1---
2stage: none
3group: unassigned
4info: To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#assignments
5---
6
7# Backend directory structure
8
9## Use namespaces to define bounded contexts
10
11A healthy application is divided into macro and sub components that represent the contexts at play,
12whether they are related to business domain or infrastructure code.
13
14As GitLab code has so many features and components it's hard to see what contexts are involved.
15We should expect any class to be defined inside a module/namespace that represents the contexts where it operates.
16
17When we namespace classes inside their domain:
18
19- Similar terminology becomes unambiguous as the domain clarifies the meaning:
20  For example, `MergeRequests::Diff` and `Notes::Diff`.
21- Top-level namespaces could be associated to one or more groups identified as domain experts.
22- We can better identify the interactions and coupling between components.
23  For example, several classes inside `MergeRequests::` domain interact more with `Ci::`
24  domain and less with `ImportExport::`.
25
26```ruby
27# bad
28class MyClass
29end
30
31# good
32module MyDomain
33  class MyClass
34  end
35end
36```
37
38### About namespace naming
39
40A good guideline for naming a top-level namespace (bounded context) is to use the related
41feature category. For example, `Continuous Integration` feature category maps to `Ci::` namespace.
42
43Alternatively a new class could be added to `Projects::` or `Groups::` if it's either:
44
45- Strictly related to one of these domains. For example `Projects::Alias`.
46- A new component that does not have yet a more specific domain. In this case, when
47  a more explicit domain does emerge we would need to move the class to a more specific
48  namespace.
49
50Do not use the [stage or group name](https://about.gitlab.com/handbook/product/categories/#devops-stages)
51since a feature category could be reassigned to a different group in the future.
52
53```ruby
54# bad
55module Create
56  class Commit
57  end
58end
59
60# good
61module Repositories
62  class Commit
63  end
64end
65```
66
67On the other hand, a feature category may sometimes be too granular. Features tend to be
68treated differently according to Product and Marketing, while they may share a lot of
69domain models and behavior under the hood. In this case, having too many bounded contexts
70could make them shallow and more coupled with other contexts.
71
72Bounded contexts (or top-level namespaces) can be seen as macro-components in the overall app.
73Good bounded contexts should be [deep](https://medium.com/@nakabonne/depth-of-module-f62dac3c2fdb)
74so consider having nested namespaces to further break down complex parts of the domain.
75For example, `Ci::Config::`.
76
77For example, instead of having separate and granular bounded contexts like: `ContainerScanning::`,
78`ContainerHostSecurity::`, `ContainerNetworkSecurity::`, we could have:
79
80```ruby
81module ContainerSecurity
82  module HostSecurity
83  end
84
85  module NetworkSecurity
86  end
87
88  module Scanning
89  end
90end
91```
92
93If classes that are defined into a namespace have a lot in common with classes in other namespaces,
94chances are that these two namespaces are part of the same bounded context.
95