Class: Gitlab::Git::Blob

Inherits:
Object
  • Object
show all
Includes:
EncodingHelper, Linguist::BlobHelper
Defined in:
lib/gitlab_git/blob.rb

Constant Summary collapse

MAX_DATA_DISPLAY_SIZE =

This number is the maximum amount of data that we want to display to the user. We load as much as we can for encoding detection (Linguist) and LFS pointer parsing. All other cases where we need full blob data should use load_all_data!.

10485760

Constants included from EncodingHelper

EncodingHelper::ENCODING_CONFIDENCE_THRESHOLD

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Methods included from EncodingHelper

#encode!, #encode_utf8

Constructor Details

#initialize(options) ⇒ Blob

Returns a new instance of Blob.



253
254
255
256
257
258
259
260
261
# File 'lib/gitlab_git/blob.rb', line 253

def initialize(options)
  %w(id name path size data mode commit_id binary).each do |key|
    self.send("#{key}=", options[key.to_sym])
  end

  @loaded_all_data = false
  # Retain the actual size before it is encoded
  @loaded_size = @data.bytesize if @data
end

Instance Attribute Details

#binaryObject

Returns the value of attribute binary.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def binary
  @binary
end

#commit_idObject

Returns the value of attribute commit_id.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def commit_id
  @commit_id
end

#dataObject

Returns the value of attribute data.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def data
  @data
end

#idObject

Returns the value of attribute id.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def id
  @id
end

#loaded_sizeObject

Returns the value of attribute loaded_size.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def loaded_size
  @loaded_size
end

#modeObject

Returns the value of attribute mode.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def mode
  @mode
end

#nameObject

Returns the value of attribute name.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def name
  @name
end

#pathObject

Returns the value of attribute path.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def path
  @path
end

#sizeObject

Returns the value of attribute size.



16
17
18
# File 'lib/gitlab_git/blob.rb', line 16

def size
  @size
end

Class Method Details

.commit(repository, options, action = :add) ⇒ Object

Commit file in repository and return commit sha

options should contain next structure:

file: {
  content: 'Lorem ipsum...',
  path: 'documents/story.txt',
  update: true
},
author: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
committer: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
commit: {
  message: 'Wow such commit',
  branch: 'master',
  update_ref: false
}


124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
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
190
191
192
193
194
195
196
# File 'lib/gitlab_git/blob.rb', line 124

def commit(repository, options, action = :add)
  file = options[:file]
  update = file[:update].nil? ? true : file[:update]
  author = options[:author]
  committer = options[:committer]
  commit = options[:commit]
  repo = repository.rugged
  ref = commit[:branch]
  update_ref = commit[:update_ref].nil? ? true : commit[:update_ref]
  parents = []
  mode = 0o100644

  unless ref.start_with?('refs/')
    ref = 'refs/heads/' + ref
  end

  path_name = PathHelper.normalize_path(file[:path])
  # Abort if any invalid characters remain (e.g. ../foo)
  raise Repository::InvalidBlobName.new("Invalid path") if path_name.each_filename.to_a.include?('..')

  filename = path_name.to_s
  index = repo.index

  unless repo.empty?
    rugged_ref = repo.references[ref]
    raise Repository::InvalidRef.new("Invalid branch name") unless rugged_ref
    last_commit = rugged_ref.target
    index.read_tree(last_commit.tree)
    parents = [last_commit]
  end

  if action == :remove
    index.remove(filename)
  else
    file_entry = index.get(filename)

    if action == :rename
      old_path_name = PathHelper.normalize_path(file[:previous_path])
      old_filename = old_path_name.to_s
      file_entry = index.get(old_filename)
      index.remove(old_filename) unless file_entry.blank?
    end

    if file_entry
      raise Repository::InvalidBlobName.new("Filename already exists; update not allowed") unless update

      # Preserve the current file mode if one is available
      mode = file_entry[:mode] if file_entry[:mode]
    end

    content = file[:content]
    detect = CharlockHolmes::EncodingDetector.new.detect(content) if content

    unless detect && detect[:type] == :binary
      # When writing to the repo directly as we are doing here,
      # the `core.autocrlf` config isn't taken into account.
      content.gsub!("\r\n", "\n") if repository.autocrlf
    end

    oid = repo.write(content, :blob)
    index.add(path: filename, oid: oid, mode: mode)
  end

  opts = {}
  opts[:tree] = index.write_tree(repo)
  opts[:author] = author
  opts[:committer] = committer
  opts[:message] = commit[:message]
  opts[:parents] = parents
  opts[:update_ref] = ref if update_ref

  Rugged::Commit.create(repo, opts)
end

.find(repository, sha, path) ⇒ Object



19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
# File 'lib/gitlab_git/blob.rb', line 19

def find(repository, sha, path)
  commit = repository.lookup(sha)
  root_tree = commit.tree

  blob_entry = find_entry_by_path(repository, root_tree.oid, path)

  return nil unless blob_entry

  if blob_entry[:type] == :commit
    submodule_blob(blob_entry, path, sha)
  else
    blob = repository.lookup(blob_entry[:oid])

    if blob
      Blob.new(
        id: blob.oid,
        name: blob_entry[:name],
        size: blob.size,
        data: blob.content(MAX_DATA_DISPLAY_SIZE),
        mode: blob_entry[:filemode].to_s(8),
        path: path,
        commit_id: sha,
        binary: blob.binary?
      )
    end
  end
end

.find_entry_by_path(repository, root_id, path) ⇒ Object

Recursive search of blob id by path

Ex.

blog/            # oid: 1a
  app/           # oid: 2a
    models/      # oid: 3a
    file.rb      # oid: 4a

Blob.find_entry_by_path(repo, ‘1a’, ‘app/file.rb’) # => ‘4a’



69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
# File 'lib/gitlab_git/blob.rb', line 69

def find_entry_by_path(repository, root_id, path)
  root_tree = repository.lookup(root_id)
  # Strip leading slashes
  path[/^\/*/] = ''
  path_arr = path.split('/')

  entry = root_tree.find do |entry|
    entry[:name] == path_arr[0]
  end

  return nil unless entry

  if path_arr.size > 1
    return nil unless entry[:type] == :tree
    path_arr.shift
    find_entry_by_path(repository, entry[:oid], path_arr.join('/'))
  else
    [:blob, :commit].include?(entry[:type]) ? entry : nil
  end
end

.raw(repository, sha) ⇒ Object



47
48
49
50
51
52
53
54
55
56
# File 'lib/gitlab_git/blob.rb', line 47

def raw(repository, sha)
  blob = repository.lookup(sha)

  Blob.new(
    id: blob.oid,
    size: blob.size,
    data: blob.content(MAX_DATA_DISPLAY_SIZE),
    binary: blob.binary?
  )
end

.remove(repository, options) ⇒ Object

Remove file from repository and return commit sha

options should contain next structure:

file: {
  path: 'documents/story.txt'
},
author: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
committer: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
commit: {
  message: 'Remove FILENAME',
  branch: 'master'
}


219
220
221
# File 'lib/gitlab_git/blob.rb', line 219

def remove(repository, options)
  commit(repository, options, :remove)
end

.rename(repository, options) ⇒ Object

Rename file from repository and return commit sha

options should contain next structure:

file: {
  previous_path: 'documents/old_story.txt'
  path: 'documents/story.txt'
  content: 'Lorem ipsum...',
  update: true
},
author: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
committer: {
  email: '[email protected]',
  name: 'Test User',
  time: Time.now
},
commit: {
  message: 'Rename FILENAME',
  branch: 'master'
}


248
249
250
# File 'lib/gitlab_git/blob.rb', line 248

def rename(repository, options)
  commit(repository, options, :rename)
end

.submodule_blob(blob_entry, path, sha) ⇒ Object



90
91
92
93
94
95
96
97
98
# File 'lib/gitlab_git/blob.rb', line 90

def submodule_blob(blob_entry, path, sha)
  Blob.new(
    id: blob_entry[:oid],
    name: blob_entry[:name],
    data: '',
    path: path,
    commit_id: sha,
  )
end

Instance Method Details

#binary?Boolean

Returns:

  • (Boolean)


263
264
265
# File 'lib/gitlab_git/blob.rb', line 263

def binary?
  @binary.nil? ? super : @binary == true
end

#empty?Boolean

Returns:

  • (Boolean)


267
268
269
# File 'lib/gitlab_git/blob.rb', line 267

def empty?
  !data || data == ''
end

#lfs_oidObject



299
300
301
302
303
304
305
306
# File 'lib/gitlab_git/blob.rb', line 299

def lfs_oid
  if has_lfs_version_key?
    oid = data.match(/(?<=sha256:)([0-9a-f]{64})/)
    return oid[1] if oid
  end

  nil
end

#lfs_pointer?Boolean

Valid LFS object pointer is a text file consisting of version oid size see github.com/github/git-lfs/blob/v1.1.0/docs/spec.md#the-pointer

Returns:

  • (Boolean)


295
296
297
# File 'lib/gitlab_git/blob.rb', line 295

def lfs_pointer?
  has_lfs_version_key? && lfs_oid.present? && lfs_size.present?
end

#lfs_sizeObject



308
309
310
311
312
313
314
315
# File 'lib/gitlab_git/blob.rb', line 308

def lfs_size
  if has_lfs_version_key?
    size = data.match(/(?<=size )([0-9]+)/)
    return size[1] if size
  end

  nil
end

#load_all_data!(repository) ⇒ Object

Load all blob data (not just the first MAX_DATA_DISPLAY_SIZE bytes) into memory as a Ruby string.



277
278
279
280
281
282
283
284
# File 'lib/gitlab_git/blob.rb', line 277

def load_all_data!(repository)
  return if @data == '' # don't mess with submodule blobs
  return @data if @loaded_all_data

  @loaded_all_data = true
  @data = repository.lookup(id).content
  @loaded_size = @data.bytesize
end

#truncated?Boolean

Returns:

  • (Boolean)


317
318
319
# File 'lib/gitlab_git/blob.rb', line 317

def truncated?
  size && (size > loaded_size)
end