merge_request_diff.rb 6.01 KB
Newer Older
1
class MergeRequestDiff < ActiveRecord::Base
2
  include Sortable
3
  include Importable
James Lopez's avatar
James Lopez committed
4
  include EncodingHelper
5

6
  # Prevent store of diff if commits amount more then 500
7
  COMMITS_SAFE_SIZE = 100
8 9 10

  belongs_to :merge_request

11 12
  delegate :source_branch_sha, :target_branch_sha,
    :target_branch, :source_branch, to: :merge_request, prefix: nil
13 14 15

  state_machine :state, initial: :empty do
    state :collected
16 17 18
    state :overflow
    # Deprecated states: these are no longer used but these values may still occur
    # in the database.
19 20 21 22 23 24 25 26 27
    state :timeout
    state :overflow_commits_safe_size
    state :overflow_diff_files_limit
    state :overflow_diff_lines_limit
  end

  serialize :st_commits
  serialize :st_diffs

28
  after_initialize :set_diff_range
29
  after_create :reload_content, unless: :importing?
30
  after_save :keep_around_commits, unless: :importing?
31

32 33 34 35 36 37
  def set_diff_range
    self.start_commit_sha ||= target_branch_sha
    self.head_commit_sha  ||= source_branch_sha
    self.base_commit_sha  ||= branch_base_sha
  end

38 39 40 41 42
  def reload_content
    reload_commits
    reload_diffs
  end

43 44
  def size
    real_size.presence || diffs.size
45 46
  end

47 48 49 50
  def diffs(options={})
    if options[:ignore_whitespace_change]
      @diffs_no_whitespace ||= begin
        compare = Gitlab::Git::Compare.new(
51
          repository.raw_repository,
52 53
          start_commit_sha,
          head_commit_sha
54 55 56 57
        )
        compare.diffs(options)
      end
    else
58
      @diffs ||= {}
59
      @diffs[options] ||= load_diffs(st_diffs, options)
60
    end
61 62
  end

63 64 65 66 67 68 69 70
  def commits
    @commits ||= load_commits(st_commits || [])
  end

  def last_commit
    commits.first
  end

71 72 73 74
  def first_commit
    commits.last
  end

75
  def base_commit
76
    return unless base_commit_sha
77

78
    project.commit(base_commit_sha)
79 80
  end

81
  def start_commit
82
    return unless start_commit_sha
83

84
    project.commit(start_commit_sha)
85 86 87
  end

  def head_commit
88
    return last_commit unless head_commit_sha
89

90
    project.commit(head_commit_sha)
91 92
  end

93 94 95 96 97 98 99 100 101 102
  def diff_refs
    return unless start_commit || base_commit

    Gitlab::Diff::DiffRefs.new(
      base_sha:  base_commit_sha,
      start_sha: start_commit_sha,
      head_sha:  head_commit_sha
    )
  end

103
  private
104

105 106 107 108 109
  def compare
    @compare ||=
      begin
        # Update ref for merge request
        merge_request.fetch_ref
110

111 112 113 114 115 116
        Gitlab::Git::Compare.new(
          repository.raw_repository,
          start_commit_sha,
          head_commit_sha
        )
      end
117 118
  end

119 120 121 122 123 124 125 126
  def dump_commits(commits)
    commits.map(&:to_hash)
  end

  def load_commits(array)
    array.map { |hash| Commit.new(Gitlab::Git::Commit.new(hash), merge_request.source_project) }
  end

127 128 129
  # Reload all commits related to current merge request from repo
  # and save it as array of hashes in st_commits db field
  def reload_commits
130 131
    new_attributes = {}

132
    commits = compare.commits
133

134 135 136
    if commits.present?
      commits = Commit.decorate(commits, merge_request.source_project).reverse
      new_attributes[:st_commits] = dump_commits(commits)
137 138
    end

139
    update_columns_serialized(new_attributes)
140 141
  end

142 143 144 145 146 147 148 149
  def dump_diffs(diffs)
    if diffs.respond_to?(:map)
      diffs.map(&:to_hash)
    end
  end

  def load_diffs(raw, options)
    if raw.respond_to?(:each)
150
      if paths = options[:paths]
151
        raw = raw.select do |diff|
152
          paths.include?(diff[:old_path]) || paths.include?(diff[:new_path])
153 154 155
        end
      end

156 157 158 159 160 161
      Gitlab::Git::DiffCollection.new(raw, options)
    else
      Gitlab::Git::DiffCollection.new([])
    end
  end

162 163 164
  # Reload diffs between branches related to current merge request from repo
  # and save it as array of hashes in st_diffs db field
  def reload_diffs
165
    new_attributes = {}
166 167 168
    new_diffs = []

    if commits.size.zero?
169
      new_attributes[:state] = :empty
170
    else
171
      diff_collection = compare.diffs(Commit.max_diff_options)
172

173 174 175
      if diff_collection.overflow?
        # Set our state to 'overflow' to make the #empty? and #collected?
        # methods (generated by StateMachine) return false.
176
        new_attributes[:state] = :overflow
177 178
      end

179
      new_attributes[:real_size] = diff_collection.real_size
180

181 182
      if diff_collection.any?
        new_diffs = dump_diffs(diff_collection)
183
        new_attributes[:state] = :collected
184
      end
185
    end
186

187
    new_attributes[:st_diffs] = new_diffs
188

189
    update_columns_serialized(new_attributes)
190 191

    keep_around_commits
192 193
  end

194 195
  def project
    merge_request.target_project
196 197 198
  end

  def repository
199
    project.repository
200
  end
201

202
  def branch_base_commit
203
    return unless source_branch_sha && target_branch_sha
204

205
    project.merge_base_commit(source_branch_sha, target_branch_sha)
206 207
  end

208 209
  def branch_base_sha
    branch_base_commit.try(:sha)
210 211
  end

James Lopez's avatar
James Lopez committed
212 213 214 215 216 217 218 219
  def utf8_st_diffs
    st_diffs.map do |diff|
      diff.each do |k, v|
        diff[k] = encode_utf8(v) if v.respond_to?(:encoding)
      end
    end
  end

220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241
  #
  # #save or #update_attributes providing changes on serialized attributes do a lot of
  # serialization and deserialization calls resulting in bad performance.
  # Using #update_columns solves the problem with just one YAML.dump per serialized attribute that we provide.
  # As a tradeoff we need to reload the current instance to properly manage time objects on those serialized
  # attributes. So to keep the same behaviour as the attribute assignment we reload the instance.
  # The difference is in the usage of
  # #write_attribute= (#update_attributes) and #raw_write_attribute= (#update_columns)
  #
  # Ex:
  #
  #   new_attributes[:st_commits].first.slice(:committed_date)
  #   => {:committed_date=>2014-02-27 11:01:38 +0200}
  #   YAML.load(YAML.dump(new_attributes[:st_commits].first.slice(:committed_date)))
  #   => {:committed_date=>2014-02-27 10:01:38 +0100}
  #
  def update_columns_serialized(new_attributes)
    return unless new_attributes.any?

    update_columns(new_attributes.merge(updated_at: current_time_from_proper_timezone))
    reload
  end
242

243
  def keep_around_commits
244 245 246
    repository.keep_around(start_commit_sha)
    repository.keep_around(head_commit_sha)
    repository.keep_around(base_commit_sha)
247
  end
248
end