debian-mirror-gitlab/lib/gitlab/git/diff.rb

284 lines
8.2 KiB
Ruby
Raw Normal View History

2017-09-10 17:25:29 +05:30
# Gitaly note: JV: needs RPC for Gitlab::Git::Diff.between.
2017-08-17 22:00:37 +05:30
# Gitlab::Git::Diff is a wrapper around native Rugged::Diff object
module Gitlab
module Git
class Diff
TimeoutError = Class.new(StandardError)
2017-09-10 17:25:29 +05:30
include Gitlab::EncodingHelper
2017-08-17 22:00:37 +05:30
# Diff properties
attr_accessor :old_path, :new_path, :a_mode, :b_mode, :diff
# Stats properties
attr_accessor :new_file, :renamed_file, :deleted_file
2017-09-10 17:25:29 +05:30
alias_method :new_file?, :new_file
alias_method :deleted_file?, :deleted_file
alias_method :renamed_file?, :renamed_file
attr_accessor :expanded
attr_writer :too_large
2017-08-17 22:00:37 +05:30
2017-09-10 17:25:29 +05:30
alias_method :expanded?, :expanded
2017-08-17 22:00:37 +05:30
2017-09-10 17:25:29 +05:30
SERIALIZE_KEYS = %i(diff new_path old_path a_mode b_mode new_file renamed_file deleted_file too_large).freeze
2017-08-17 22:00:37 +05:30
2018-03-17 18:26:18 +05:30
# The maximum size of a diff to display.
SIZE_LIMIT = 100.kilobytes
2017-09-10 17:25:29 +05:30
2018-03-17 18:26:18 +05:30
# The maximum size before a diff is collapsed.
COLLAPSE_LIMIT = 10.kilobytes
2017-09-10 17:25:29 +05:30
2018-03-17 18:26:18 +05:30
class << self
2017-08-17 22:00:37 +05:30
def between(repo, head, base, options = {}, *paths)
straight = options.delete(:straight) || false
common_commit = if straight
base
else
# Only show what is new in the source branch
# compared to the target branch, not the other way
# around. The linex below with merge_base is
# equivalent to diff with three dots (git diff
# branch1...branch2) From the git documentation:
# "git diff A...B" is equivalent to "git diff
# $(git-merge-base A B) B"
2018-03-17 18:26:18 +05:30
repo.merge_base(head, base)
2017-08-17 22:00:37 +05:30
end
options ||= {}
actual_options = filter_diff_options(options)
repo.diff(common_commit, head, actual_options, *paths)
end
# Return a copy of the +options+ hash containing only keys that can be
# passed to Rugged. Allowed options are:
#
# :ignore_whitespace_change ::
# If true, changes in amount of whitespace will be ignored.
#
# :disable_pathspec_match ::
# If true, the given +*paths+ will be applied as exact matches,
# instead of as fnmatch patterns.
#
def filter_diff_options(options, default_options = {})
2017-09-10 17:25:29 +05:30
allowed_options = [:ignore_whitespace_change,
:disable_pathspec_match, :paths,
:max_files, :max_lines, :limits, :expanded]
2017-08-17 22:00:37 +05:30
if default_options
actual_defaults = default_options.dup
actual_defaults.keep_if do |key|
allowed_options.include?(key)
end
else
actual_defaults = {}
end
if options
filtered_opts = options.dup
filtered_opts.keep_if do |key|
allowed_options.include?(key)
end
filtered_opts = actual_defaults.merge(filtered_opts)
else
filtered_opts = actual_defaults
end
filtered_opts
end
2018-03-17 18:26:18 +05:30
# Return a binary diff message like:
#
# "Binary files a/file/path and b/file/path differ\n"
# This is used when we detect that a diff is binary
# using CharlockHolmes when Rugged treats it as text.
def binary_message(old_path, new_path)
"Binary files #{old_path} and #{new_path} differ\n"
end
2017-08-17 22:00:37 +05:30
end
2017-09-10 17:25:29 +05:30
def initialize(raw_diff, expanded: true)
@expanded = expanded
2017-08-17 22:00:37 +05:30
case raw_diff
when Hash
init_from_hash(raw_diff)
2017-09-10 17:25:29 +05:30
prune_diff_if_eligible
2017-08-17 22:00:37 +05:30
when Rugged::Patch, Rugged::Diff::Delta
2017-09-10 17:25:29 +05:30
init_from_rugged(raw_diff)
when Gitlab::GitalyClient::Diff
init_from_gitaly(raw_diff)
prune_diff_if_eligible
when Gitaly::CommitDelta
2017-08-17 22:00:37 +05:30
init_from_gitaly(raw_diff)
when nil
raise "Nil as raw diff passed"
else
raise "Invalid raw diff type: #{raw_diff.class}"
end
end
def to_hash
hash = {}
2017-09-10 17:25:29 +05:30
SERIALIZE_KEYS.each do |key|
2018-03-17 18:26:18 +05:30
hash[key] = send(key) # rubocop:disable GitlabSecurity/PublicSend
2017-08-17 22:00:37 +05:30
end
hash
end
2017-09-10 17:25:29 +05:30
def mode_changed?
a_mode && b_mode && a_mode != b_mode
end
2017-08-17 22:00:37 +05:30
def submodule?
a_mode == '160000' || b_mode == '160000'
end
def line_count
@line_count ||= Util.count_lines(@diff)
end
def too_large?
if @too_large.nil?
2018-03-17 18:26:18 +05:30
@too_large = @diff.bytesize >= SIZE_LIMIT
2017-08-17 22:00:37 +05:30
else
@too_large
end
end
2017-09-10 17:25:29 +05:30
# This is used by `to_hash` and `init_from_hash`.
alias_method :too_large, :too_large?
2017-08-17 22:00:37 +05:30
2017-09-10 17:25:29 +05:30
def too_large!
2017-08-17 22:00:37 +05:30
@diff = ''
@line_count = 0
@too_large = true
end
def collapsed?
return @collapsed if defined?(@collapsed)
2017-09-10 17:25:29 +05:30
2018-03-17 18:26:18 +05:30
@collapsed = !expanded && @diff.bytesize >= COLLAPSE_LIMIT
2017-08-17 22:00:37 +05:30
end
2017-09-10 17:25:29 +05:30
def collapse!
2017-08-17 22:00:37 +05:30
@diff = ''
@line_count = 0
@collapsed = true
end
2018-03-17 18:26:18 +05:30
def json_safe_diff
return @diff unless detect_binary?(@diff)
# the diff is binary, let's make a message for it
Diff.binary_message(@old_path, @new_path)
end
def has_binary_notice?
@diff.start_with?('Binary')
end
2017-08-17 22:00:37 +05:30
private
2017-09-10 17:25:29 +05:30
def init_from_rugged(rugged)
2017-08-17 22:00:37 +05:30
if rugged.is_a?(Rugged::Patch)
2017-09-10 17:25:29 +05:30
init_from_rugged_patch(rugged)
2017-08-17 22:00:37 +05:30
d = rugged.delta
else
d = rugged
end
@new_path = encode!(d.new_file[:path])
@old_path = encode!(d.old_file[:path])
@a_mode = d.old_file[:mode].to_s(8)
@b_mode = d.new_file[:mode].to_s(8)
@new_file = d.added?
@renamed_file = d.renamed?
@deleted_file = d.deleted?
end
2017-09-10 17:25:29 +05:30
def init_from_rugged_patch(patch)
2017-08-17 22:00:37 +05:30
# Don't bother initializing diffs that are too large. If a diff is
# binary we're not going to display anything so we skip the size check.
2017-09-10 17:25:29 +05:30
return if !patch.delta.binary? && prune_large_patch(patch)
2017-08-17 22:00:37 +05:30
@diff = encode!(strip_diff_headers(patch.to_s))
end
def init_from_hash(hash)
raw_diff = hash.symbolize_keys
2017-09-10 17:25:29 +05:30
SERIALIZE_KEYS.each do |key|
2018-03-17 18:26:18 +05:30
send(:"#{key}=", raw_diff[key.to_sym]) # rubocop:disable GitlabSecurity/PublicSend
2017-08-17 22:00:37 +05:30
end
end
2017-09-10 17:25:29 +05:30
def init_from_gitaly(diff)
@diff = encode!(diff.patch) if diff.respond_to?(:patch)
@new_path = encode!(diff.to_path.dup)
@old_path = encode!(diff.from_path.dup)
@a_mode = diff.old_mode.to_s(8)
@b_mode = diff.new_mode.to_s(8)
@new_file = diff.from_id == BLANK_SHA
@renamed_file = diff.from_path != diff.to_path
@deleted_file = diff.to_id == BLANK_SHA
collapse! if diff.respond_to?(:collapsed) && diff.collapsed
2017-08-17 22:00:37 +05:30
end
2017-09-10 17:25:29 +05:30
def prune_diff_if_eligible
if too_large?
too_large!
elsif collapsed?
collapse!
end
2017-08-17 22:00:37 +05:30
end
# If the patch surpasses any of the diff limits it calls the appropiate
# prune method and returns true. Otherwise returns false.
2017-09-10 17:25:29 +05:30
def prune_large_patch(patch)
2017-08-17 22:00:37 +05:30
size = 0
patch.each_hunk do |hunk|
hunk.each_line do |line|
size += line.content.bytesize
2018-03-17 18:26:18 +05:30
if size >= SIZE_LIMIT
2017-09-10 17:25:29 +05:30
too_large!
2018-10-15 14:42:47 +05:30
return true # rubocop:disable Cop/AvoidReturnFromBlocks
2017-08-17 22:00:37 +05:30
end
end
end
2018-03-17 18:26:18 +05:30
if !expanded && size >= COLLAPSE_LIMIT
2017-09-10 17:25:29 +05:30
collapse!
2017-08-17 22:00:37 +05:30
return true
end
false
end
# Strip out the information at the beginning of the patch's text to match
# Grit's output
def strip_diff_headers(diff_text)
# Delete everything up to the first line that starts with '---' or
# 'Binary'
diff_text.sub!(/\A.*?^(---|Binary)/m, '\1')
if diff_text.start_with?('---', 'Binary')
diff_text
else
# If the diff_text did not contain a line starting with '---' or
# 'Binary', return the empty string. No idea why; we are just
# preserving behavior from before the refactor.
''
end
end
end
end
end