Reroute batch blobs to single blob RPC

Given the priorities shifted for the Gitaly team, this endpoint does not
get a dedicated endpoint yet. To make it work in a cloud native
environment the request needs to go to Gitaly, not rugged. This is
achieved by rerouting to the generic TreeEntry endpoint.

2 files changed, 29 insertions, 17 deletions

diff --git a/lib/gitlab/git/blob.rb b/lib/gitlab/git/blob.rb
index 228d97a87ab..bd91125d3b6 100644
--- a/lib/gitlab/git/blob.rb
+++ b/lib/gitlab/git/blob.rb
@@ -50,10 +50,19 @@ module Gitlab
         # to the caller to limit the number of blobs and blob_size_limit.
         #
         # Gitaly migration issue: https://gitlab.com/gitlab-org/gitaly/issues/798
-        def batch(repository, blob_references, blob_size_limit: nil)
-          blob_size_limit ||= MAX_DATA_DISPLAY_SIZE
-          blob_references.map do |sha, path|
-            find_by_rugged(repository, sha, path, limit: blob_size_limit)
+        def batch(repository, blob_references, blob_size_limit: MAX_DATA_DISPLAY_SIZE)
+          Gitlab::GitalyClient.migrate(:list_blobs_by_sha_path) do |is_enabled|
+            if is_enabled
+              Gitlab::GitalyClient.allow_n_plus_1_calls do
+                blob_references.map do |sha, path|
+                  find_by_gitaly(repository, sha, path, limit: blob_size_limit)
+                end
+              end
+            else
+              blob_references.map do |sha, path|
+                find_by_rugged(repository, sha, path, limit: blob_size_limit)
+              end
+            end
           end
         end
 
@@ -122,13 +131,23 @@ module Gitlab
           )
         end
 
-        def find_by_gitaly(repository, sha, path)
+        def find_by_gitaly(repository, sha, path, limit: MAX_DATA_DISPLAY_SIZE)
           path = path.sub(/\A\/*/, '')
           path = '/' if path.empty?
           name = File.basename(path)
-          entry = Gitlab::GitalyClient::CommitService.new(repository).tree_entry(sha, path, MAX_DATA_DISPLAY_SIZE)
+
+          # Gitaly will think that setting the limit to 0 means unlimited, while
+          # the client might only need the metadata and thus set the limit to 0.
+          # In this method we'll than set the limit to 1, but clear the byte of data
+          # that we got back so fot the outside world it looks like the limit was
+          # actually 0.
+          req_limit = limit == 0 ? 1 : limit
+
+          entry = Gitlab::GitalyClient::CommitService.new(repository).tree_entry(sha, path, req_limit)
           return unless entry
 
+          entry.data = "" if limit == 0
+
           case entry.type
           when :COMMIT
             new(
diff --git a/spec/lib/gitlab/git/blob_spec.rb b/spec/lib/gitlab/git/blob_spec.rb
index c04a9688503..7f5946b1658 100644
--- a/spec/lib/gitlab/git/blob_spec.rb
+++ b/spec/lib/gitlab/git/blob_spec.rb
@@ -202,16 +202,6 @@ describe Gitlab::Git::Blob, seed_helper: true do
     context 'limiting' do
       subject { described_class.batch(repository, blob_references, blob_size_limit: blob_size_limit) }
 
-      context 'default' do
-        let(:blob_size_limit) { nil }
-
-        it 'limits to MAX_DATA_DISPLAY_SIZE' do
-          stub_const('Gitlab::Git::Blob::MAX_DATA_DISPLAY_SIZE', 100)
-
-          expect(subject.first.data.size).to eq(100)
-        end
-      end
-
       context 'positive' do
         let(:blob_size_limit) { 10 }
 
@@ -221,7 +211,10 @@ describe Gitlab::Git::Blob, seed_helper: true do
       context 'zero' do
         let(:blob_size_limit) { 0 }
 
-        it { expect(subject.first.data).to eq('') }
+        it 'only loads the metadata' do
+          expect(subject.first.size).not_to be(0)
+          expect(subject.first.data).to eq('')
+        end
       end
 
       context 'negative' do