Class: ConnectorsSdk::GitLab::Extractor

Inherits:
Base::Extractor show all
Defined in:
lib/connectors_sdk/gitlab/extractor.rb

Constant Summary collapse

PAGE_SIZE =

max is 100

100

Constants inherited from Base::Extractor

Base::Extractor::DEFAULT_CURSOR_KEY, Base::Extractor::MAX_CONNECTION_ATTEMPTS, Base::Extractor::TRANSIENT_SERVER_ERROR_CLASSES

Instance Attribute Summary

Attributes inherited from Base::Extractor

#client_proc, #completed, #config, #content_source_id, #features, #monitor, #original_cursors, #service_type

Instance Method Summary collapse

Methods inherited from Base::Extractor

#authorization_data, #authorization_data!, #client, #client!, #convert_transient_server_errors, #cursors_modified_since_start?, #deleted_ids, #document_changes, #download_args_and_proc, #evictable?, #identifying_error_message, #initialize, #permissions, #retrieve_latest_cursors, #transient_error?, #with_auth_tokens_and_retry, #yield_single_document_change

Constructor Details

This class inherits a constructor from ConnectorsSdk::Base::Extractor

Instance Method Details

#yield_deleted_ids(ids) ⇒ Object 



59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
 
# File 'lib/connectors_sdk/gitlab/extractor.rb', line 59

def yield_deleted_ids(ids)
  if ids.present?
    ids.each do |id|
      response = client.get("projects/#{id}")
      if response.status == 404
        # not found - assume deleted
        yield id
      else
        unless response.success?
          raise "Could not get a project by ID: #{id}, response code: #{response.status}, response: #{response.body}"
        end
      end
    end
  end
end

#yield_document_changes(modified_since: nil) ⇒ Object 



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
46
47
48
49
50
51
52
53
54
55
56
57
 
# File 'lib/connectors_sdk/gitlab/extractor.rb', line 20

def yield_document_changes(modified_since: nil)
  query_params = {
    :pagination => :keyset,
    :per_page => PAGE_SIZE,
    :order_by => :id,
    :sort => :desc
  }
  # looks like it's an incremental sync
  if modified_since.present?
    date_since = modified_since.is_a?(Time) ? modified_since : Time.new(modified_since)
    query_params[:last_activity_after] = date_since.iso8601
  end

  next_page_link = nil

  loop do
    if next_page_link.present?
      if (matcher = /(https?:[^>]*)/.match(next_page_link))
        clean_query = URI.parse(matcher.captures[0]).query
        query_params = Rack::Utils.parse_query(clean_query)
      else
        raise "Next page link has unexpected format: #{next_page_link}"
      end
    end
    response = client.get('projects', query_params)

    JSON.parse(response.body).map do |doc|
      doc = doc.with_indifferent_access
      if config.index_permissions
        doc = doc.merge(project_permissions(doc[:id], doc[:visibility]))
      end
      yield :create_or_update, ConnectorsSdk::GitLab::Adapter.to_es_document(:project, doc), nil
    end

    next_page_link = response.headers['Link'] || nil
    break unless next_page_link.present?
  end
end

#yield_permissions(source_user_id) {|result| ... } ⇒ Object

Yields:

  • (result) 


75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
 
# File 'lib/connectors_sdk/gitlab/extractor.rb', line 75

def yield_permissions(source_user_id)
  result = []
  if source_user_id.present?
    result.push("user:#{source_user_id}")

    user_response = client.get("users/#{source_user_id}")
    if user_response.success?
      username = JSON.parse(user_response.body).with_indifferent_access[:username]
      query = { :external => true, :username => username }
      external_response = client.get('users', query)
      if external_response.success?
        external_users = Hashie::Array.new(JSON.parse(external_response.body))
        if external_users.empty?
          # the user is not external
          result.push('type:internal')
        end
      else
        raise "Could not check external user status by ID: #{source_user_id}"
      end
    else
      raise "User isn't found by ID: #{source_user_id}"
    end
  end
  yield result
end