Commit a00578ce authored by Robert Speicher's avatar Robert Speicher

Absorb gitlab_git

parent aec04a47
...@@ -16,6 +16,8 @@ gem 'default_value_for', '~> 3.0.0' ...@@ -16,6 +16,8 @@ gem 'default_value_for', '~> 3.0.0'
gem 'mysql2', '~> 0.3.16', group: :mysql gem 'mysql2', '~> 0.3.16', group: :mysql
gem 'pg', '~> 0.18.2', group: :postgres gem 'pg', '~> 0.18.2', group: :postgres
gem 'rugged', '~> 0.24.0'
# Authentication libraries # Authentication libraries
gem 'devise', '~> 4.2' gem 'devise', '~> 4.2'
gem 'doorkeeper', '~> 4.2.0' gem 'doorkeeper', '~> 4.2.0'
...@@ -49,10 +51,6 @@ gem 'u2f', '~> 0.2.1' ...@@ -49,10 +51,6 @@ gem 'u2f', '~> 0.2.1'
# Browser detection # Browser detection
gem 'browser', '~> 2.2' gem 'browser', '~> 2.2'
# Extracting information from a git repository
# Provide access to Gitlab::Git library
gem 'gitlab_git', '~> 10.7.0'
# LDAP Auth # LDAP Auth
# GitLab fork with several improvements to original library. For full list of changes # GitLab fork with several improvements to original library. For full list of changes
# see https://github.com/intridea/omniauth-ldap/compare/master...gitlabhq:master # see https://github.com/intridea/omniauth-ldap/compare/master...gitlabhq:master
......
...@@ -255,11 +255,6 @@ GEM ...@@ -255,11 +255,6 @@ GEM
mime-types (>= 1.16, < 3) mime-types (>= 1.16, < 3)
posix-spawn (~> 0.3) posix-spawn (~> 0.3)
gitlab-markup (1.5.0) gitlab-markup (1.5.0)
gitlab_git (10.7.0)
activesupport (~> 4.0)
charlock_holmes (~> 0.7.3)
github-linguist (~> 4.7.0)
rugged (~> 0.24.0)
gitlab_omniauth-ldap (1.2.1) gitlab_omniauth-ldap (1.2.1)
net-ldap (~> 0.9) net-ldap (~> 0.9)
omniauth (~> 1.0) omniauth (~> 1.0)
...@@ -857,7 +852,6 @@ DEPENDENCIES ...@@ -857,7 +852,6 @@ DEPENDENCIES
github-linguist (~> 4.7.0) github-linguist (~> 4.7.0)
gitlab-flowdock-git-hook (~> 1.0.1) gitlab-flowdock-git-hook (~> 1.0.1)
gitlab-markup (~> 1.5.0) gitlab-markup (~> 1.5.0)
gitlab_git (~> 10.7.0)
gitlab_omniauth-ldap (~> 1.2.1) gitlab_omniauth-ldap (~> 1.2.1)
gollum-lib (~> 4.2) gollum-lib (~> 4.2)
gollum-rugged_adapter (~> 0.4.2) gollum-rugged_adapter (~> 0.4.2)
...@@ -942,6 +936,7 @@ DEPENDENCIES ...@@ -942,6 +936,7 @@ DEPENDENCIES
rubocop-rspec (~> 1.5.0) rubocop-rspec (~> 1.5.0)
ruby-fogbugz (~> 0.2.1) ruby-fogbugz (~> 0.2.1)
ruby-prof (~> 0.16.2) ruby-prof (~> 0.16.2)
rugged (~> 0.24.0)
sanitize (~> 2.0) sanitize (~> 2.0)
sass-rails (~> 5.0.6) sass-rails (~> 5.0.6)
scss_lint (~> 0.47.0) scss_lint (~> 0.47.0)
...@@ -988,4 +983,4 @@ DEPENDENCIES ...@@ -988,4 +983,4 @@ DEPENDENCIES
wikicloth (= 0.8.1) wikicloth (= 0.8.1)
BUNDLED WITH BUNDLED WITH
1.13.6 1.13.7
class MergeRequestDiff < ActiveRecord::Base class MergeRequestDiff < ActiveRecord::Base
include Sortable include Sortable
include Importable include Importable
include EncodingHelper include Gitlab::Git::EncodingHelper
# Prevent store of diff if commits amount more then 500 # Prevent store of diff if commits amount more then 500
COMMITS_SAFE_SIZE = 100 COMMITS_SAFE_SIZE = 100
......
module Gitlab
module Git
# Class for parsing Git attribute files and extracting the attributes for
# file patterns.
#
# Unlike Rugged this parser only needs a single IO call (a call to `open`),
# vastly reducing the time spent in extracting attributes.
#
# This class _only_ supports parsing the attributes file located at
# `$GIT_DIR/info/attributes` as GitLab doesn't use any other files
# (`.gitattributes` is copied to this particular path).
#
# Basic usage:
#
# attributes = Gitlab::Git::Attributes.new(some_repo.path)
#
# attributes.attributes('README.md') # => { "eol" => "lf }
class Attributes
# path - The path to the Git repository.
def initialize(path)
@path = File.expand_path(path)
@patterns = nil
end
# Returns all the Git attributes for the given path.
#
# path - A path to a file for which to get the attributes.
#
# Returns a Hash.
def attributes(path)
full_path = File.join(@path, path)
patterns.each do |pattern, attrs|
return attrs if File.fnmatch?(pattern, full_path)
end
{}
end
# Returns a Hash containing the file patterns and their attributes.
def patterns
@patterns ||= parse_file
end
# Parses an attribute string.
#
# These strings can be in the following formats:
#
# text # => { "text" => true }
# -text # => { "text" => false }
# key=value # => { "key" => "value" }
#
# string - The string to parse.
#
# Returns a Hash containing the attributes and their values.
def parse_attributes(string)
values = {}
dash = '-'
equal = '='
binary = 'binary'
string.split(/\s+/).each do |chunk|
# Data such as "foo = bar" should be treated as "foo" and "bar" being
# separate boolean attributes.
next if chunk == equal
key = chunk
# Input: "-foo"
if chunk.start_with?(dash)
key = chunk.byteslice(1, chunk.length - 1)
value = false
# Input: "foo=bar"
elsif chunk.include?(equal)
key, value = chunk.split(equal, 2)
# Input: "foo"
else
value = true
end
values[key] = value
# When the "binary" option is set the "diff" option should be set to
# the inverse. If "diff" is later set it should overwrite the
# automatically set value.
values['diff'] = false if key == binary && value
end
values
end
# Iterates over every line in the attributes file.
def each_line
full_path = File.join(@path, 'info/attributes')
return unless File.exist?(full_path)
File.open(full_path, 'r') do |handle|
handle.each_line do |line|
break unless line.valid_encoding?
yield line.strip
end
end
end
private
# Parses the Git attributes file.
def parse_file
pairs = []
comment = '#'
each_line do |line|
next if line.start_with?(comment) || line.empty?
pattern, attrs = line.split(/\s+/, 2)
parsed = attrs ? parse_attributes(attrs) : {}
pairs << [File.join(@path, pattern), parsed]
end
# Newer entries take precedence over older entries.
pairs.reverse.to_h
end
end
end
end
require_relative 'encoding_helper'
module Gitlab
module Git
class Blame
include Gitlab::Git::EncodingHelper
attr_reader :lines, :blames
def initialize(repository, sha, path)
@repo = repository
@sha = sha
@path = path
@lines = []
@blames = load_blame
end
def each
@blames.each do |blame|
yield(
Gitlab::Git::Commit.new(blame.commit),
blame.line
)
end
end
private
def load_blame
cmd = %W(git --git-dir=#{@repo.path} blame -p #{@sha} -- #{@path})
# Read in binary mode to ensure ASCII-8BIT
raw_output = IO.popen(cmd, 'rb') {|io| io.read }
output = encode_utf8(raw_output)
process_raw_blame output
end
def process_raw_blame(output)
lines, final = [], []
info, commits = {}, {}
# process the output
output.split("\n").each do |line|
if line[0, 1] == "\t"
lines << line[1, line.size]
elsif m = /^(\w{40}) (\d+) (\d+)/.match(line)
commit_id, old_lineno, lineno = m[1], m[2].to_i, m[3].to_i
commits[commit_id] = nil unless commits.key?(commit_id)
info[lineno] = [commit_id, old_lineno]
end
end
# load all commits in single call
commits.keys.each do |key|
commits[key] = @repo.lookup(key)
end
# get it together
info.sort.each do |lineno, (commit_id, old_lineno)|
commit = commits[commit_id]
final << BlameLine.new(lineno, old_lineno, commit, lines[lineno - 1])
end
@lines = final
end
end
class BlameLine
attr_accessor :lineno, :oldlineno, :commit, :line
def initialize(lineno, oldlineno, commit, line)
@lineno = lineno
@oldlineno = oldlineno
@commit = commit
@line = line
end
end
end
end
This diff is collapsed.
module Gitlab
module Git
class BlobSnippet
include Linguist::BlobHelper
attr_accessor :ref
attr_accessor :lines
attr_accessor :filename
attr_accessor :startline
def initialize(ref, lines, startline, filename)
@ref, @lines, @startline, @filename = ref, lines, startline, filename
end
def data
lines.join("\n") if lines
end
def name
filename
end
def size
data.length
end
def mode
nil
end
end
end
end
module Gitlab
module Git
class Branch < Ref
end
end
end
# Gitlab::Git::Commit is a wrapper around native Rugged::Commit object
module Gitlab
module Git
class Commit
include Gitlab::Git::EncodingHelper
attr_accessor :raw_commit, :head, :refs
SERIALIZE_KEYS = [
:id, :message, :parent_ids,
:authored_date, :author_name, :author_email,
:committed_date, :committer_name, :committer_email
].freeze
attr_accessor *SERIALIZE_KEYS # rubocop:disable Lint/AmbiguousOperator
def ==(other)
return false unless other.is_a?(Gitlab::Git::Commit)
methods = [:message, :parent_ids, :authored_date, :author_name,
:author_email, :committed_date, :committer_name,
:committer_email]
methods.all? do |method|
send(method) == other.send(method)
end
end
class << self
# Get commits collection
#
# Ex.
# Commit.where(
# repo: repo,
# ref: 'master',
# path: 'app/models',
# limit: 10,
# offset: 5,
# )
#
def where(options)
repo = options.delete(:repo)
raise 'Gitlab::Git::Repository is required' unless repo.respond_to?(:log)
repo.log(options).map { |c| decorate(c) }
end
# Get single commit
#
# Ex.
# Commit.find(repo, '29eda46b')
#
# Commit.find(repo, 'master')
#
def find(repo, commit_id = "HEAD")
return decorate(commit_id) if commit_id.is_a?(Rugged::Commit)
obj = if commit_id.is_a?(String)
repo.rev_parse_target(commit_id)
else
Ref.dereference_object(commit_id)
end
return nil unless obj.is_a?(Rugged::Commit)
decorate(obj)
rescue Rugged::ReferenceError, Rugged::InvalidError, Rugged::ObjectError, Gitlab::Git::Repository::NoRepository
nil
end
# Get last commit for HEAD
#
# Ex.
# Commit.last(repo)
#
def last(repo)
find(repo)
end
# Get last commit for specified path and ref
#
# Ex.
# Commit.last_for_path(repo, '29eda46b', 'app/models')
#
# Commit.last_for_path(repo, 'master', 'Gemfile')
#
def last_for_path(repo, ref, path = nil)
where(
repo: repo,
ref: ref,
path: path,
limit: 1
).first
end
# Get commits between two revspecs
# See also #repository.commits_between
#
# Ex.
# Commit.between(repo, '29eda46b', 'master')
#
def between(repo, base, head)
repo.commits_between(base, head).map do |commit|
decorate(commit)
end
rescue Rugged::ReferenceError
[]
end
# Delegate Repository#find_commits
def find_all(repo, options = {})
repo.find_commits(options)
end
def decorate(commit, ref = nil)
Gitlab::Git::Commit.new(commit, ref)
end
# Returns a diff object for the changes introduced by +rugged_commit+.
# If +rugged_commit+ doesn't have a parent, then the diff is between
# this commit and an empty repo. See Repository#diff for the keys
# allowed in the +options+ hash.
def diff_from_parent(rugged_commit, options = {})
options ||= {}
break_rewrites = options[:break_rewrites]
actual_options = Diff.filter_diff_options(options)
diff = if rugged_commit.parents.empty?
rugged_commit.diff(actual_options.merge(reverse: true))
else
rugged_commit.parents[0].diff(rugged_commit, actual_options)
end
diff.find_similar!(break_rewrites: break_rewrites)
diff
end
end
def initialize(raw_commit, head = nil)
raise "Nil as raw commit passed" unless raw_commit
if raw_commit.is_a?(Hash)
init_from_hash(raw_commit)
elsif raw_commit.is_a?(Rugged::Commit)
init_from_rugged(raw_commit)
else
raise "Invalid raw commit type: #{raw_commit.class}"
end
@head = head
end
def sha
id
end
def short_id(length = 10)
id.to_s[0..length]
end
def safe_message
@safe_message ||= message
end
def created_at
committed_date
end
# Was this commit committed by a different person than the original author?
def different_committer?
author_name != committer_name || author_email != committer_email
end
def parent_id
parent_ids.first
end
# Shows the diff between the commit's parent and the commit.
#
# Cuts out the header and stats from #to_patch and returns only the diff.
def to_diff(options = {})
diff_from_parent(options).patch
end
# Returns a diff object for the changes from this commit's first parent.
# If there is no parent, then the diff is between this commit and an
# empty repo. See Repository#diff for keys allowed in the +options+
# hash.
def diff_from_parent(options = {})
Commit.diff_from_parent(raw_commit, options)
end
def has_zero_stats?
stats.total.zero?
rescue
true
end
def no_commit_message
"--no commit message"
end
def to_hash
serialize_keys.map.with_object({}) do |key, hash|
hash[key] = send(key)
end
end
def date
committed_date
end
def diffs(options = {})
DiffCollection.new(diff_from_parent(options), options)
end
def parents
raw_commit.parents.map { |c| Gitlab::Git::Commit.new(c) }
end
def tree
raw_commit.tree
end
def stats
Gitlab::Git::CommitStats.new(self)
end
def to_patch(options = {})
begin
raw_commit.to_mbox(options)
rescue Rugged::InvalidError => ex
if ex.message =~ /Commit \w+ is a merge commit/
'Patch format is not currently supported for merge commits.'
end
end
end
# Get a collection of Rugged::Reference objects for this commit.
#
# Ex.
# commit.ref(repo)
#
def refs(repo)
repo.refs_hash[id]
end
# Get ref names collection
#
# Ex.
# commit.ref_names(repo)
#
def ref_names(repo)
refs(repo).map do |ref|
ref.name.sub(%r{^refs/(heads|remotes|tags)/}, "")
end
end
def message
encode! @message
end
def author_name
encode! @author_name
end
def author_email
encode! @author_email
end
def committer_name
encode! @committer_name
end
def committer_email
encode! @committer_email
end
private
def init_from_hash(hash)
raw_commit = hash.symbolize_keys
serialize_keys.each do |key|
send("#{key}=", raw_commit[key])
end
end
def init_from_rugged(commit)
author = commit.author
committer = commit.committer
@raw_commit = commit
@id = commit.oid
@message = commit.message
@authored_date = author[:time]
@committed_date = committer[:time]
@author_name = author[:name]
@author_email = author[:email]
@committer_name = committer[:name]
@committer_email = committer[:email]
@parent_ids = commit.parents.map(&:oid)
end
def serialize_keys
SERIALIZE_KEYS
end
end
end
end
# Gitlab::Git::CommitStats counts the additions, deletions, and total changes
# in a commit.
module Gitlab
module Git
class CommitStats
attr_reader :id, :additions, :deletions, :total
# Instantiate a CommitStats object
def initialize(commit)
@id = commit.id
@additions = 0
@deletions = 0
@total = 0
diff = commit.diff_from_parent
diff.each_patch do |p|
# TODO: Use the new Rugged convenience methods when they're released
@additions += p.stat[0]
@deletions += p.stat[1]
@total += p.changes
end
end
end
end
end
module Gitlab
module Git
class Compare
attr_reader :head, :base, :straight
def initialize(repository, base, head, straight = false)
@repository = repository
@straight = straight
unless base && head
@commits = []
return
end
@base = Gitlab::Git::Commit.find(repository, base.try(:strip))
@head = Gitlab::Git::Commit.find(repository, head.try(:strip))
@commits = [] unless @base && @head
@commits = [] if same
end
def same
@base && @head && @base.id == @head.id
end
def commits
return @commits if defined?(@commits)
@commits = Gitlab::Git::Commit.between(@repository, @base.id, @head.id)
end
def diffs(options = {})
unless @head && @base
return Gitlab::Git::DiffCollection.new([])
end
paths = options.delete(:paths) || []
options[:straight] = @straight
Gitlab::Git::Diff.between(@repository, @head.id, @base.id, options, *paths)
end
end
end
end
This diff is collapsed.
module Gitlab
module Git
class DiffCollection
include Enumerable
DEFAULT_LIMITS = { max_files: 100, max_lines: 5000 }.freeze
def initialize(iterator, options = {})
@iterator = iterator
@max_files = options.fetch(:max_files, DEFAULT_LIMITS[:max_files])
@max_lines = options.fetch(:max_lines, DEFAULT_LIMITS[:max_lines])
@max_bytes = @max_files * 5120 # Average 5 KB per file
@safe_max_files = [@max_files, DEFAULT_LIMITS[:max_files]].min
@safe_max_lines = [@max_lines, DEFAULT_LIMITS[:max_lines]].min
@safe_max_bytes = @safe_max_files * 5120 # Average 5 KB per file
@all_diffs = !!options.fetch(:all_diffs, false)
@no_collapse = !!options.fetch(:no_collapse, true)
@deltas_only = !!options.fetch(:deltas_only, false)
@line_count = 0
@byte_count = 0
@overflow = false
@array = Array.new
end
def each(&block)
if @populated
# @iterator.each is slower than just iterating the array in place
@array.each(&block)
elsif @deltas_only
each_delta(&block)
else
each_patch(&block)
end
end
def empty?
!@iterator.any?
end
def overflow?
populate!
!!@overflow
end
def size
@size ||= count # forces a loop using each method
end
def real_size
populate!
if @overflow
"#{size}+"
else
size.to_s
end
end
def decorate!
collection = each_with_index do |element, i|
@array[i] = yield(element)
end
@populated = true
collection
end
private
def populate!
return if @populated
each { nil } # force a loop through all diffs
@populated = true
nil
end
def over_safe_limits?(files)
files >= @safe_max_files || @line_count > @safe_max_lines || @byte_count >= @safe_max_bytes
end
def each_delta
@iterator.each_delta.with_index do |delta, i|
diff = Gitlab::Git::Diff.new(delta)
yield @array[i] = diff
end
end
def each_patch
@iterator.each_with_index do |raw, i|
# First yield cached Diff instances from @array
if @array[i]
yield @array[i]
next
end
# We have exhausted @array, time to create new Diff instances or stop.
break if @overflow
if !@all_diffs && i >= @max_files
@overflow = true
break
end
collapse = !@all_diffs && !@no_collapse
diff = Gitlab::Git::Diff.new(raw, collapse: collapse)
if collapse && over_safe_limits?(i)
diff.prune_collapsed_diff!
end
@line_count += diff.line_count
@byte_count += diff.diff.bytesize
if !@all_diffs && (@line_count >= @max_lines || @byte_count >= @max_bytes)
# This last Diff instance pushes us over the lines limit. We stop and
# discard it.
@overflow = true
break
end
yield @array[i] = diff
end
end
end
end
end
module Gitlab
module Git
module EncodingHelper
extend self
# This threshold is carefully tweaked to prevent usage of encodings detected
# by CharlockHolmes with low confidence. If CharlockHolmes confidence is low,
# we're better off sticking with utf8 encoding.
# Reason: git diff can return strings with invalid utf8 byte sequences if it
# truncates a diff in the middle of a multibyte character. In this case
# CharlockHolmes will try to guess the encoding and will likely suggest an
# obscure encoding with low confidence.
# There is a lot more info with this merge request:
# https://gitlab.com/gitlab-org/gitlab_git/merge_requests/77#note_4754193
ENCODING_CONFIDENCE_THRESHOLD = 40
def encode!(message)
return nil unless message.respond_to? :force_encoding
# if message is utf-8 encoding, just return it
message.force_encoding("UTF-8")
return message if message.valid_encoding?
# return message if message type is binary
detect = CharlockHolmes::EncodingDetector.detect(message)
return message.force_encoding("BINARY") if detect && detect[:type] == :binary
# force detected encoding if we have sufficient confidence.
if detect && detect[:encoding] && detect[:confidence] > ENCODING_CONFIDENCE_THRESHOLD
message.force_encoding(detect[:encoding])
end
# encode and clean the bad chars
message.replace clean(message)
rescue
encoding = detect ? detect[:encoding] : "unknown"
"--broken encoding: #{encoding}"
end
def encode_utf8(message)
detect = CharlockHolmes::EncodingDetector.detect(message)
if detect
CharlockHolmes::Converter.convert(message, detect[:encoding], 'UTF-8')
else
clean(message)
end
end
private
def clean(message)
message.encode("UTF-16BE", undef: :replace, invalid: :replace, replace: "")
.encode("UTF-8")
.gsub("\0".encode("UTF-8"), "")
end
end
end
end
module Gitlab
module Git
class PathHelper
class << self
def normalize_path(filename)
# Strip all leading slashes so that //foo -> foo
filename[/^\/*/] = ''
# Expand relative paths (e.g. foo/../bar)
filename = Pathname.new(filename)
filename.relative_path_from(Pathname.new(''))
end
end
end
end
end
require 'open3'
module Gitlab
module Git
module Popen
def popen(cmd, path)
unless cmd.is_a?(Array)
raise "System commands must be given as an array of strings"
end
vars = { "PWD" => path }
options = { chdir: path }
@cmd_output = ""
@cmd_status = 0
Open3.popen3(vars, *cmd, options) do |stdin, stdout, stderr, wait_thr|
@cmd_output << stdout.read
@cmd_output << stderr.read
@cmd_status = wait_thr.value.exitstatus
end
[@cmd_output, @cmd_status]
end
end
end
end
module Gitlab
module Git
class Ref
include Gitlab::Git::EncodingHelper
# Branch or tag name
# without "refs/tags|heads" prefix
attr_reader :name
# Target sha.
# Usually it is commit sha but in case
# when tag reference on other tag it can be tag sha
attr_reader :target
# Dereferenced target
# Commit object to which the Ref points to
attr_reader :dereferenced_target
# Extract branch name from full ref path
#
# Ex.
# Ref.extract_branch_name('refs/heads/master') #=> 'master'
def self.extract_branch_name(str)
str.gsub(/\Arefs\/heads\//, '')
end
def self.dereference_object(object)
object = object.target while object.is_a?(Rugged::Tag::Annotation)
object
end
def initialize(repository, name, target)
encode! name
@name = name.gsub(/\Arefs\/(tags|heads)\//, '')
@dereferenced_target = Commit.find(repository, target)
@target = if target.respond_to?(:oid)
target.oid
elsif target.respond_to?(:name)
target.name
elsif target.is_a? String
target
else
nil
end
end
end
end
end
This diff is collapsed.
module Gitlab
module Git
class Tag < Ref
attr_reader :object_sha
def initialize(repository, name, target, message = nil)
super(repository, name, target)
@message = message
end
def message
encode! @message
end
end
end
end
module Gitlab
module Git
class Tree
include Gitlab::Git::EncodingHelper
attr_accessor :id, :root_id, :name, :path, :type,
:mode, :commit_id, :submodule_url
class << self
# Get list of tree objects
# for repository based on commit sha and path
# Uses rugged for raw objects
def where(repository, sha, path = nil)
path = nil if path == '' || path == '/'
commit = repository.lookup(sha)
root_tree = commit.tree
tree = if path
id = Tree.find_id_by_path(repository, root_tree.oid, path)
if id
repository.lookup(id)
else
[]
end
else
root_tree
end
tree.map do |entry|
Tree.new(
id: entry[:oid],
root_id: root_tree.oid,
name: entry[:name],
type: entry[:type],
mode: entry[:filemode],
path: path ? File.join(path, entry[:name]) : entry[:name],
commit_id: sha,
)
end
end
# Recursive search of tree id for path
#
# Ex.
# blog/ # oid: 1a
# app/ # oid: 2a
# models/ # oid: 3a
# views/ # oid: 4a
#
#
# Tree.find_id_by_path(repo, '1a', 'app/models') # => '3a'
#
def find_id_by_path(repository, root_id, path)
root_tree = repository.lookup(root_id)
path_arr = path.split('/')
entry = root_tree.find do |entry|
entry[:name] == path_arr[0] && entry[:type] == :tree
end
return nil unless entry
if path_arr.size > 1
path_arr.shift
find_id_by_path(repository, entry[:oid], path_arr.join('/'))
else
entry[:oid]
end
end
end
def initialize(options)
%w(id root_id name path type mode commit_id).each do |key|
self.send("#{key}=", options[key.to_sym])
end
end
def name
encode! @name
end
def dir?
type == :tree
end
def file?
type == :blob
end
def submodule?
type == :commit
end
def readme?
name =~ /^readme/i
end
def contributing?
name =~ /^contributing/i
end
end
end
end
module Gitlab
module Git
module Util
LINE_SEP = "\n".freeze
def self.count_lines(string)
case string[-1]
when nil
0
when LINE_SEP
string.count(LINE_SEP)
else
string.count(LINE_SEP) + 1
end
end
end
end
end
require 'spec_helper'
describe Gitlab::Git::Attributes, seed_helper: true do
let(:path) do
File.join(SEED_REPOSITORY_PATH, 'with-git-attributes.git')
end
subject { described_class.new(path) }
describe '#attributes' do
context 'using a path with attributes' do
it 'returns the attributes as a Hash' do
expect(subject.attributes('test.txt')).to eq({ 'text' => true })
end
it 'returns a Hash containing multiple attributes' do
expect(subject.attributes('test.sh')).
to eq({ 'eol' => 'lf', 'gitlab-language' => 'shell' })
end
it 'returns a Hash containing attributes for a file with multiple extensions' do
expect(subject.attributes('test.haml.html')).
to eq({ 'gitlab-language' => 'haml' })
end
it 'returns a Hash containing attributes for a file in a directory' do
expect(subject.attributes('foo/bar.txt')).to eq({ 'foo' => true })
end
it 'returns a Hash containing attributes with query string parameters' do
expect(subject.attributes('foo.cgi')).
to eq({ 'key' => 'value?p1=v1&p2=v2' })
end
it 'returns a Hash containing the attributes for an absolute path' do
expect(subject.attributes('/test.txt')).to eq({ 'text' => true })
end
it 'returns a Hash containing the attributes when a pattern is defined using an absolute path' do
# When a path is given without a leading slash it should still match
# patterns defined with a leading slash.
expect(subject.attributes('foo.png')).
to eq({ 'gitlab-language' => 'png' })
expect(subject.attributes('/foo.png')).
to eq({ 'gitlab-language' => 'png' })
end
it 'returns an empty Hash for a defined path without attributes' do
expect(subject.attributes('bla/bla.txt')).to eq({})
end
context 'when the "binary" option is set for a path' do
it 'returns true for the "binary" option' do
expect(subject.attributes('test.binary')['binary']).to eq(true)
end
it 'returns false for the "diff" option' do
expect(subject.attributes('test.binary')['diff']).to eq(false)
end
end
end
context 'using a path without any attributes' do
it 'returns an empty Hash' do
expect(subject.attributes('test.foo')).to eq({})
end
end
end
describe '#patterns' do
it 'parses a file with entries' do
expect(subject.patterns).to be_an_instance_of(Hash)
end
it 'parses an entry that uses a tab to separate the pattern and attributes' do
expect(subject.patterns[File.join(path, '*.md')]).
to eq({ 'gitlab-language' => 'markdown' })
end
it 'stores patterns in reverse order' do
first = subject.patterns.to_a[0]
expect(first[0]).to eq(File.join(path, 'bla/bla.txt'))
end
# It's a bit hard to test for something _not_ being processed. As such we'll
# just test the number of entries.
it 'ignores any comments and empty lines' do
expect(subject.patterns.length).to eq(10)
end
it 'does not parse anything when the attributes file does not exist' do
expect(File).to receive(:exist?).
with(File.join(path, 'info/attributes')).
and_return(false)
expect(subject.patterns).to eq({})
end
end
describe '#parse_attributes' do
it 'parses a boolean attribute' do
expect(subject.parse_attributes('text')).to eq({ 'text' => true })
end
it 'parses a negated boolean attribute' do
expect(subject.parse_attributes('-text')).to eq({ 'text' => false })
end
it 'parses a key-value pair' do
expect(subject.parse_attributes('foo=bar')).to eq({ 'foo' => 'bar' })
end
it 'parses multiple attributes' do
input = 'boolean key=value -negated'
expect(subject.parse_attributes(input)).
to eq({ 'boolean' => true, 'key' => 'value', 'negated' => false })
end
it 'parses attributes with query string parameters' do
expect(subject.parse_attributes('foo=bar?baz=1')).
to eq({ 'foo' => 'bar?baz=1' })
end
end
describe '#each_line' do
it 'iterates over every line in the attributes file' do
args = [String] * 14 # the number of lines in the file
expect { |b| subject.each_line(&b) }.to yield_successive_args(*args)
end
it 'does not yield when the attributes file does not exist' do
expect(File).to receive(:exist?).
with(File.join(path, 'info/attributes')).
and_return(false)
expect { |b| subject.each_line(&b) }.not_to yield_control
end
it 'does not yield when the attributes file has an unsupported encoding' do
path = File.join(SEED_REPOSITORY_PATH, 'with-invalid-git-attributes.git')
attrs = described_class.new(path)
expect { |b| attrs.each_line(&b) }.not_to yield_control
end
end
end
# coding: utf-8
require "spec_helper"
describe Gitlab::Git::Blame, seed_helper: true do
let(:repository) { Gitlab::Git::Repository.new(TEST_REPO_PATH) }
let(:blame) do
Gitlab::Git::Blame.new(repository, SeedRepo::Commit::ID, "CONTRIBUTING.md")
end
context "each count" do
it do
data = []
blame.each do |commit, line|
data << {
commit: commit,
line: line
}
end
expect(data.size).to eq(95)
expect(data.first[:commit]).to be_kind_of(Gitlab::Git::Commit)
expect(data.first[:line]).to eq("# Contribute to GitLab")
end
end
context "ISO-8859 encoding" do
let(:blame) do
Gitlab::Git::Blame.new(repository, SeedRepo::EncodingCommit::ID, "encoding/iso8859.txt")
end
it 'converts to UTF-8' do
data = []
blame.each do |commit, line|
data << {
commit: commit,
line: line
}
end
expect(data.size).to eq(1)
expect(data.first[:commit]).to be_kind_of(Gitlab::Git::Commit)
expect(data.first[:line]).to eq("Ä ü")
end
end
context "unknown encoding" do
let(:blame) do
Gitlab::Git::Blame.new(repository, SeedRepo::EncodingCommit::ID, "encoding/iso8859.txt")
end
it 'converts to UTF-8' do
expect(CharlockHolmes::EncodingDetector).to receive(:detect).and_return(nil)
data = []
blame.each do |commit, line|
data << {
commit: commit,
line: line
}
end
expect(data.size).to eq(1)
expect(data.first[:commit]).to be_kind_of(Gitlab::Git::Commit)
expect(data.first[:line]).to eq(" ")
end
end
end
# encoding: UTF-8
require "spec_helper"
describe Gitlab::Git::BlobSnippet, seed_helper: true do
describe :data do
context 'empty lines' do
let(:snippet) { Gitlab::Git::BlobSnippet.new('master', nil, nil, nil) }
it { expect(snippet.data).to be_nil }
end
context 'present lines' do
let(:snippet) { Gitlab::Git::BlobSnippet.new('master', ['wow', 'much'], 1, 'wow.rb') }
it { expect(snippet.data).to eq("wow\nmuch") }
end
end
end
This diff is collapsed.
require "spec_helper"
describe Gitlab::Git::Branch, seed_helper: true do
let(:repository) { Gitlab::Git::Repository.new(TEST_REPO_PATH) }
subject { repository.branches }
it { is_expected.to be_kind_of Array }
describe '#size' do
subject { super().size }
it { is_expected.to eq(SeedRepo::Repo::BRANCHES.size) }
end
describe 'first branch' do
let(:branch) { repository.branches.first }
it { expect(branch.name).to eq(SeedRepo::Repo::BRANCHES.first) }
it { expect(branch.dereferenced_target.sha).to eq("0b4bc9a49b562e85de7cc9e834518ea6828729b9") }
end
describe 'master branch' do
let(:branch) do
repository.branches.find { |branch| branch.name == 'master' }
end
it { expect(branch.dereferenced_target.sha).to eq(SeedRepo::LastCommit::ID) }
end
it { expect(repository.branches.size).to eq(SeedRepo::Repo::BRANCHES.size) }
end
This diff is collapsed.
require "spec_helper"
describe Gitlab::Git::Compare, seed_helper: true do
let(:repository) { Gitlab::Git::Repository.new(TEST_REPO_PATH) }
let(:compare) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, SeedRepo::Commit::ID, false) }
let(:compare_straight) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, SeedRepo::Commit::ID, true) }
describe :commits do
subject do
compare.commits.map(&:id)
end
it 'has 8 elements' do
expect(subject.size).to eq(8)
end
it { is_expected.to include(SeedRepo::Commit::PARENT_ID) }
it { is_expected.not_to include(SeedRepo::BigCommit::PARENT_ID) }
context 'non-existing base ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, 'no-such-branch', SeedRepo::Commit::ID) }
it { is_expected.to be_empty }
end
context 'non-existing head ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, '1234567890') }
it { is_expected.to be_empty }
end
context 'base ref is equal to head ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, SeedRepo::BigCommit::ID) }
it { is_expected.to be_empty }
end
context 'providing nil as base ref or head ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, nil, nil) }
it { is_expected.to be_empty }
end
end
describe :diffs do
subject do
compare.diffs.map(&:new_path)
end
it 'has 10 elements' do
expect(subject.size).to eq(10)
end
it { is_expected.to include('files/ruby/popen.rb') }
it { is_expected.not_to include('LICENSE') }
context 'non-existing base ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, 'no-such-branch', SeedRepo::Commit::ID) }
it { is_expected.to be_empty }
end
context 'non-existing head ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, '1234567890') }
it { is_expected.to be_empty }
end
end
describe :same do
subject do
compare.same
end
it { is_expected.to eq(false) }
context 'base ref is equal to head ref' do
let(:compare) { Gitlab::Git::Compare.new(repository, SeedRepo::BigCommit::ID, SeedRepo::BigCommit::ID) }
it { is_expected.to eq(true) }
end
end
describe :commits_straight do
subject do
compare_straight.commits.map(&:id)
end
it 'has 8 elements' do
expect(subject.size).to eq(8)
end
it { is_expected.to include(SeedRepo::Commit::PARENT_ID) }
it { is_expected.not_to include(SeedRepo::BigCommit::PARENT_ID) }
end
describe :diffs_straight do
subject do
compare_straight.diffs.map(&:new_path)
end
it 'has 10 elements' do
expect(subject.size).to eq(10)
end
it { is_expected.to include('files/ruby/popen.rb') }
it { is_expected.not_to include('LICENSE') }
end
end
This diff is collapsed.
This diff is collapsed.
require "spec_helper"
describe Gitlab::Git::EncodingHelper do
let(:ext_class) { Class.new { extend EncodingHelper } }
let(:binary_string) { File.join(SEED_REPOSITORY_PATH, 'gitlab_logo.png') }
describe '#encode!' do
[
[
'leaves ascii only string as is',
'ascii only string',
'ascii only string'
],
[
'leaves valid utf8 string as is',
'multibyte string №∑∉',
'multibyte string №∑∉'
],
[
'removes invalid bytes from ASCII-8bit encoded multibyte string. This can occur when a git diff match line truncates in the middle of a multibyte character. This occurs after the second word in this example. The test string is as short as we can get while still triggering the error condition when not looking at `detect[:confidence]`.',
"mu ns\xC3\n Lorem ipsum dolor sit amet, consectetur adipisicing ut\xC3\xA0y\xC3\xB9abcd\xC3\xB9efg kia elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non p\n {: .normal_pn}\n \n-Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in\n# *Lorem ipsum\xC3\xB9l\xC3\xB9l\xC3\xA0 dolor\xC3\xB9k\xC3\xB9 sit\xC3\xA8b\xC3\xA8 N\xC3\xA8 amet b\xC3\xA0d\xC3\xAC*\n+# *consectetur\xC3\xB9l\xC3\xB9l\xC3\xA0 adipisicing\xC3\xB9k\xC3\xB9 elit\xC3\xA8b\xC3\xA8 N\xC3\xA8 sed do\xC3\xA0d\xC3\xAC*{: .italic .smcaps}\n \n \xEF\x9B\xA1 eiusmod tempor incididunt, ut\xC3\xAAn\xC3\xB9 labore et dolore. Tw\xC4\x83nj\xC3\xAC magna aliqua. Ut enim ad minim veniam\n {: .normal}\n@@ -9,5 +9,5 @@ quis nostrud\xC3\xAAt\xC3\xB9 exercitiation ullamco laboris m\xC3\xB9s\xC3\xB9k\xC3\xB9abc\xC3\xB9 nisi ".force_encoding('ASCII-8BIT'),
"mu ns\n Lorem ipsum dolor sit amet, consectetur adipisicing ut\xC3\xA0y\xC3\xB9abcd\xC3\xB9efg kia elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non p\n {: .normal_pn}\n \n-Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in\n# *Lorem ipsum\xC3\xB9l\xC3\xB9l\xC3\xA0 dolor\xC3\xB9k\xC3\xB9 sit\xC3\xA8b\xC3\xA8 N\xC3\xA8 amet b\xC3\xA0d\xC3\xAC*\n+# *consectetur\xC3\xB9l\xC3\xB9l\xC3\xA0 adipisicing\xC3\xB9k\xC3\xB9 elit\xC3\xA8b\xC3\xA8 N\xC3\xA8 sed do\xC3\xA0d\xC3\xAC*{: .italic .smcaps}\n \n \xEF\x9B\xA1 eiusmod tempor incididunt, ut\xC3\xAAn\xC3\xB9 labore et dolore. Tw\xC4\x83nj\xC3\xAC magna aliqua. Ut enim ad minim veniam\n {: .normal}\n@@ -9,5 +9,5 @@ quis nostrud\xC3\xAAt\xC3\xB9 exercitiation ullamco laboris m\xC3\xB9s\xC3\xB9k\xC3\xB9abc\xC3\xB9 nisi ",
],
].each do |description, test_string, xpect|
it description do
expect(ext_class.encode!(test_string)).to eq(xpect)
end
end
it 'leaves binary string as is' do
expect(ext_class.encode!(binary_string)).to eq(binary_string)
end
end
describe '#encode_utf8' do
[
[
"encodes valid utf8 encoded string to utf8",
"λ, λ, λ".encode("UTF-8"),
"λ, λ, λ".encode("UTF-8"),
],
[
"encodes valid ASCII-8BIT encoded string to utf8",
"ascii only".encode("ASCII-8BIT"),
"ascii only".encode("UTF-8"),
],
[
"encodes valid ISO-8859-1 encoded string to utf8",
"Rüby ist eine Programmiersprache. Wir verlängern den text damit ICU die Sprache erkennen kann.".encode("ISO-8859-1", "UTF-8"),
"Rüby ist eine Programmiersprache. Wir verlängern den text damit ICU die Sprache erkennen kann.".encode("UTF-8"),
],
].each do |description, test_string, xpect|
it description do
r = ext_class.encode_utf8(test_string.force_encoding('UTF-8'))
expect(r).to eq(xpect)
expect(r.encoding.name).to eq('UTF-8')
end
end
end
describe '#clean' do
[
[
'leaves ascii only string as is',
'ascii only string',
'ascii only string'
],
[
'leaves valid utf8 string as is',
'multibyte string №∑∉',
'multibyte string №∑∉'
],
[
'removes invalid bytes from ASCII-8bit encoded multibyte string.',
"Lorem ipsum\xC3\n dolor sit amet, xy\xC3\xA0y\xC3\xB9abcd\xC3\xB9efg".force_encoding('ASCII-8BIT'),
"Lorem ipsum\n dolor sit amet, xyàyùabcdùefg",
],
].each do |description, test_string, xpect|
it description do
expect(ext_class.encode!(test_string)).to eq(xpect)
end
end
end
end
This diff is collapsed.
require "spec_helper"
describe Gitlab::Git::Tag, seed_helper: true do
let(:repository) { Gitlab::Git::Repository.new(TEST_REPO_PATH) }
describe 'first tag' do
let(:tag) { repository.tags.first }
it { expect(tag.name).to eq("v1.0.0") }
it { expect(tag.target).to eq("f4e6814c3e4e7a0de82a9e7cd20c626cc963a2f8") }
it { expect(tag.dereferenced_target.sha).to eq("6f6d7e7ed97bb5f0054f2b1df789b39ca89b6ff9") }
it { expect(tag.message).to eq("Release") }
end
describe 'last tag' do
let(:tag) { repository.tags.last }
it { expect(tag.name).to eq("v1.2.1") }
it { expect(tag.target).to eq("2ac1f24e253e08135507d0830508febaaccf02ee") }
it { expect(tag.dereferenced_target.sha).to eq("fa1b1e6c004a68b7d8763b86455da9e6b23e36d6") }
it { expect(tag.message).to eq("Version 1.2.1") }
end
it { expect(repository.tags.size).to eq(SeedRepo::Repo::TAGS.size) }
end
require "spec_helper"
describe Gitlab::Git::Tree, seed_helper: true do
context :repo do
let(:repository) { Gitlab::Git::Repository.new(TEST_REPO_PATH) }
let(:tree) { Gitlab::Git::Tree.where(repository, SeedRepo::Commit::ID) }
it { expect(tree).to be_kind_of Array }
it { expect(tree.empty?).to be_falsey }
it { expect(tree.select(&:dir?).size).to eq(2) }
it { expect(tree.select(&:file?).size).to eq(10) }
it { expect(tree.select(&:submodule?).size).to eq(2) }
describe :dir do
let(:dir) { tree.select(&:dir?).first }
it { expect(dir).to be_kind_of Gitlab::Git::Tree }
it { expect(dir.id).to eq('3c122d2b7830eca25235131070602575cf8b41a1') }
it { expect(dir.commit_id).to eq(SeedRepo::Commit::ID) }
it { expect(dir.name).to eq('encoding') }
it { expect(dir.path).to eq('encoding') }
context :subdir do
let(:subdir) { Gitlab::Git::Tree.where(repository, SeedRepo::Commit::ID, 'files').first }
it { expect(subdir).to be_kind_of Gitlab::Git::Tree }
it { expect(subdir.id).to eq('a1e8f8d745cc87e3a9248358d9352bb7f9a0aeba') }
it { expect(subdir.commit_id).to eq(SeedRepo::Commit::ID) }
it { expect(subdir.name).to eq('html') }
it { expect(subdir.path).to eq('files/html') }
end
context :subdir_file do
let(:subdir_file) { Gitlab::Git::Tree.where(repository, SeedRepo::Commit::ID, 'files/ruby').first }
it { expect(subdir_file).to be_kind_of Gitlab::Git::Tree }
it { expect(subdir_file.id).to eq('7e3e39ebb9b2bf433b4ad17313770fbe4051649c') }
it { expect(subdir_file.commit_id).to eq(SeedRepo::Commit::ID) }
it { expect(subdir_file.name).to eq('popen.rb') }
it { expect(subdir_file.path).to eq('files/ruby/popen.rb') }
end
end
describe :file do
let(:file) { tree.select(&:file?).first }
it { expect(file).to be_kind_of Gitlab::Git::Tree }
it { expect(file.id).to eq('dfaa3f97ca337e20154a98ac9d0be76ddd1fcc82') }
it { expect(file.commit_id).to eq(SeedRepo::Commit::ID) }
it { expect(file.name).to eq('.gitignore') }
end
describe :readme do
let(:file) { tree.select(&:readme?).first }
it { expect(file).to be_kind_of Gitlab::Git::Tree }
it { expect(file.name).to eq('README.md') }
end
describe :contributing do
let(:file) { tree.select(&:contributing?).first }
it { expect(file).to be_kind_of Gitlab::Git::Tree }
it { expect(file.name).to eq('CONTRIBUTING.md') }
end
describe :submodule do
let(:submodule) { tree.select(&:submodule?).first }
it { expect(submodule).to be_kind_of Gitlab::Git::Tree }
it { expect(submodule.id).to eq('79bceae69cb5750d6567b223597999bfa91cb3b9') }
it { expect(submodule.commit_id).to eq('570e7b2abdd848b95f2f578043fc23bd6f6fd24d') }
it { expect(submodule.name).to eq('gitlab-shell') }
end
end
end
require 'spec_helper'
describe Gitlab::Git::Util do
describe :count_lines do
[
["", 0],
["foo", 1],
["foo\n", 1],
["foo\n\n", 2],
].each do |string, line_count|
it "counts #{line_count} lines in #{string.inspect}" do
expect(Gitlab::Git::Util.count_lines(string)).to eq(line_count)
end
end
end
end
RSpec::Matchers.define :be_valid_commit do
match do |actual|
actual &&
actual.id == SeedRepo::Commit::ID &&
actual.message == SeedRepo::Commit::MESSAGE &&
actual.author_name == SeedRepo::Commit::AUTHOR_FULL_NAME
end
end
This diff is collapsed.
This diff is collapsed.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment