sync-to-kv

Push the contents of a folder to Cloudflare KV.

If you have a nice little static website, and you want to host the files somewhere, you can shove them in CFKV to be accessed by a small fast Worker.

Install

The usual way (global for CLI access):

npm install -g sync-to-kv

You will need the KV "Namespace ID", and some environment variables:

CF_ACCOUNT_ID - Your Cloudflare account id.
CF_NAMESPACE_ID - The "Namespace ID" of that KV store.
Either this one (recommended):
- CF_API_TOKEN - A generated API token with Workers/KV storage permissions.
Or this one:
- CF_AUTH_EMAIL - Your Cloudflare account email address.
- CF_AUTH_KEY - Your global API key.

(Note: the CLOUDFLARE_ prefix is also supported.)

Then you simply run the command like this:

sync-to-kv ./path/to/folder

You can run --help for more details, but the flags you can pass in are:

-i, --ignore : Ignore one or more files, using glob syntax. Example -i='**/*.test.js'.
-p, --prefix : This is a prefix used for partitioning groups of KV entries within the same store. E.g. site A may use -p='a:' while site B may use -p='b:'.
-f, --file (default file:) : Prefix used for all file entries, e.g. a file index.html with the default prefix (file:) would be stored with a key file:index.html.
-h, --hash (default hashes) : Name used to store the dictionary hash map of key names to hashes, which is used to improve upload time.
-d, --dryrun : Just print out what would have been updated or removed, but do not actually update KV.

If you have a folder like site and you want to exclude all *.test.js files, you would write something like:

sync-to-kv ./site -i '**/*.test.js'

If you have multiple sites in one KV store, it might be good to store them by domain name, so your --prefix would be the domain:

sync-to-kv ./site -p site.com

Here's a longer example with many flags:

sync-to-kv ./site \
  -i '**/*.test.js' \
  -i '**/*.md' \
  -i '.DS_Store' \
  -p site.com \
  -f 'bin:' \
  -h 'hashmap'

Published and released under the Very Open License.