Skip to main content

Upload Presets

Overview​

Upload presets enable us to centrally define a set of asset upload options instead of specifying them in each upload call. We can define multiple upload presets and apply different presets in different upload scenarios.

Use Cases​

Some use-cases for using upload presets include:

  • Define how the assets will be stored and accessed, such as the way the public ID value will be determined (use_filename), which folders they'll be stored in (folder), and whether existing assets can be replaced by those uploaded with the preset (overwrite).
  • Generate eager transformations for uploaded files (eager). This is especially valuable for video transformations that may take longer to generate and helps to ensure your uploaded files will be ready for delivery.
  • Apply incoming transformations while uploading a file (transformation), for example to limit format types or file size before storing uploaded files in your Cloudinary account.
  • Apply add-on capabilities, such as auto-tagging (categorization) or automatic moderation (moderation), to your uploads.
  • Update the behavior of user-generated content uploads without modifying your website or mobile app code.
  • Set up default upload behavior for uploads via the Media Library or via upload method commands.
  • Use as upload preferences for uploads via the Upload widget.

Creating and managing upload presets​

You can create, modify, or delete upload presets using the Upload Preset UI in the Management Console.

Notes​

  • When defining an upload preset that may be used for both image and video uploads, you can include both image-specific and video-specific upload options (e.g., for moderation or tagging add-ons) and only the relevant options will be used when a file is uploaded with that preset.

Our Use Cases​

Event​

This will be used by the Digital Brand team for assets created during Rapha events.

  • Use Filename

    • Specific naming convention of <SEASON-OR-YEAR>_event_<EVENT-NAME>_<COUNTRY>_<FILE-NUMBER>.filetype

    OR

    • Specific naming convention of <SEASON-OR-YEAR>_<EVENT-NAME>_<COUNTRY>_<ANY-TEXT>.filetype

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
      • Year taken from filename
      • Event name taken from filename
      • Country taken from filename
    • Color analysis
    • Quality analysis
    • Accessibility analysis
    • Image metadata
    • Face analysis

Location​

This will be used by the Digital Brand team for assets created on location.

  • Use Filename

    • Specific naming convention of <SEASON-OR-YEAR>_location_<COLLECTION>_<LOCATION>_<FILE-NUMBER>.filetypeOR
    • Specific naming convention of <SEASON-OR-YEAR>_<COLLECTION>_<LOCATION>_<ANY-TEXT>.filetype

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
      • SKU taken from filename
    • Color analysis
    • Quality analysis
    • Accessibility analysis
    • Image metadata
    • Face analysis

Triptych video (2:3)​

This will be used by the Digital Merchandising team for any assets that will be used in the triptych component. This could include on-model videos, mannequin videos and macro videos.

If you update any of the values below, please cross reference with the Cloudinary Restore Asset lambda which uses these values and would also need updating if anything changes: https://gitlab.com/rapharacing/rapha-middleware/cloudinary/restore/-/tree/master/services/restore-asset

  • Use Filename

    • Specific naming convention of <BASE-PRODUCT_SKU>_<PRODUCT-NAME>_<SEASON>_<VIDEO-TYPE>_<ORDER>.filetype
    • VIDEO-TYPE is one of MANNEQUIN, MACRO or MODEL
    • <ORDER> is optional

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
      • SKU taken from filename
      • Season taken from filename
      • Collection name taken from filename

  • Upload Manipulations:

    • Incoming transformations:
      • Audio codec set to opus
      • Audio frequency set to 48000 (max for opus)
      • Aspect ratio set to 2:3
      • Crop to fill
    • Eager transformations:
      • c_fill,g_center,vc_vp9,w_675 (webM, video codec vp9, 675 width)
      • c_fill,g_center,q_auto,vc_h265,w_675/mp4 (mp4, video codec h.265, 675 width)
      • c_fill,g_center,q_auto,vc_h264,w_675/mp4 (mp4, video codec h.264, 675 width)

Triptych image​

This will be used by the Digital Merchandising team for any assets that will be used in the triptych component. This could include on-model videos, mannequin videos and macro images.

If you update any of the values below, please cross reference with the Cloudinary Restore Asset lambda which uses these values and would also need updating if anything changes: https://gitlab.com/rapharacing/rapha-middleware/cloudinary/restore/-/tree/master/services/restore-asset

  • Use Filename

    • Specific naming convention of <BASE-PRODUCT-SKU>_<PRODUCT-NAME>_<SEASON>_<IMAGE-TYPE>_<ORDER>.filetype
    • IMAGE-TYPE is one of MANNEQUIN, MACRO or MODEL
    • ORDER is optional

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
      • SKU taken from filename
      • Season taken from filename
      • Collection name taken from filename
    • Color analysis
    • Quality analysis
    • Accessibility analysis
    • Image metadata
    • Face analysis

Mannequin​

This will be used by both Digital Merchandising and Digital Brand teams for any assets that were produced by our studio to be included as Mannequin assets in the PDP. This would include on-model images, mannequin images and macro images.

If you update any of the values below, please cross reference with the Cloudinary Restore Asset lambda which uses these values and would also need updating if anything changes: https://gitlab.com/rapharacing/rapha-middleware/cloudinary/restore/-/tree/master/services/restore-asset

  • Use Filename

    • Specific naming convention of <BASE-PRODUCT-SKU>_<VARIANT-COLOR>_<PRODUCT-NAME>_<SEASON>_<IMAGE-TYPE>_<ORDER>.filetype
    • IMAGE-TYPE is one of MANNEQUIN, MACRO or MODEL
    • Filename must finish with a number for ordering.
    • Underscore character must only be used as a separator between the different parts of the file name, never within one of the parts.

Upload manipulations:

  • Eager transformations:

    • Aspect ratio 1:2
    • Crop to fill
    • DPR 2
    • Width 992px
    • Width 1920px

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
      • SKU taken from filename
      • Variant taken from filename
      • Season taken from filename
      • Collection name taken from filename
    • Color analysis
    • Quality analysis
    • Accessibility analysis
    • Image metadata
    • Face analysis

Standard video​

This will be used by the Brand team for assets that do not require a specific naming convention or need to be synced with Hybris.

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags

  • Upload Manipulations:

    • Incoming transformations:
      • Audio codec set to opus
      • Audio frequency set to 48000 (max for opus)
      • Aspect ratio set to 2:3
      • Crop to fill
    • Eager transformations:
      • g_center,vc_vp9 (webM, video codec vp9)
      • g_center,q_auto,vc_h265/mp4 (mp4, video codec h.265)
      • g_center,q_auto,vc_h264/mp4 (mp4, video codec h.264)

Bundles​

This will be used by both Digital Merchandising team for assets related to bundles as they are shown in the PLP.

  • Use Filename

    • Specific naming convention of BUN-<YEAR-SEASON-NUMBER-COMBINATION>.filetype
    • BUN is the letters BUN
    • SEASON-NUMBER-COMBINATION is a 6 character number and letter combination that details the season and number for the bundle e.g. H12204
    • SEASON-NUMBER-COMBINATION must be uppercase

Upload manipulations:

  • Eager transformations:

    • Aspect ratio 500:150
    • Crop to fill
    • DPR 2
    • Width 992px
    • Width 1920px

  • Analysis and AI

    • Categorization
      • Amazon Rekognition Tags
    • Color analysis
    • Quality analysis
    • Accessibility analysis
    • Image metadata

Resources​