_ru"));
   }
 
+  @Test
+  public void testAllowlist() throws Exception {
+    ModifiableSolrParams parameters = new ModifiableSolrParams();
+    parameters.add("langid.fl", "name,subject");
+    parameters.add("langid.langField", "language_s");
+    parameters.add("langid.allowlist", "no,en ,, ,sv, sv");
+    liProcessor = createLangIdProcessor(parameters);
+
+    // Make sure that empty language codes have been filtered out and others trimmed.
+    assertEquals(Set.of("no", "en", "sv"), liProcessor.langAllowlist);
+  }
+
+  @Test
+  public void testAllowlistBackwardsCompatabilityWithLegacyAllowlist() throws Exception {
+    // The "legacy allowlist" is "langid.whitelist"
+    ModifiableSolrParams parameters = new ModifiableSolrParams();
+    parameters.add("langid.fl", "name,subject");
+    parameters.add("langid.langField", "language_s");
+    parameters.add("langid.whitelist", "no,en ,, ,sv, sv");
+    liProcessor = createLangIdProcessor(parameters);
+
+    // Make sure that empty language codes have been filtered out and others trimmed.
+    assertEquals(Set.of("no", "en", "sv"), liProcessor.langAllowlist);
+  }
+
   // Various utility methods
 
   private SolrInputDocument englishDoc() {

diff --git a/solr/solr-ref-guide/modules/deployment-guide/pages/jwt-authentication-plugin.adoc b/solr/solr-ref-guide/modules/deployment-guide/pages/jwt-authentication-plugin.adoc
@@ -169,7 +169,7 @@ Let's comment on this config:
 <1> Plugin class
 <2> Make sure to block anyone without a valid token (this is also the default)
 <3> Fetch the user id from another claim than the default `sub`
-<4> Require that the `roles` claim is one of "A" or "B" and that the `dept` claim is "IT"
+<4> Require that the `foo` claim is one of "A" or "B" and that the `dept` claim is "IT"
 <5> Require one of the scopes `solr:read`, `solr:write` or `solr:admin`
 <6> Only accept RSA algorithms for signatures
 <7> Array of issuer configurations