only log archive provenance for namespace once; #52

globalbioticinteractions · Nov 30, 2023 · 0ca2f7f · 0ca2f7f
1 parent 3fd598b
commit 0ca2f7f
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 11 deletions.
diff --git a/src/main/java/org/globalbioticinteractions/elton/cmd/CmdLog.java b/src/main/java/org/globalbioticinteractions/elton/cmd/CmdLog.java
@@ -39,6 +39,7 @@
 import java.util.Arrays;
 import java.util.Collections;
 import java.util.concurrent.atomic.AtomicBoolean;
+import java.util.concurrent.atomic.AtomicReference;
 
 @CommandLine.Command(
         name = "log",
@@ -88,6 +89,7 @@ private static class LoggingResourceService implements ResourceService {
         private final PrintStream out;
         private HashType sha256;
         private final ResourceService local;
+        private final AtomicReference<IRI> archiveContentId = new AtomicReference<>(null);
 
         public LoggingResourceService(PrintStream out, ResourceService resourceService) {
             this.out = out;
@@ -119,17 +121,20 @@ private URI getLocationInDataset(URI uri, Dataset dataset) throws IOException {
                 if (CacheUtil.isLocalDir(archiveURI)) {
                     resourceLocation = ResourceUtil.getAbsoluteResourceURI(archiveURI, uri);
                 } else {
-                    IRI archiveContentId = Hasher.calcHashIRI(local.retrieve(archiveURI), NullOutputStream.NULL_OUTPUT_STREAM, sha256);
-                    Quad quad = RefNodeFactory.toStatement(
-                            RefNodeFactory.toIRI(archiveURI),
-                            RefNodeConstants.HAS_VERSION,
-                            archiveContentId
-                    );
-                    out.println(quad.toString());
+                    if (this.archiveContentId.get() == null) {
+                        IRI archiveContentId = Hasher.calcHashIRI(local.retrieve(archiveURI), NullOutputStream.NULL_OUTPUT_STREAM, sha256);
+                        this.archiveContentId.set(archiveContentId);
+                        Quad quad = RefNodeFactory.toStatement(
+                                RefNodeFactory.toIRI(archiveURI),
+                                RefNodeConstants.HAS_VERSION,
+                                archiveContentId
+                        );
+                        out.println(quad.toString());
+                    }
 
                     String localDatasetRoot = DatasetFinderUtil.getLocalDatasetURIRoot(local.retrieve(archiveURI));
 
-                    URI localArchiveRoot = URI.create("zip:" + archiveContentId.getIRIString() + "!/" + localDatasetRoot);
+                    URI localArchiveRoot = URI.create("zip:" + archiveContentId.get().getIRIString() + "!/" + localDatasetRoot);
                     resourceLocation = ResourceUtil.getAbsoluteResourceURI(localArchiveRoot, uri);
                 }
             }

diff --git a/src/test/java/org/globalbioticinteractions/elton/cmd/CmdLogTest.java b/src/test/java/org/globalbioticinteractions/elton/cmd/CmdLogTest.java
@@ -21,11 +21,10 @@ public void logTemplate() throws URISyntaxException {
         PrintStream out = new PrintStream(out1);
         cmd.run(out);
         String[] split = out1.toString().split("\n");
-        assertThat(split.length, is(4));
+        assertThat(split.length, is(3));
         assertThat(split[0], is("<https://zenodo.org/record/207958/files/globalbioticinteractions/template-dataset-0.0.2.zip> <http://purl.org/pav/hasVersion> <hash://sha256/631d3777cf83e1abea848b59a6589c470cf0c7d0fd99682c4c104481ad9a543f> ."));
         assertThat(split[1], is("<zip:hash://sha256/631d3777cf83e1abea848b59a6589c470cf0c7d0fd99682c4c104481ad9a543f!/globalbioticinteractions-template-dataset-e68f448/globi.json> <http://purl.org/pav/hasVersion> <hash://sha256/1cc8eff62af0e6bb3e7771666e2e4109f351b7dfc6fc1dc8314e5671a8eecb80> ."));
-        assertThat(split[2], is("<https://zenodo.org/record/207958/files/globalbioticinteractions/template-dataset-0.0.2.zip> <http://purl.org/pav/hasVersion> <hash://sha256/631d3777cf83e1abea848b59a6589c470cf0c7d0fd99682c4c104481ad9a543f> ."));
-        assertThat(split[3], is("<zip:hash://sha256/631d3777cf83e1abea848b59a6589c470cf0c7d0fd99682c4c104481ad9a543f!/globalbioticinteractions-template-dataset-e68f448/interactions.tsv> <http://purl.org/pav/hasVersion> <hash://sha256/c1b37add5ee5f30916f19811c59c2960e3b68ecf1a3846afe1776014c4c96271> ."));
+        assertThat(split[2], is("<zip:hash://sha256/631d3777cf83e1abea848b59a6589c470cf0c7d0fd99682c4c104481ad9a543f!/globalbioticinteractions-template-dataset-e68f448/interactions.tsv> <http://purl.org/pav/hasVersion> <hash://sha256/c1b37add5ee5f30916f19811c59c2960e3b68ecf1a3846afe1776014c4c96271> ."));
 
     }