This approach separates all languages, which can make it easier to find specific files. More configuration entries are required, however, resulting in a longer site navigation.
Hugo calls this Translation by content directory. Your files will be placed inside a top-level content folder for each language (e.g. content/es/posts/my-post.md).
The best practice here is to define a collection per language and folder of related files (e.g. Pages EN, Pages ES, Pages FR, Posts EN, Posts ES, Posts FR).
Here's an example with Pages and Posts:
content/
└─ en/
├─ about.md
└─ blog/
└─ first-post.md
└─ es/
├─ about.md
└─ blog/
└─ first-post.md
└─ fr/
├─ about.md
└─ blog/
└─ first-post.md
collections_config:
pages_en:
name: Pages EN
icon: wysiwyg
path: content/en
parse_branch_index: true
filter: strict
pages_es:
name: Pages ES
icon: wysiwyg
path: content/es
parse_branch_index: true
filter: strict
pages_fr:
name: Pages FR
icon: wysiwyg
path: content/en
parse_branch_index: true
filter: strict
posts_en:
name: Posts EN
icon: event_available
paths: content/en/blog
posts_es:
name: Posts ES
icon: event_available
paths: content/es/blog
posts_fr:
name: Posts FR
icon: event_available
paths: content/fr/blog
The parse_branch_index option is specific to Hugo sites on CloudCannon, and affects what collection _index.* files get assigned to.
By default these files do not get assigned to the closest collection, as they usually represent the landing pages of your site.
As such, they will be assigned to the global pages collection unless otherwise configured.
If a collection is tagged as parse_branch_index: true, any _index.* files in the collection path (and subdirectories) will be assigned
to it.
In this case, we want the content/*/blog/_index.md files to be in the pages_* collection for their language,
rather than in the global pages collection, or the inner posts_* collection.
When you browse a collection in CloudCannon, all files within that folder will be shown by default.
Applying the strict filter to a collection instead limits the UI to show only the files that exactly match that collection.
In this case, it prevents files from the posts_* collections in the content/*/blog/ subdirectories
from appearing when browsing the pages collection for that language.
{
"collections_config": {
"pages_en": {
"name": "Pages EN",
"icon": "wysiwyg",
"path": "content/en",
"parse_branch_index": true,
"filter": "strict"
},
"pages_es": {
"name": "Pages ES",
"icon": "wysiwyg",
"path": "content/es",
"parse_branch_index": true,
"filter": "strict"
},
"pages_fr": {
"name": "Pages FR",
"icon": "wysiwyg",
"path": "content/en",
"parse_branch_index": true,
"filter": "strict"
},
"posts_en": {
"name": "Posts EN",
"icon": "event_available",
"paths": "content/en/blog"
},
"posts_es": {
"name": "Posts ES",
"icon": "event_available",
"paths": "content/es/blog"
},
"posts_fr": {
"name": "Posts FR",
"icon": "event_available",
"paths": "content/fr/blog"
}
}
}The parse_branch_index option is specific to Hugo sites on CloudCannon, and affects what collection _index.* files get assigned to.
By default these files do not get assigned to the closest collection, as they usually represent the landing pages of your site.
As such, they will be assigned to the global pages collection unless otherwise configured.
If a collection is tagged as parse_branch_index: true, any _index.* files in the collection path (and subdirectories) will be assigned
to it.
In this case, we want the content/*/blog/_index.md files to be in the pages_* collection for their language,
rather than in the global pages collection, or the inner posts_* collection.
When you browse a collection in CloudCannon, all files within that folder will be shown by default.
Applying the strict filter to a collection instead limits the UI to show only the files that exactly match that collection.
In this case, it prevents files from the posts_* collections in the content/*/blog/ subdirectories
from appearing when browsing the pages collection for that language.