(site-info) Try to discover the schema of the website with a site:-query

The site info view can't blindly assume that every website supports https.  To figure out which schema to use when linking to a site, execute a single-result search for site:domain.name and then grab the schema off the result.

To allow this, a count parameter is introduced to doSiteSearch() in SearchOperator.

code/services-application/search-service/src/main/java/nu/marginalia/search/SearchOperator.java

@@ -68,13 +68,15 @@ public class SearchOperator {
     }
 
     public List<UrlDetails> doSiteSearch(Context ctx,
-                                        String domain) {
+                                        String domain,
+                                        int count) {
 
-        var queryParams = paramFactory.forSiteSearch(domain);
+        var queryParams = paramFactory.forSiteSearch(domain, count);
         var queryResponse = queryClient.search(ctx, queryParams);
 
         return searchQueryService.getResultsFromQuery(queryResponse);
     }
+
     public List<UrlDetails> doBacklinkSearch(Context ctx,
                                          String domain) {
 

code/services-application/search-service/src/main/java/nu/marginalia/search/SearchQueryParamFactory.java

@@ -36,7 +36,7 @@ public class SearchQueryParamFactory {
 
     }
 
-    public QueryParams forSiteSearch(String domain) {
+    public QueryParams forSiteSearch(String domain, int count) {
         return new QueryParams("site:"+domain,
                 null,
                 List.of(),
@@ -48,12 +48,11 @@ public class SearchQueryParamFactory {
                 SpecificationLimit.none(),
                 SpecificationLimit.none(),
                 List.of(),
-                new QueryLimits(100, 100, 100, 512),
+                new QueryLimits(count, count, 100, 512),
                 SearchSetIdentifier.NONE
         );
     }
 
-
     public QueryParams forBacklinkSearch(String domain) {
         return new QueryParams("links:"+domain,
                 null,

code/services-application/search-service/src/main/java/nu/marginalia/search/svc/SearchSiteInfoService.java

@@ -119,6 +119,7 @@ public class SearchSiteInfoService {
         final DomainInformation domainInfo;
         final List<SimilarDomain> similarSet;
         final List<SimilarDomain> linkingDomains;
+        String url = "https://" + domainName + "/";;
 
         if (domainId < 0 || !assistantClient.isAccepting()) {
             domainInfo = createDummySiteInfo(domainName);
@@ -133,10 +134,16 @@ public class SearchSiteInfoService {
             linkingDomains = assistantClient
                     .linkedDomains(ctx, domainId, 100)
                     .blockingFirst();
+
+            List<UrlDetails> sampleResults = searchOperator.doSiteSearch(ctx, domainName, 1);
+            if (!sampleResults.isEmpty()) {
+                url = sampleResults.getFirst().url.withPathAndParam("/", null).toString();
+            }
         }
 
         return new SiteInfoWithContext(domainName,
                 domainId,
+                url,
                 domainInfo,
                 similarSet,
                 linkingDomains
@@ -154,7 +161,7 @@ public class SearchSiteInfoService {
     private Docs listDocs(Context ctx, String domainName) {
         return new Docs(domainName,
                 domainQueries.tryGetDomainId(new EdgeDomain(domainName)).orElse(-1),
-                searchOperator.doSiteSearch(ctx, domainName));
+                searchOperator.doSiteSearch(ctx, domainName, 100));
     }
 
     public record Docs(Map<String, Boolean> view,
@@ -190,11 +197,13 @@ public class SearchSiteInfoService {
                                       Map<String, Boolean> domainState,
                                       String domain,
                                       long domainId,
+                                      String siteUrl,
                                       DomainInformation domainInformation,
                                       List<SimilarDomain> similar,
                                       List<SimilarDomain> linking) {
         public SiteInfoWithContext(String domain,
                                    long domainId,
+                                   String siteUrl,
                                    DomainInformation domainInformation,
                                    List<SimilarDomain> similar,
                                    List<SimilarDomain> linking
@@ -204,6 +213,7 @@ public class SearchSiteInfoService {
                     Map.of(domainInfoState(domainInformation), true),
                     domain,
                     domainId,
+                    siteUrl,
                     domainInformation,
                     similar,
                     linking);

code/services-application/search-service/src/main/resources/templates/search/site-info/site-info-summary.hdb

@@ -5,10 +5,10 @@
 
 <div id="similar-view" data-layout="{{layout}}">
     <div id="similar-info">
-        <h2><span title="External Link">&#x1F30E;</span>&nbsp;<a rel="external noopener" href="https://{{domain}}/">{{domain}}</a></h2>
+        <h2><span title="External Link">&#x1F30E;</span>&nbsp;<a rel="external noopener" href="{{siteUrl}}">{{domain}}</a></h2>
 
 
-        <a rel="external noopener" href="https://{{domain}}/">
+        <a rel="external noopener" href="{{siteUrl}}">
             <img class="screenshot" width="300" height="225" src="/screenshot/{{domainId}}" alt="Screenshot of {{domain}}" />
         </a>
         {{#with domainInformation}}