Use sha256 and new Fedora bucket

digirati-co-uk · Mar 11, 2024 · 9852de7 · 9852de7
1 parent 7ea79e8
commit 9852de7
Show file tree

Hide file tree

Showing 3 changed files with 8 additions and 6 deletions.
diff --git a/LeedsExperiment/Preservation.API/Controllers/ImportController.cs b/LeedsExperiment/Preservation.API/Controllers/ImportController.cs
@@ -363,8 +363,10 @@ private async Task<ImportSource> GetImportSource(string source, Uri intendedPare
 
             // Unless coming from other information, we *require* that S3 source folders have sha256 hashes in their metadata
             // so we don't have to do this:
+
+            // TODAY
             var s3Stream = await s3Client!.GetObjectStreamAsync(obj.BucketName, obj.Key, null);
-            var sha512Digest = Checksum.Sha512FromStream(s3Stream);
+            var sha256Digest = Checksum.Sha256FromStream(s3Stream);
             // (and all our Fedora objects have sha-256)
             // We can also do an eTag comparison for smaller files
             // We can also do a size comparison as a sanity check - this can't catch all changes obvs
@@ -384,7 +386,7 @@ private async Task<ImportSource> GetImportSource(string source, Uri intendedPare
                 Path = sourcePath,
                 StorageType = StorageTypes.S3,
                 ExternalLocation = $"s3://{obj.BucketName}/{obj.Key}",
-                Digest = sha512Digest,
+                Digest = sha256Digest,
                 ContentType = GetDefaultContentType(nameAndParentPath.Name) // we may overwrite this later, e.g., from PREMIS data
             });
         }

diff --git a/LeedsExperiment/Preservation.API/appsettings.Production.json b/LeedsExperiment/Preservation.API/appsettings.Production.json
@@ -10,7 +10,7 @@
     "Region": "eu-west-1"
   },
   "Fedora-AWS-S3": {
-    "bucket": "uol-expts-fedora-01"
+    "bucket": "uol-expts-fedora-650"
   },
   "AllowedHosts": "*"
 }
diff --git a/LeedsExperiment/Preservation/FedoraWrapper.cs b/LeedsExperiment/Preservation/FedoraWrapper.cs
@@ -218,7 +218,7 @@ private async void EnsureChecksum(BinaryFile binaryFile)
         {
             case StorageTypes.FileSystem:
                 var fi = new FileInfo(binaryFile.ExternalLocation);
-                expected = Checksum.Sha512FromFile(fi);
+                expected = Checksum.Sha256FromFile(fi);
                 break;
             case StorageTypes.S3:
                 // TODO - get the SHA256 algorithm from AWS directly rather than compute it here
@@ -233,7 +233,7 @@ private async void EnsureChecksum(BinaryFile binaryFile)
                 // This would be an efficient way of doing this - but with this naive implementation
                 // we're going to read the object twice
                 var s3Stream = await s3Client!.GetObjectStreamAsync(s3Uri.Bucket, s3Uri.Key, null);
-                expected = Checksum.Sha512FromStream(s3Stream);
+                expected = Checksum.Sha256FromStream(s3Stream);
                 // could get a byte array here and then pass it along eventually to MakeBinaryPutOrPost
                 // for now just read it twice.
                 // Later we'll get the sha256 checksum from metadata
@@ -321,7 +321,7 @@ private async Task<HttpRequestMessage> MakeBinaryPutOrPost(HttpMethod httpMethod
     {
         var req = MakeHttpRequestMessage(location, httpMethod)
             .InTransaction(transaction)
-            .WithDigest(binaryFile.Digest, "sha-512"); // move algorithm choice to config
+            .WithDigest(binaryFile.Digest, "sha-256"); // move algorithm choice to constant
         if (httpMethod == HttpMethod.Post)
         {
             req.WithSlug(binaryFile.Slug);