Class: MergeRequestDiffCommit

Inherits:
ApplicationRecord show all
Extended by:
SuppressCompositePrimaryKeyWarning
Includes:
BulkInsertSafe, CachedCommit, FromUnion, ShaAttribute
Defined in:
app/models/merge_request_diff_commit.rb

Constant Summary collapse

CouldNotCreateMetadataError = 
Class.new(StandardError)
TRIM_USER_KEYS =

A list of keys of which their values need to be trimmed before they can be inserted into the merge_request_diff_commit_users table.

i[author_name author_email committer_name committer_email].freeze

Constants included from BulkInsertSafe

BulkInsertSafe::ALLOWED_CALLBACKS, BulkInsertSafe::DEFAULT_BATCH_SIZE, BulkInsertSafe::MethodNotAllowedError, BulkInsertSafe::PrimaryKeySetError

Constants inherited from ApplicationRecord

ApplicationRecord::MAX_PLUCK

Constants included from HasCheckConstraints

HasCheckConstraints::NOT_NULL_CHECK_PATTERN

Constants included from ResetOnColumnErrors

ResetOnColumnErrors::MAX_RESET_PERIOD

Class Method Summary collapse

Instance Method Summary collapse

Methods included from CachedCommit

#extended_trailers, #parent_ids, #referenced_by

Methods inherited from ApplicationRecord

===, cached_column_list, #create_or_load_association, current_transaction, declarative_enum, default_select_columns, delete_all_returning, #deleted_from_database?, id_in, id_not_in, iid_in, nullable_column?, primary_key_in, #readable_by?, safe_ensure_unique, safe_find_or_create_by, safe_find_or_create_by!, #to_ability_name, underscore, where_exists, where_not_exists, with_fast_read_statement_timeout, without_order

Methods included from Organizations::Sharding

#sharding_organization

Methods included from ResetOnColumnErrors

#reset_on_union_error, #reset_on_unknown_attribute_error

Methods included from Gitlab::SensitiveSerializableHash

#serializable_hash

Class Method Details

.commit_rows_with_metadata(project_id, merge_request_diff_id, rows) ⇒ Object 



160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
 
# File 'app/models/merge_request_diff_commit.rb', line 160

def self.commit_rows_with_metadata(project_id, merge_request_diff_id, rows)
  commits_metadata_mapping = MergeRequest::CommitsMetadata.bulk_find_or_create(
    project_id,
    rows
  )

  rows.each do |row|
    row[:merge_request_commits_metadata_id] = commits_metadata_mapping[row[:raw_sha]]

    # At this point, we no longer need the `raw_sha` so we delete it from
    # the row that will be inserted into `merge_request_diff_commits` table.
    row.delete(:raw_sha)
  end

  rows_without_metadata = rows.select { |row| row[:merge_request_commits_metadata_id].nil? }

  if rows_without_metadata.any?
    Gitlab::ErrorTracking.track_exception(
      CouldNotCreateMetadataError.new,
      message: 'Failed to create metadata',
      failed_count: rows_without_metadata.size,
      total_count: rows.size,
      merge_request_diff_id: merge_request_diff_id,
      project_id: project_id,
      relative_orders: rows_without_metadata.filter_map { |r| r[:relative_order] }
    )
  end

  rows
end

.commit_shas_from_metadata(project_id:, limit:) ⇒ Object 



139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
 
# File 'app/models/merge_request_diff_commit.rb', line 139

def self.commit_shas_from_metadata(project_id:, limit:)
  # Until `merge_request_commits_metadata` records are backfilled, SHAs data may be in found in either table
  metadata_join_sql = "    LEFT JOIN merge_request_commits_metadata\n    ON merge_request_commits_metadata.id = merge_request_diff_commits.merge_request_commits_metadata_id\n    AND merge_request_commits_metadata.project_id = ?\n  SQL\n\n  # raw SQL in pluck() bypass ActiveRecord's type casting, so encode() is needed to convert bytea to hex\n  shas_sql = Arel.sql(\"encode(COALESCE(merge_request_commits_metadata.sha, merge_request_diff_commits.sha), 'hex')\")\n\n  relation = self.joins(self.sanitize_sql_array([metadata_join_sql, project_id]))\n    .order(:relative_order)\n\n  relation = relation.limit(limit) if limit\n\n  # rubocop:disable Database/AvoidUsingPluckWithoutLimit -- limit may be applied in the caller\n  relation.pluck(shas_sql)\n  # rubocop:enable Database/AvoidUsingPluckWithoutLimit\nend\n".squish

.create_bulk(merge_request_diff_id, commits, project, skip_commit_data: false) ⇒ Object

Deprecated; use bulk_insert! from BulkInsertSafe mixin instead. cf. gitlab.com/gitlab-org/gitlab/issues/207989 for progress



51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
 
# File 'app/models/merge_request_diff_commit.rb', line 51

def self.create_bulk(merge_request_diff_id, commits, project, skip_commit_data: false)
  organization_id = project.organization_id
  dedup_enabled = Feature.enabled?(:merge_request_diff_commits_dedup, project)
  partition_enabled = Feature.enabled?(:merge_request_diff_commits_partition, project)
  commit_hashes, user_triples = prepare_commits_for_bulk_insert(commits, organization_id)
  users = MergeRequest::DiffCommitUser.bulk_find_or_create(user_triples)

  rows = commit_hashes.map.with_index do |commit_hash, index|
    raw_sha = commit_hash.delete(:id)
    trailers = commit_hash.fetch(:trailers, {})

    author = users[[commit_hash[:author_name], commit_hash[:author_email], organization_id]]
    committer = users[[commit_hash[:committer_name], commit_hash[:committer_email], organization_id]]

    # These fields are only used to determine the author/committer IDs, we
    # don't store them in the DB.
    #
    # Trailers are stored in the DB here in order to allow changelog parsing.
    # Rather than add an additional column for :extended_trailers, we're instead
    # ignoring it for now until we deprecate the :trailers field and replace it with
    # the new functionality.
    commit_hash = commit_hash
      .except(:author_name, :author_email, :committer_name, :committer_email, :extended_trailers)

    commit_hash = commit_hash.merge(
      commit_author_id: author.id,
      committer_id: committer.id,
      merge_request_diff_id: merge_request_diff_id,
      relative_order: index,
      sha: Gitlab::Database::ShaAttribute.serialize(raw_sha),
      authored_date: Gitlab::Database.sanitize_timestamp(commit_hash[:authored_date]),
      committed_date: Gitlab::Database.sanitize_timestamp(commit_hash[:committed_date]),
      trailers: Gitlab::Json.dump(trailers)
    )

    # Need to add `raw_sha` to commit_hash as we will use that when
    # inserting the `sha` in `merge_request_commits_metadata` table. We
    # only need to do this when dedup is enabled.
    commit_hash[:raw_sha] = raw_sha if dedup_enabled

    commit_hash[:project_id] = project.id if partition_enabled
    commit_hash = commit_hash.merge(message: '') if skip_commit_data

    commit_hash
  end

  rows = commit_rows_with_metadata(project.id, merge_request_diff_id, rows) if dedup_enabled

  ApplicationRecord.legacy_bulk_insert(self.table_name, rows) # rubocop:disable Gitlab/BulkInsert
end

.oldest_merge_request_id_per_commit(project_id, shas) ⇒ Object 



120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
 
# File 'app/models/merge_request_diff_commit.rb', line 120

def self.oldest_merge_request_id_per_commit(project_id, shas)
  # This method is defined here and not on MergeRequest, otherwise the SHA
  # values used in the WHERE below won't be encoded correctly.
  select(['merge_request_diff_commits.sha AS sha', 'min(merge_requests.id) AS merge_request_id'])
    .joins(:merge_request_diff)
    .joins(
      'INNER JOIN merge_requests ' \
        'ON merge_requests.latest_merge_request_diff_id = merge_request_diffs.id'
    )
    .where(sha: shas)
    .where(
      merge_requests: {
        target_project_id: project_id,
        state_id: MergeRequest.available_states[:merged]
      }
    )
    .group(:sha)
end

.prepare_commits_for_bulk_insert(commits, organization_id) ⇒ Object 



102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
 
# File 'app/models/merge_request_diff_commit.rb', line 102

def self.prepare_commits_for_bulk_insert(commits, organization_id)
  user_triples = Set.new
  hashes = commits.map do |commit|
    hash = commit.to_hash.except(:parent_ids, :referenced_by)

    TRIM_USER_KEYS.each do |key|
      hash[key] = MergeRequest::DiffCommitUser.prepare(hash[key])
    end

    user_triples << [hash[:author_name], hash[:author_email], organization_id]
    user_triples << [hash[:committer_name], hash[:committer_email], organization_id]

    hash
  end

  [hashes, user_triples]
end

Instance Method Details

#author_email ⇒ Object 



195
196
197
 
# File 'app/models/merge_request_diff_commit.rb', line 195

def author_email
  commit_author&.email
end

#author_name ⇒ Object 



191
192
193
 
# File 'app/models/merge_request_diff_commit.rb', line 191

def author_name
  commit_author&.name
end

#authored_date ⇒ Object 



222
223
224
 
# File 'app/models/merge_request_diff_commit.rb', line 222

def authored_date
  has_commit_metadata? ? merge_request_commits_metadata.authored_date : super
end

#commit_author ⇒ Object 



234
235
236
 
# File 'app/models/merge_request_diff_commit.rb', line 234

def commit_author
  has_commit_metadata? ? merge_request_commits_metadata.commit_author : super
end

#committed_date ⇒ Object 



226
227
228
 
# File 'app/models/merge_request_diff_commit.rb', line 226

def committed_date
  has_commit_metadata? ? merge_request_commits_metadata.committed_date : super
end

#committer ⇒ Object 



238
239
240
 
# File 'app/models/merge_request_diff_commit.rb', line 238

def committer
  has_commit_metadata? ? merge_request_commits_metadata.committer : super
end

#committer_email ⇒ Object 



203
204
205
 
# File 'app/models/merge_request_diff_commit.rb', line 203

def committer_email
  committer&.email
end

#committer_name ⇒ Object 



199
200
201
 
# File 'app/models/merge_request_diff_commit.rb', line 199

def committer_name
  committer&.name
end

#message ⇒ Object 



207
208
209
 
# File 'app/models/merge_request_diff_commit.rb', line 207

def message
  fetch_message
end

#project_id ⇒ Object 



218
219
220
 
# File 'app/models/merge_request_diff_commit.rb', line 218

def project_id
  project.id
end

#sha ⇒ Object 



230
231
232
 
# File 'app/models/merge_request_diff_commit.rb', line 230

def sha
  has_commit_metadata? ? merge_request_commits_metadata.sha : super
end

#to_hash ⇒ Object 



211
212
213
214
215
216
 
# File 'app/models/merge_request_diff_commit.rb', line 211

def to_hash
  super(exclude_keys: [:message]).merge({
    'id' => sha,
    message: fetch_message
  })
end