summaryrefslogtreecommitdiff
path: root/lib/banzai
diff options
context:
space:
mode:
Diffstat (limited to 'lib/banzai')
-rw-r--r--lib/banzai/filter/abstract_reference_filter.rb12
-rw-r--r--lib/banzai/filter/commit_range_reference_filter.rb20
-rw-r--r--lib/banzai/filter/commit_reference_filter.rb20
-rw-r--r--lib/banzai/filter/external_issue_reference_filter.rb14
-rw-r--r--lib/banzai/filter/inline_diff_filter.rb12
-rw-r--r--lib/banzai/filter/issue_reference_filter.rb10
-rw-r--r--lib/banzai/filter/label_reference_filter.rb2
-rw-r--r--lib/banzai/filter/merge_request_reference_filter.rb2
-rw-r--r--lib/banzai/filter/milestone_reference_filter.rb2
-rw-r--r--lib/banzai/filter/redactor_filter.rb31
-rw-r--r--lib/banzai/filter/reference_filter.rb31
-rw-r--r--lib/banzai/filter/reference_gatherer_filter.rb65
-rw-r--r--lib/banzai/filter/snippet_reference_filter.rb2
-rw-r--r--lib/banzai/filter/user_reference_filter.rb73
-rw-r--r--lib/banzai/lazy_reference.rb25
-rw-r--r--lib/banzai/pipeline/reference_extraction_pipeline.rb11
-rw-r--r--lib/banzai/reference_extractor.rb48
-rw-r--r--lib/banzai/reference_parser.rb14
-rw-r--r--lib/banzai/reference_parser/base_parser.rb204
-rw-r--r--lib/banzai/reference_parser/commit_parser.rb34
-rw-r--r--lib/banzai/reference_parser/commit_range_parser.rb38
-rw-r--r--lib/banzai/reference_parser/external_issue_parser.rb25
-rw-r--r--lib/banzai/reference_parser/issue_parser.rb40
-rw-r--r--lib/banzai/reference_parser/label_parser.rb11
-rw-r--r--lib/banzai/reference_parser/merge_request_parser.rb11
-rw-r--r--lib/banzai/reference_parser/milestone_parser.rb11
-rw-r--r--lib/banzai/reference_parser/snippet_parser.rb11
-rw-r--r--lib/banzai/reference_parser/user_parser.rb92
28 files changed, 599 insertions, 272 deletions
diff --git a/lib/banzai/filter/abstract_reference_filter.rb b/lib/banzai/filter/abstract_reference_filter.rb
index b8962379cb5..db95d7c908b 100644
--- a/lib/banzai/filter/abstract_reference_filter.rb
+++ b/lib/banzai/filter/abstract_reference_filter.rb
@@ -18,10 +18,6 @@ module Banzai
@object_sym ||= object_name.to_sym
end
- def self.data_reference
- @data_reference ||= "data-#{object_name.dasherize}"
- end
-
def self.object_class_title
@object_title ||= object_class.name.titleize
end
@@ -45,10 +41,6 @@ module Banzai
end
end
- def self.referenced_by(node)
- { object_sym => LazyReference.new(object_class, node.attr(data_reference)) }
- end
-
def object_class
self.class.object_class
end
@@ -236,7 +228,9 @@ module Banzai
if cache.key?(key)
cache[key]
else
- cache[key] = yield
+ value = yield
+ cache[key] = value if key.present?
+ value
end
end
end
diff --git a/lib/banzai/filter/commit_range_reference_filter.rb b/lib/banzai/filter/commit_range_reference_filter.rb
index b469ea0f626..bbb88c979cc 100644
--- a/lib/banzai/filter/commit_range_reference_filter.rb
+++ b/lib/banzai/filter/commit_range_reference_filter.rb
@@ -4,6 +4,8 @@ module Banzai
#
# This filter supports cross-project references.
class CommitRangeReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :commit_range
+
def self.object_class
CommitRange
end
@@ -14,34 +16,18 @@ module Banzai
end
end
- def self.referenced_by(node)
- project = Project.find(node.attr("data-project")) rescue nil
- return unless project
-
- id = node.attr("data-commit-range")
- range = find_object(project, id)
-
- return unless range
-
- { commit_range: range }
- end
-
def initialize(*args)
super
@commit_map = {}
end
- def self.find_object(project, id)
+ def find_object(project, id)
range = CommitRange.new(id, project)
range.valid_commits? ? range : nil
end
- def find_object(*args)
- self.class.find_object(*args)
- end
-
def url_for_object(range, project)
h = Gitlab::Routing.url_helpers
h.namespace_project_compare_url(project.namespace, project,
diff --git a/lib/banzai/filter/commit_reference_filter.rb b/lib/banzai/filter/commit_reference_filter.rb
index bd88207326c..2ce1816672b 100644
--- a/lib/banzai/filter/commit_reference_filter.rb
+++ b/lib/banzai/filter/commit_reference_filter.rb
@@ -4,6 +4,8 @@ module Banzai
#
# This filter supports cross-project references.
class CommitReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :commit
+
def self.object_class
Commit
end
@@ -14,28 +16,12 @@ module Banzai
end
end
- def self.referenced_by(node)
- project = Project.find(node.attr("data-project")) rescue nil
- return unless project
-
- id = node.attr("data-commit")
- commit = find_object(project, id)
-
- return unless commit
-
- { commit: commit }
- end
-
- def self.find_object(project, id)
+ def find_object(project, id)
if project && project.valid_repo?
project.commit(id)
end
end
- def find_object(*args)
- self.class.find_object(*args)
- end
-
def url_for_object(commit, project)
h = Gitlab::Routing.url_helpers
h.namespace_project_commit_url(project.namespace, project, commit,
diff --git a/lib/banzai/filter/external_issue_reference_filter.rb b/lib/banzai/filter/external_issue_reference_filter.rb
index 37344b90576..eaa702952cc 100644
--- a/lib/banzai/filter/external_issue_reference_filter.rb
+++ b/lib/banzai/filter/external_issue_reference_filter.rb
@@ -4,6 +4,8 @@ module Banzai
# References are ignored if the project doesn't use an external issue
# tracker.
class ExternalIssueReferenceFilter < ReferenceFilter
+ self.reference_type = :external_issue
+
# Public: Find `JIRA-123` issue references in text
#
# ExternalIssueReferenceFilter.references_in(text) do |match, issue|
@@ -21,18 +23,6 @@ module Banzai
end
end
- def self.referenced_by(node)
- project = Project.find(node.attr("data-project")) rescue nil
- return unless project
-
- id = node.attr("data-external-issue")
- external_issue = ExternalIssue.new(id, project)
-
- return unless external_issue
-
- { external_issue: external_issue }
- end
-
def call
# Early return if the project isn't using an external tracker
return doc if project.nil? || default_issues_tracker?
diff --git a/lib/banzai/filter/inline_diff_filter.rb b/lib/banzai/filter/inline_diff_filter.rb
index 9e75edd4d4c..beb21b19ab3 100644
--- a/lib/banzai/filter/inline_diff_filter.rb
+++ b/lib/banzai/filter/inline_diff_filter.rb
@@ -8,15 +8,19 @@ module Banzai
next if has_ancestor?(node, IGNORED_ANCESTOR_TAGS)
content = node.to_html
- content = content.gsub(/(?:\[\-(.*?)\-\]|\{\-(.*?)\-\})/, '<span class="idiff left right deletion">\1\2</span>')
- content = content.gsub(/(?:\[\+(.*?)\+\]|\{\+(.*?)\+\})/, '<span class="idiff left right addition">\1\2</span>')
+ html_content = inline_diff_filter(content)
- next if html == content
+ next if content == html_content
- node.replace(content)
+ node.replace(html_content)
end
doc
end
+
+ def inline_diff_filter(text)
+ html_content = text.gsub(/(?:\[\-(.*?)\-\]|\{\-(.*?)\-\})/, '<span class="idiff left right deletion">\1\2</span>')
+ html_content.gsub(/(?:\[\+(.*?)\+\]|\{\+(.*?)\+\})/, '<span class="idiff left right addition">\1\2</span>')
+ end
end
end
end
diff --git a/lib/banzai/filter/issue_reference_filter.rb b/lib/banzai/filter/issue_reference_filter.rb
index 59c5e89c546..2496e704002 100644
--- a/lib/banzai/filter/issue_reference_filter.rb
+++ b/lib/banzai/filter/issue_reference_filter.rb
@@ -5,18 +5,12 @@ module Banzai
#
# This filter supports cross-project references.
class IssueReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :issue
+
def self.object_class
Issue
end
- def self.user_can_see_reference?(user, node, context)
- # It is not possible to check access rights for external issue trackers
- return true if context[:project].try(:external_issue_tracker)
-
- issue = Issue.find(node.attr('data-issue')) rescue nil
- Ability.abilities.allowed?(user, :read_issue, issue)
- end
-
def find_object(project, id)
project.get_issue(id)
end
diff --git a/lib/banzai/filter/label_reference_filter.rb b/lib/banzai/filter/label_reference_filter.rb
index 8488a493b55..e4d3f87d0aa 100644
--- a/lib/banzai/filter/label_reference_filter.rb
+++ b/lib/banzai/filter/label_reference_filter.rb
@@ -2,6 +2,8 @@ module Banzai
module Filter
# HTML filter that replaces label references with links.
class LabelReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :label
+
def self.object_class
Label
end
diff --git a/lib/banzai/filter/merge_request_reference_filter.rb b/lib/banzai/filter/merge_request_reference_filter.rb
index cad38a51851..ac5216d9cfb 100644
--- a/lib/banzai/filter/merge_request_reference_filter.rb
+++ b/lib/banzai/filter/merge_request_reference_filter.rb
@@ -5,6 +5,8 @@ module Banzai
#
# This filter supports cross-project references.
class MergeRequestReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :merge_request
+
def self.object_class
MergeRequest
end
diff --git a/lib/banzai/filter/milestone_reference_filter.rb b/lib/banzai/filter/milestone_reference_filter.rb
index dad0768f51b..ca686c87d97 100644
--- a/lib/banzai/filter/milestone_reference_filter.rb
+++ b/lib/banzai/filter/milestone_reference_filter.rb
@@ -2,6 +2,8 @@ module Banzai
module Filter
# HTML filter that replaces milestone references with links.
class MilestoneReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :milestone
+
def self.object_class
Milestone
end
diff --git a/lib/banzai/filter/redactor_filter.rb b/lib/banzai/filter/redactor_filter.rb
index e589b5df6ec..c753a84a20d 100644
--- a/lib/banzai/filter/redactor_filter.rb
+++ b/lib/banzai/filter/redactor_filter.rb
@@ -7,8 +7,11 @@ module Banzai
#
class RedactorFilter < HTML::Pipeline::Filter
def call
- Querying.css(doc, 'a.gfm').each do |node|
- unless user_can_see_reference?(node)
+ nodes = Querying.css(doc, 'a.gfm[data-reference-type]')
+ visible = nodes_visible_to_user(nodes)
+
+ nodes.each do |node|
+ unless visible.include?(node)
# The reference should be replaced by the original text,
# which is not always the same as the rendered text.
text = node.attr('data-original') || node.text
@@ -21,20 +24,30 @@ module Banzai
private
- def user_can_see_reference?(node)
- if node.has_attribute?('data-reference-filter')
- reference_type = node.attr('data-reference-filter')
- reference_filter = Banzai::Filter.const_get(reference_type)
+ def nodes_visible_to_user(nodes)
+ per_type = Hash.new { |h, k| h[k] = [] }
+ visible = Set.new
+
+ nodes.each do |node|
+ per_type[node.attr('data-reference-type')] << node
+ end
+
+ per_type.each do |type, nodes|
+ parser = Banzai::ReferenceParser[type].new(project, current_user)
- reference_filter.user_can_see_reference?(current_user, node, context)
- else
- true
+ visible.merge(parser.nodes_visible_to_user(current_user, nodes))
end
+
+ visible
end
def current_user
context[:current_user]
end
+
+ def project
+ context[:project]
+ end
end
end
end
diff --git a/lib/banzai/filter/reference_filter.rb b/lib/banzai/filter/reference_filter.rb
index 31386cf851c..2d6f34c9cd8 100644
--- a/lib/banzai/filter/reference_filter.rb
+++ b/lib/banzai/filter/reference_filter.rb
@@ -8,24 +8,8 @@ module Banzai
# :project (required) - Current project, ignored if reference is cross-project.
# :only_path - Generate path-only links.
class ReferenceFilter < HTML::Pipeline::Filter
- def self.user_can_see_reference?(user, node, context)
- if node.has_attribute?('data-project')
- project_id = node.attr('data-project').to_i
- return true if project_id == context[:project].try(:id)
-
- project = Project.find(project_id) rescue nil
- Ability.abilities.allowed?(user, :read_project, project)
- else
- true
- end
- end
-
- def self.user_can_reference?(user, node, context)
- true
- end
-
- def self.referenced_by(node)
- raise NotImplementedError, "#{self} does not implement #{__method__}"
+ class << self
+ attr_accessor :reference_type
end
# Returns a data attribute String to attach to a reference link
@@ -43,7 +27,9 @@ module Banzai
#
# Returns a String
def data_attribute(attributes = {})
- attributes[:reference_filter] = self.class.name.demodulize
+ attributes = attributes.reject { |_, v| v.nil? }
+
+ attributes[:reference_type] = self.class.reference_type
attributes.delete(:original) if context[:no_original_data]
attributes.map { |key, value| %Q(data-#{key.to_s.dasherize}="#{escape_once(value)}") }.join(" ")
end
@@ -82,6 +68,8 @@ module Banzai
# by `ignore_ancestor_query`. Link tags are not processed if they have a
# "gfm" class or the "href" attribute is empty.
def each_node
+ return to_enum(__method__) unless block_given?
+
query = %Q{descendant-or-self::text()[not(#{ignore_ancestor_query})]
| descendant-or-self::a[
not(contains(concat(" ", @class, " "), " gfm ")) and not(@href = "")
@@ -92,6 +80,11 @@ module Banzai
end
end
+ # Returns an Array containing all HTML nodes.
+ def nodes
+ @nodes ||= each_node.to_a
+ end
+
# Yields the link's URL and text whenever the node is a valid <a> tag.
def yield_valid_link(node)
link = CGI.unescape(node.attr('href').to_s)
diff --git a/lib/banzai/filter/reference_gatherer_filter.rb b/lib/banzai/filter/reference_gatherer_filter.rb
deleted file mode 100644
index 96fdb06304e..00000000000
--- a/lib/banzai/filter/reference_gatherer_filter.rb
+++ /dev/null
@@ -1,65 +0,0 @@
-module Banzai
- module Filter
- # HTML filter that gathers all referenced records that the current user has
- # permission to view.
- #
- # Expected to be run in its own post-processing pipeline.
- #
- class ReferenceGathererFilter < HTML::Pipeline::Filter
- def initialize(*)
- super
-
- result[:references] ||= Hash.new { |hash, type| hash[type] = [] }
- end
-
- def call
- Querying.css(doc, 'a.gfm').each do |node|
- gather_references(node)
- end
-
- load_lazy_references unless ReferenceExtractor.lazy?
-
- doc
- end
-
- private
-
- def gather_references(node)
- return unless node.has_attribute?('data-reference-filter')
-
- reference_type = node.attr('data-reference-filter')
- reference_filter = Banzai::Filter.const_get(reference_type)
-
- return if context[:reference_filter] && reference_filter != context[:reference_filter]
-
- return if author && !reference_filter.user_can_reference?(author, node, context)
-
- return unless reference_filter.user_can_see_reference?(current_user, node, context)
-
- references = reference_filter.referenced_by(node)
- return unless references
-
- references.each do |type, values|
- Array.wrap(values).each do |value|
- result[:references][type] << value
- end
- end
- end
-
- def load_lazy_references
- refs = result[:references]
- refs.each do |type, values|
- refs[type] = ReferenceExtractor.lazily(values)
- end
- end
-
- def current_user
- context[:current_user]
- end
-
- def author
- context[:author]
- end
- end
- end
-end
diff --git a/lib/banzai/filter/snippet_reference_filter.rb b/lib/banzai/filter/snippet_reference_filter.rb
index d507eb5ebe1..212a0bbf2a0 100644
--- a/lib/banzai/filter/snippet_reference_filter.rb
+++ b/lib/banzai/filter/snippet_reference_filter.rb
@@ -5,6 +5,8 @@ module Banzai
#
# This filter supports cross-project references.
class SnippetReferenceFilter < AbstractReferenceFilter
+ self.reference_type = :snippet
+
def self.object_class
Snippet
end
diff --git a/lib/banzai/filter/user_reference_filter.rb b/lib/banzai/filter/user_reference_filter.rb
index eea3af842b6..5b0a6d8541b 100644
--- a/lib/banzai/filter/user_reference_filter.rb
+++ b/lib/banzai/filter/user_reference_filter.rb
@@ -4,6 +4,8 @@ module Banzai
#
# A special `@all` reference is also supported.
class UserReferenceFilter < ReferenceFilter
+ self.reference_type = :user
+
# Public: Find `@user` user references in text
#
# UserReferenceFilter.references_in(text) do |match, username|
@@ -21,50 +23,13 @@ module Banzai
end
end
- def self.referenced_by(node)
- if node.has_attribute?('data-group')
- group = Group.find(node.attr('data-group')) rescue nil
- return unless group
-
- { user: group.users }
- elsif node.has_attribute?('data-user')
- { user: LazyReference.new(User, node.attr('data-user')) }
- elsif node.has_attribute?('data-project')
- project = Project.find(node.attr('data-project')) rescue nil
- return unless project
-
- { user: project.team.members.flatten }
- end
- end
-
- def self.user_can_see_reference?(user, node, context)
- if node.has_attribute?('data-group')
- group = Group.find(node.attr('data-group')) rescue nil
- Ability.abilities.allowed?(user, :read_group, group)
- else
- super
- end
- end
-
- def self.user_can_reference?(user, node, context)
- # Only team members can reference `@all`
- if node.has_attribute?('data-project')
- project = Project.find(node.attr('data-project')) rescue nil
- return false unless project
-
- user && project.team.member?(user)
- else
- super
- end
- end
-
def call
return doc if project.nil?
ref_pattern = User.reference_pattern
ref_pattern_start = /\A#{ref_pattern}\z/
- each_node do |node|
+ nodes.each do |node|
if text_node?(node)
replace_text_when_pattern_matches(node, ref_pattern) do |content|
user_link_filter(content)
@@ -94,7 +59,7 @@ module Banzai
self.class.references_in(text) do |match, username|
if username == 'all'
link_to_all(link_text: link_text)
- elsif namespace = Namespace.find_by(path: username)
+ elsif namespace = namespaces[username]
link_to_namespace(namespace, link_text: link_text) || match
else
match
@@ -102,6 +67,31 @@ module Banzai
end
end
+ # Returns a Hash containing all Namespace objects for the username
+ # references in the current document.
+ #
+ # The keys of this Hash are the namespace paths, the values the
+ # corresponding Namespace objects.
+ def namespaces
+ @namespaces ||=
+ Namespace.where(path: usernames).each_with_object({}) do |row, hash|
+ hash[row.path] = row
+ end
+ end
+
+ # Returns all usernames referenced in the current document.
+ def usernames
+ refs = Set.new
+
+ nodes.each do |node|
+ node.to_html.scan(User.reference_pattern) do
+ refs << $~[:user]
+ end
+ end
+
+ refs.to_a
+ end
+
private
def urls
@@ -114,9 +104,12 @@ module Banzai
def link_to_all(link_text: nil)
project = context[:project]
+ author = context[:author]
+
url = urls.namespace_project_url(project.namespace, project,
only_path: context[:only_path])
- data = data_attribute(project: project.id)
+
+ data = data_attribute(project: project.id, author: author.try(:id))
text = link_text || User.reference_prefix + 'all'
link_tag(url, data, text)
diff --git a/lib/banzai/lazy_reference.rb b/lib/banzai/lazy_reference.rb
deleted file mode 100644
index 1095b4debc7..00000000000
--- a/lib/banzai/lazy_reference.rb
+++ /dev/null
@@ -1,25 +0,0 @@
-module Banzai
- class LazyReference
- def self.load(refs)
- lazy_references, values = refs.partition { |ref| ref.is_a?(self) }
-
- lazy_values = lazy_references.group_by(&:klass).flat_map do |klass, refs|
- ids = refs.flat_map(&:ids)
- klass.where(id: ids)
- end
-
- values + lazy_values
- end
-
- attr_reader :klass, :ids
-
- def initialize(klass, ids)
- @klass = klass
- @ids = Array.wrap(ids).map(&:to_i)
- end
-
- def load
- self.klass.where(id: self.ids)
- end
- end
-end
diff --git a/lib/banzai/pipeline/reference_extraction_pipeline.rb b/lib/banzai/pipeline/reference_extraction_pipeline.rb
deleted file mode 100644
index 919998380e4..00000000000
--- a/lib/banzai/pipeline/reference_extraction_pipeline.rb
+++ /dev/null
@@ -1,11 +0,0 @@
-module Banzai
- module Pipeline
- class ReferenceExtractionPipeline < BasePipeline
- def self.filters
- FilterArray[
- Filter::ReferenceGathererFilter
- ]
- end
- end
- end
-end
diff --git a/lib/banzai/reference_extractor.rb b/lib/banzai/reference_extractor.rb
index f4079538ec5..bf366962aef 100644
--- a/lib/banzai/reference_extractor.rb
+++ b/lib/banzai/reference_extractor.rb
@@ -1,28 +1,6 @@
module Banzai
# Extract possible GFM references from an arbitrary String for further processing.
class ReferenceExtractor
- class << self
- LAZY_KEY = :banzai_reference_extractor_lazy
-
- def lazy?
- Thread.current[LAZY_KEY]
- end
-
- def lazily(values = nil, &block)
- return (values || block.call).uniq if lazy?
-
- begin
- Thread.current[LAZY_KEY] = true
-
- values ||= block.call
-
- Banzai::LazyReference.load(values.uniq).uniq
- ensure
- Thread.current[LAZY_KEY] = false
- end
- end
- end
-
def initialize
@texts = []
end
@@ -31,23 +9,21 @@ module Banzai
@texts << Renderer.render(text, context)
end
- def references(type, context = {})
- filter = Banzai::Filter["#{type}_reference"]
+ def references(type, project, current_user = nil)
+ processor = Banzai::ReferenceParser[type].
+ new(project, current_user)
+
+ processor.process(html_documents)
+ end
- context.merge!(
- pipeline: :reference_extraction,
+ private
- # ReferenceGathererFilter
- reference_filter: filter
- )
+ def html_documents
+ # This ensures that we don't memoize anything until we have a number of
+ # text blobs to parse.
+ return [] if @texts.empty?
- self.class.lazily do
- @texts.flat_map do |html|
- text_context = context.dup
- result = Renderer.render_result(html, text_context)
- result[:references][type]
- end.uniq
- end
+ @html_documents ||= @texts.map { |html| Nokogiri::HTML.fragment(html) }
end
end
end
diff --git a/lib/banzai/reference_parser.rb b/lib/banzai/reference_parser.rb
new file mode 100644
index 00000000000..557bec4316e
--- /dev/null
+++ b/lib/banzai/reference_parser.rb
@@ -0,0 +1,14 @@
+module Banzai
+ module ReferenceParser
+ # Returns the reference parser class for the given type
+ #
+ # Example:
+ #
+ # Banzai::ReferenceParser['issue']
+ #
+ # This would return the `Banzai::ReferenceParser::IssueParser` class.
+ def self.[](name)
+ const_get("#{name.to_s.camelize}Parser")
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/base_parser.rb b/lib/banzai/reference_parser/base_parser.rb
new file mode 100644
index 00000000000..3d7b9c4a024
--- /dev/null
+++ b/lib/banzai/reference_parser/base_parser.rb
@@ -0,0 +1,204 @@
+module Banzai
+ module ReferenceParser
+ # Base class for reference parsing classes.
+ #
+ # Each parser should also specify its reference type by calling
+ # `self.reference_type = ...` in the body of the class. The value of this
+ # method should be a symbol such as `:issue` or `:merge_request`. For
+ # example:
+ #
+ # class IssueParser < BaseParser
+ # self.reference_type = :issue
+ # end
+ #
+ # The reference type is used to determine what nodes to pass to the
+ # `referenced_by` method.
+ #
+ # Parser classes should either implement the instance method
+ # `references_relation` or overwrite `referenced_by`. The
+ # `references_relation` method is supposed to return an
+ # ActiveRecord::Relation used as a base relation for retrieving the objects
+ # referenced in a set of HTML nodes.
+ #
+ # Each class can implement two additional methods:
+ #
+ # * `nodes_user_can_reference`: returns an Array of nodes the given user can
+ # refer to.
+ # * `nodes_visible_to_user`: returns an Array of nodes that are visible to
+ # the given user.
+ #
+ # You only need to overwrite these methods if you want to tweak who can see
+ # which references. For example, the IssueParser class defines its own
+ # `nodes_visible_to_user` method so it can ensure users can only see issues
+ # they have access to.
+ class BaseParser
+ class << self
+ attr_accessor :reference_type
+ end
+
+ # Returns the attribute name containing the value for every object to be
+ # parsed by the current parser.
+ #
+ # For example, for a parser class that returns "Animal" objects this
+ # attribute would be "data-animal".
+ def self.data_attribute
+ @data_attribute ||= "data-#{reference_type.to_s.dasherize}"
+ end
+
+ def initialize(project = nil, current_user = nil)
+ @project = project
+ @current_user = current_user
+ end
+
+ # Returns all the nodes containing references that the user can refer to.
+ def nodes_user_can_reference(user, nodes)
+ nodes
+ end
+
+ # Returns all the nodes that are visible to the given user.
+ def nodes_visible_to_user(user, nodes)
+ projects = lazy { projects_for_nodes(nodes) }
+ project_attr = 'data-project'
+
+ nodes.select do |node|
+ if node.has_attribute?(project_attr)
+ node_id = node.attr(project_attr).to_i
+
+ if project && project.id == node_id
+ true
+ else
+ can?(user, :read_project, projects[node_id])
+ end
+ else
+ true
+ end
+ end
+ end
+
+ # Returns an Array of objects referenced by any of the given HTML nodes.
+ def referenced_by(nodes)
+ ids = unique_attribute_values(nodes, self.class.data_attribute)
+
+ references_relation.where(id: ids)
+ end
+
+ # Returns the ActiveRecord::Relation to use for querying references in the
+ # DB.
+ def references_relation
+ raise NotImplementedError,
+ "#{self.class} does not implement #{__method__}"
+ end
+
+ # Returns a Hash containing attribute values per project ID.
+ #
+ # The returned Hash uses the following format:
+ #
+ # { project id => [value1, value2, ...] }
+ #
+ # nodes - An Array of HTML nodes to process.
+ # attribute - The name of the attribute (as a String) for which to gather
+ # values.
+ #
+ # Returns a Hash.
+ def gather_attributes_per_project(nodes, attribute)
+ per_project = Hash.new { |hash, key| hash[key] = Set.new }
+
+ nodes.each do |node|
+ project_id = node.attr('data-project').to_i
+ id = node.attr(attribute)
+
+ per_project[project_id] << id if id
+ end
+
+ per_project
+ end
+
+ # Returns a Hash containing objects for an attribute grouped per their
+ # IDs.
+ #
+ # The returned Hash uses the following format:
+ #
+ # { id value => row }
+ #
+ # nodes - An Array of HTML nodes to process.
+ #
+ # collection - The model or ActiveRecord relation to use for retrieving
+ # rows from the database.
+ #
+ # attribute - The name of the attribute containing the primary key values
+ # for every row.
+ #
+ # Returns a Hash.
+ def grouped_objects_for_nodes(nodes, collection, attribute)
+ return {} if nodes.empty?
+
+ ids = unique_attribute_values(nodes, attribute)
+
+ collection.where(id: ids).each_with_object({}) do |row, hash|
+ hash[row.id] = row
+ end
+ end
+
+ # Returns an Array containing all unique values of an attribute of the
+ # given nodes.
+ def unique_attribute_values(nodes, attribute)
+ values = Set.new
+
+ nodes.each do |node|
+ if node.has_attribute?(attribute)
+ values << node.attr(attribute)
+ end
+ end
+
+ values.to_a
+ end
+
+ # Processes the list of HTML documents and returns an Array containing all
+ # the references.
+ def process(documents)
+ type = self.class.reference_type
+
+ nodes = documents.flat_map do |document|
+ Querying.css(document, "a[data-reference-type='#{type}'].gfm").to_a
+ end
+
+ gather_references(nodes)
+ end
+
+ # Gathers the references for the given HTML nodes.
+ def gather_references(nodes)
+ nodes = nodes_user_can_reference(current_user, nodes)
+ nodes = nodes_visible_to_user(current_user, nodes)
+
+ referenced_by(nodes)
+ end
+
+ # Returns a Hash containing the projects for a given list of HTML nodes.
+ #
+ # The returned Hash uses the following format:
+ #
+ # { project ID => project }
+ #
+ def projects_for_nodes(nodes)
+ @projects_for_nodes ||=
+ grouped_objects_for_nodes(nodes, Project, 'data-project')
+ end
+
+ def can?(user, permission, subject)
+ Ability.abilities.allowed?(user, permission, subject)
+ end
+
+ def find_projects_for_hash_keys(hash)
+ Project.where(id: hash.keys)
+ end
+
+ private
+
+ attr_reader :current_user, :project
+
+ def lazy(&block)
+ Gitlab::Lazy.new(&block)
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/commit_parser.rb b/lib/banzai/reference_parser/commit_parser.rb
new file mode 100644
index 00000000000..0fee9d267de
--- /dev/null
+++ b/lib/banzai/reference_parser/commit_parser.rb
@@ -0,0 +1,34 @@
+module Banzai
+ module ReferenceParser
+ class CommitParser < BaseParser
+ self.reference_type = :commit
+
+ def referenced_by(nodes)
+ commit_ids = commit_ids_per_project(nodes)
+ projects = find_projects_for_hash_keys(commit_ids)
+
+ projects.flat_map do |project|
+ find_commits(project, commit_ids[project.id])
+ end
+ end
+
+ def commit_ids_per_project(nodes)
+ gather_attributes_per_project(nodes, self.class.data_attribute)
+ end
+
+ def find_commits(project, ids)
+ commits = []
+
+ return commits unless project.valid_repo?
+
+ ids.each do |id|
+ commit = project.commit(id)
+
+ commits << commit if commit
+ end
+
+ commits
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/commit_range_parser.rb b/lib/banzai/reference_parser/commit_range_parser.rb
new file mode 100644
index 00000000000..69d01f8db15
--- /dev/null
+++ b/lib/banzai/reference_parser/commit_range_parser.rb
@@ -0,0 +1,38 @@
+module Banzai
+ module ReferenceParser
+ class CommitRangeParser < BaseParser
+ self.reference_type = :commit_range
+
+ def referenced_by(nodes)
+ range_ids = commit_range_ids_per_project(nodes)
+ projects = find_projects_for_hash_keys(range_ids)
+
+ projects.flat_map do |project|
+ find_ranges(project, range_ids[project.id])
+ end
+ end
+
+ def commit_range_ids_per_project(nodes)
+ gather_attributes_per_project(nodes, self.class.data_attribute)
+ end
+
+ def find_ranges(project, range_ids)
+ ranges = []
+
+ range_ids.each do |id|
+ range = find_object(project, id)
+
+ ranges << range if range
+ end
+
+ ranges
+ end
+
+ def find_object(project, id)
+ range = CommitRange.new(id, project)
+
+ range.valid_commits? ? range : nil
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/external_issue_parser.rb b/lib/banzai/reference_parser/external_issue_parser.rb
new file mode 100644
index 00000000000..a1264db2111
--- /dev/null
+++ b/lib/banzai/reference_parser/external_issue_parser.rb
@@ -0,0 +1,25 @@
+module Banzai
+ module ReferenceParser
+ class ExternalIssueParser < BaseParser
+ self.reference_type = :external_issue
+
+ def referenced_by(nodes)
+ issue_ids = issue_ids_per_project(nodes)
+ projects = find_projects_for_hash_keys(issue_ids)
+ issues = []
+
+ projects.each do |project|
+ issue_ids[project.id].each do |id|
+ issues << ExternalIssue.new(id, project)
+ end
+ end
+
+ issues
+ end
+
+ def issue_ids_per_project(nodes)
+ gather_attributes_per_project(nodes, self.class.data_attribute)
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/issue_parser.rb b/lib/banzai/reference_parser/issue_parser.rb
new file mode 100644
index 00000000000..24076e3d9ec
--- /dev/null
+++ b/lib/banzai/reference_parser/issue_parser.rb
@@ -0,0 +1,40 @@
+module Banzai
+ module ReferenceParser
+ class IssueParser < BaseParser
+ self.reference_type = :issue
+
+ def nodes_visible_to_user(user, nodes)
+ # It is not possible to check access rights for external issue trackers
+ return nodes if project && project.external_issue_tracker
+
+ issues = issues_for_nodes(nodes)
+
+ nodes.select do |node|
+ issue = issue_for_node(issues, node)
+
+ issue ? can?(user, :read_issue, issue) : false
+ end
+ end
+
+ def referenced_by(nodes)
+ issues = issues_for_nodes(nodes)
+
+ nodes.map { |node| issue_for_node(issues, node) }.uniq
+ end
+
+ def issues_for_nodes(nodes)
+ @issues_for_nodes ||= grouped_objects_for_nodes(
+ nodes,
+ Issue.all.includes(:author, :assignee, :project),
+ self.class.data_attribute
+ )
+ end
+
+ private
+
+ def issue_for_node(issues, node)
+ issues[node.attr(self.class.data_attribute).to_i]
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/label_parser.rb b/lib/banzai/reference_parser/label_parser.rb
new file mode 100644
index 00000000000..e5d1eb11d7f
--- /dev/null
+++ b/lib/banzai/reference_parser/label_parser.rb
@@ -0,0 +1,11 @@
+module Banzai
+ module ReferenceParser
+ class LabelParser < BaseParser
+ self.reference_type = :label
+
+ def references_relation
+ Label
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/merge_request_parser.rb b/lib/banzai/reference_parser/merge_request_parser.rb
new file mode 100644
index 00000000000..c9a9ca79c09
--- /dev/null
+++ b/lib/banzai/reference_parser/merge_request_parser.rb
@@ -0,0 +1,11 @@
+module Banzai
+ module ReferenceParser
+ class MergeRequestParser < BaseParser
+ self.reference_type = :merge_request
+
+ def references_relation
+ MergeRequest.includes(:author, :assignee, :target_project)
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/milestone_parser.rb b/lib/banzai/reference_parser/milestone_parser.rb
new file mode 100644
index 00000000000..a000ac61e5c
--- /dev/null
+++ b/lib/banzai/reference_parser/milestone_parser.rb
@@ -0,0 +1,11 @@
+module Banzai
+ module ReferenceParser
+ class MilestoneParser < BaseParser
+ self.reference_type = :milestone
+
+ def references_relation
+ Milestone
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/snippet_parser.rb b/lib/banzai/reference_parser/snippet_parser.rb
new file mode 100644
index 00000000000..fa71b3c952a
--- /dev/null
+++ b/lib/banzai/reference_parser/snippet_parser.rb
@@ -0,0 +1,11 @@
+module Banzai
+ module ReferenceParser
+ class SnippetParser < BaseParser
+ self.reference_type = :snippet
+
+ def references_relation
+ Snippet
+ end
+ end
+ end
+end
diff --git a/lib/banzai/reference_parser/user_parser.rb b/lib/banzai/reference_parser/user_parser.rb
new file mode 100644
index 00000000000..a12b0d19560
--- /dev/null
+++ b/lib/banzai/reference_parser/user_parser.rb
@@ -0,0 +1,92 @@
+module Banzai
+ module ReferenceParser
+ class UserParser < BaseParser
+ self.reference_type = :user
+
+ def referenced_by(nodes)
+ group_ids = []
+ user_ids = []
+ project_ids = []
+
+ nodes.each do |node|
+ if node.has_attribute?('data-group')
+ group_ids << node.attr('data-group').to_i
+ elsif node.has_attribute?(self.class.data_attribute)
+ user_ids << node.attr(self.class.data_attribute).to_i
+ elsif node.has_attribute?('data-project')
+ project_ids << node.attr('data-project').to_i
+ end
+ end
+
+ find_users_for_groups(group_ids) | find_users(user_ids) |
+ find_users_for_projects(project_ids)
+ end
+
+ def nodes_visible_to_user(user, nodes)
+ group_attr = 'data-group'
+ groups = lazy { grouped_objects_for_nodes(nodes, Group, group_attr) }
+ visible = []
+ remaining = []
+
+ nodes.each do |node|
+ if node.has_attribute?(group_attr)
+ node_group = groups[node.attr(group_attr).to_i]
+
+ if node_group &&
+ can?(user, :read_group, node_group)
+ visible << node
+ end
+ # Remaining nodes will be processed by the parent class'
+ # implementation of this method.
+ else
+ remaining << node
+ end
+ end
+
+ visible + super(current_user, remaining)
+ end
+
+ def nodes_user_can_reference(current_user, nodes)
+ project_attr = 'data-project'
+ author_attr = 'data-author'
+
+ projects = lazy { projects_for_nodes(nodes) }
+ users = lazy { grouped_objects_for_nodes(nodes, User, author_attr) }
+
+ nodes.select do |node|
+ project_id = node.attr(project_attr)
+ user_id = node.attr(author_attr)
+
+ if project && project_id && project.id == project_id.to_i
+ true
+ elsif project_id && user_id
+ project = projects[project_id.to_i]
+ user = users[user_id.to_i]
+
+ project && user ? project.team.member?(user) : false
+ else
+ true
+ end
+ end
+ end
+
+ def find_users(ids)
+ return [] if ids.empty?
+
+ User.where(id: ids).to_a
+ end
+
+ def find_users_for_groups(ids)
+ return [] if ids.empty?
+
+ User.joins(:group_members).where(members: { source_id: ids }).to_a
+ end
+
+ def find_users_for_projects(ids)
+ return [] if ids.empty?
+
+ Project.where(id: ids).flat_map { |p| p.team.members.to_a }
+ end
+ end
+ end
+end