Encode when migrating ProcessCommitWorker jobsprocess-commit-worker-migration-encoding

If the source encoding is not UTF-8 we need to encode the data as
`JSON.dump` may throw an error if the input can not be converted to
UTF-8. We only encode when necessary to reduce the overhead.

Fixes gitlab-org/gitlab-ce#25489

1 files changed, 15 insertions, 5 deletions

diff --git a/db/migrate/20161124141322_migrate_process_commit_worker_jobs.rb b/db/migrate/20161124141322_migrate_process_commit_worker_jobs.rb
index 453a44e271a..77e0c40d850 100644
--- a/db/migrate/20161124141322_migrate_process_commit_worker_jobs.rb
+++ b/db/migrate/20161124141322_migrate_process_commit_worker_jobs.rb
@@ -47,14 +47,14 @@ class MigrateProcessCommitWorkerJobs < ActiveRecord::Migration
 
         hash = {
           id: commit.oid,
-          message: commit.message,
+          message: encode(commit.message),
           parent_ids: commit.parent_ids,
           authored_date: commit.author[:time],
-          author_name: commit.author[:name],
-          author_email: commit.author[:email],
+          author_name: encode(commit.author[:name]),
+          author_email: encode(commit.author[:email]),
           committed_date: commit.committer[:time],
-          committer_email: commit.committer[:email],
-          committer_name: commit.committer[:name]
+          committer_email: encode(commit.committer[:email]),
+          committer_name: encode(commit.committer[:name])
         }
 
         payload['args'][2] = hash
@@ -89,4 +89,14 @@ class MigrateProcessCommitWorkerJobs < ActiveRecord::Migration
       end
     end
   end
+
+  def encode(data)
+    encoding = Encoding::UTF_8
+
+    if data.encoding == encoding
+      data
+    else
+      data.encode(encoding, invalid: :replace, undef: :replace)
+    end
+  end
 end