Command line tools for running and using the Local CMS Server
yarn add @hubspot/local-cms-server-cli yarn hs-cms-server init # accept the defaults for all prompts yarn hs-cms-server start
NOTE: if this is your first starting the server, it may take a minute to pull down the docker image. Once you see the line "Local HubL Server started, running at http://0.0.0.0:8080/" printed to the console, navigate to that URL in your browser
yarn add @hubspot/local-cms-server-cli yarn hs-cms-server init --config --context # respond to the CLI prompts (hapikey, FTP creds, portalId) yarn hs-cms-server run-tasks download-all yarn hs-cms-server start
Your project should have two directories:
designs (These names can be configured to whatever you'd like, more on that later)
context- This directory contains HubSpot entities, as well as some default stylesheets to be rendered with your templates. e.g., your
contextdirectory could have a
hubdb/<portalId>directory. In your templates, you will be able to use HubL tags and functions that reference these HubDB tables.
designs- This directory will contain all your HubL templates, stylesheets, js, custom modules in whatever organization you'd like. It is meant to mimic the structure of your project within the design manager UI.
Pulls down the latest docker image of the local cms server (unless you already have it locally), checks the current directory for
server-config.yaml and then starts the server on `http://0.0.0.0:8080.
This will bring you to an index page with three sections:
context/content/<portalId>/) with their specified template. If your template references content variables, those will be rendered
context/blogs/<portalId>/) with their specified template.
--configflag initializes a
--contextflag initializes a `context directory. This includes default hubspot stylesheets, some generic objects to be rendered with your templates (e.g. site-settings.json), as well as some example objects that will be rendered with portal 123
--designsflag initializes a
designsdirectory with some example templates, that reference some of portal 123's data in the
--allwill initialize all of the above. You can also omit a flag all together for the same effect
This command is used to run gulp tasks for downloading data from your portal to your project. It uses the configs specified in
--tasks flag lists the valid tasks you can run, which are:
download-all- Runs all the
download-blogs- Downloads blog listing page objects, which allows you to render your templates with blog context
download-content- Downloads page content, which allows you to render your templates with page context
download-default-modules- Downloads the default modules for your portal, in a
.moduledirectory structure. e.g. Rich Text module
download-designs- Uses FTP to download your designs directory
download-hubdb- Downloads your hubdb tables/rows
download-layouts- Downloads drag and drop layout objects, to locally render pages built in the DnD editor
download-menus- Downloads navigation menus
download-resource-mappings- Downloads a mapping file of resource ID -> file location in the design manager. This is needed for HubL functions that reference resources by their ID.
This command uploads the list of files/directories specified under the
You shouldn't need to worry about the other properties in that file
This config is used for the
run-tasks command described above. Here you can specify your
hapikey (for connecting to HubSpot APIs) and your
contextDirPath for where you want these local entities to be stored. This should match the
server-config.yaml. Your hapikey can be found here: https://app.hubspot.com/l/integrations-settings/api-key
download-designsasks for your FTP username/password, portalId, and the local output dir.
limitis a property for most tasks. If the limit is set to null for a task, then all the entities in that task will be downloaded via multiple calls to the API. If you don't need all entities, consider setting this limit property.