Improve site info for unknown domains:

* Placeholder screenshot should work
* Add a link to git-repo for submitting the site for crawling

code/common/db/src/main/java/nu/marginalia/db/DbDomainQueries.java

@@ -47,6 +47,33 @@ public class DbDomainQueries {
         }
     }
 
+    @SneakyThrows
+    public Optional<EdgeId<EdgeDomain>> tryGetDomainId(EdgeDomain domain) {
+
+        var maybe = Optional.ofNullable(domainIdCache.getIfPresent(domain));
+
+        if (maybe.isPresent())
+            return maybe;
+
+        try (var connection = dataSource.getConnection()) {
+
+            try (var stmt = connection.prepareStatement("SELECT ID FROM EC_DOMAIN WHERE DOMAIN_NAME=?")) {
+                stmt.setString(1, domain.toString());
+                var rsp = stmt.executeQuery();
+                if (rsp.next()) {
+                    var id = new EdgeId<EdgeDomain>(rsp.getInt(1));
+
+                    domainIdCache.put(domain, id);
+                    return Optional.of(id);
+                }
+            }
+            return Optional.empty();
+        }
+        catch (UncheckedExecutionException ex) {
+            return Optional.empty();
+        }
+    }
+
     @SneakyThrows
     public Optional<EdgeDomain> getDomain(EdgeId<EdgeDomain> id) {
         try (var connection = dataSource.getConnection()) {

code/services-core/search-service/src/main/java/nu/marginalia/search/command/commands/SiteListCommand.java

@@ -2,6 +2,7 @@ package nu.marginalia.search.command.commands;
 
 import com.google.inject.Inject;
 import nu.marginalia.db.DbDomainQueries;
+import nu.marginalia.model.EdgeDomain;
 import nu.marginalia.search.model.UrlDetails;
 import nu.marginalia.search.command.SearchCommandInterface;
 import nu.marginalia.search.command.SearchParameters;
@@ -64,8 +65,15 @@ public class SiteListCommand implements SearchCommandInterface {
         if (null != domain) {
             var dumbQuery = queryFactory.createQuery(SearchProfile.CORPO, 100, 100, "site:"+domain);
             resultSet = searchQueryIndexService.executeQuery(ctx, dumbQuery);
-            domainId = domainQueries.getDomainId(domain).id();
+            var maybeId = domainQueries.tryGetDomainId(domain);
-            screenshotPath = Path.of("/screenshot/" + domainId);
+            if (maybeId.isPresent()) {
+                domainId = maybeId.get().id();
+                screenshotPath = Path.of("/screenshot/" + domainId);
+            }
+            else {
+                domainId = -1;
+                screenshotPath = Path.of("/screenshot/0");
+            }
         }
         else {
             resultSet = Collections.emptyList();
@@ -90,7 +98,10 @@ public class SiteListCommand implements SearchCommandInterface {
         String word = humanQuery.substring(definePrefix.length()).toLowerCase();
 
         logger.info("Fetching Site Info: {}", word);
-        var results = domainInformationService.domainInfo(word).orElseGet(DomainInformation::new);
+
+        var results = domainInformationService
+                .domainInfo(word)
+                .orElseGet(() -> unknownSite(word));
 
         logger.debug("Results = {}", results);
 
@@ -98,4 +109,11 @@ public class SiteListCommand implements SearchCommandInterface {
 
     }
 
+    private DomainInformation unknownSite(String url) {
+        return DomainInformation.builder()
+                .domain(new EdgeDomain(url))
+                .suggestForCrawling(true)
+                .unknownDomain(true)
+                .build();
+    }
 }

code/services-core/search-service/src/main/java/nu/marginalia/search/model/DomainInformation.java

@@ -20,6 +20,7 @@ public class DomainInformation {
 
     boolean suggestForCrawling;
     boolean inCrawlQueue;
+    boolean unknownDomain;
 
     String state;
     List<EdgeDomain> linkingDomains;

code/services-core/search-service/src/main/resources/templates/search/parts/site-info-index.hdb

@@ -26,19 +26,34 @@
         {{/if}}
 
         {{#if suggestForCrawling}}
-        <form method="POST" action="/site/suggest/">
+            {{#if unknownDomain}}
+
         <fieldset>
             <legend>Crawling</legend>
-            This website is not queued for crawling. If you would like it to be crawled,
+            This website is not known to the search engine.
-            use the checkbox and button below.<p/>
+
-            <input type="hidden" name="id" value="{{domainId}}" />
+            To submit the website for crawling, follow <a
-            <input type="checkbox" id="nomisclick" name="nomisclick" /> <label for="nomisclick"> This is not a mis-click </label>
+                    rel="noopener noreferrer"
-            <br/>
+                    target="_blank"
-            <br/>
+                    href="https://github.com/MarginaliaSearch/submit-site-to-marginalia-search">these instructions</a>.
-            <input type="submit" value="Add {{domain}} to queue" />
         </fieldset>
+            {{/if}}
+            {{#unless unknownDomain}}
+        <form method="POST" action="/site/suggest/">
+            <fieldset>
+                <legend>Crawling</legend>
+                This website is not queued for crawling. If you would like it to be crawled,
+                use the checkbox and button below.<p/>
+                <input type="hidden" name="id" value="{{domainId}}" />
+                <input type="checkbox" id="nomisclick" name="nomisclick" /> <label for="nomisclick"> This is not a mis-click </label>
+                <br/>
+                <br/>
+                <input type="submit" value="Add {{domain}} to queue" />
+            </fieldset>
         </form>
+            {{/unless}}
         {{/if}}
+
         {{#if pagesFetched}}
         <p>
         If you've found a reason why this website should not be indexed,