The WP2Static plugin has been closed since August 10, 2020 and is not available for download.

We had a PR in the old Workers Docs repo here for additional instructions for Windows users: cloudflare/workers-docs#738

Opening an issue to update.

products/access/src/content/getting-started/access-setup.md says:

"Choose an authentication domain. The authentication domain will represent you Cloudflare for Teams and the URL will serve as the launch page for end users."

Should that be "represent you to Cloudflare for Teams"? Or perhaps "represent you on Cloudflare for Teams"?

As stated in the README, we deploy projects to both a testing site (cloudflare-docs.workers.dev) and a production site (workers.cloudflare.com):

We should instruct search engines to ignore the test sites via a robots.txt with:

User-agent: * Disallow: /

So the below handler allows us to do things with element. I am wondering how can we append a nested/child element to element?

class ElementHandler {
  element(element) {
    // An incoming element, such as `div`
    console.log(`Incoming element: ${element.tagName}`)
  }
}

"The QR code example located here: https://developers.cloudflare.com/workers/tutorials/build-a-serverless-function seems to be missing the correct Content-Type header.

Here's what the example code looks like:

const generate = async request => {
// ...
return new Response(qr_png, { headers })
}
The paragraph right above it mentions the header that should be there, but the code doesn't match.

By default, the QR code is generated as a PNG. Construct a new instance of Response, passing in the PNG data as the body, and a Content-Type header of image/png: this will allow browsers to properly parse the data coming back from your serverless function, as an image:"

cloudflare/workers-docs#965

Hello,

I prepared a simple web app template by using rust on Cloudflare workers. Do you think it would make sense to add this as documentation?

cf-workers-web-app

Sorry if this is the wrong place to post this, but I think this example should be updated to respect file paths:

https://developers.cloudflare.com/workers/examples/respond-with-another-site

So if you go to /404, it relays that and displays the proper page.

Need to migrate over more content about namespaces, usecases, etc. in a new guide.
I would prioritize this.

https://developers.cloudflare.com/stream/video-playback/player-api/

Linked to from here and in google search results.

It has been reported that this fixes the issue:

{
  "private": true,
  "dependencies": {
    "cloudflare-docs-engine": "git+https://github.com/cloudflare/cloudflare-docs-engine.git"
  },
  "scripts": {
    "bootstrap": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh bootstrap",
    "build": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh build",
    "develop": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh develop",
    "ghactionsbootstrap": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh ghactionsbootstrap",
    "savechanges": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh savechanges",
    "serve": "sh ./node_modules/cloudflare-docs-engine/bin/commands.sh serve"
  }
}

• Step 5 of the getting started guide currently recommends wrangler preview but we should be pushing people towards wrangler dev

• Step 6 should recommend using wrangler login for expediency :)

The documentation for the argo tunnel command line arguments is very outdated. Please update documentation.

Here's the relevant output from cloudflared.exe tunnel --help:

OPTIONS:
   --config value                                      Specifies a config file in YAML format.
   --origincert value                                  Path to the certificate generated for your origin when you run cloudflared login. (default: "C:\\Users\\vakadm.STOLTUAT\\.cloudflared\\cert.pem") [%TUNNEL_ORIGIN_CERT%]
   --autoupdate-freq value                             Autoupdate frequency. Default is 24h0m0s. (default: 24h0m0s)
   --no-autoupdate                                     Disable periodic check for updates, restarting the server with the new version. (default: false) [%NO_AUTOUPDATE%]
   --metrics value                                     Listen address for metrics reporting. (default: "localhost:") [%TUNNEL_METRICS%]
   --pidfile value                                     Write the application's PID to this file after first successful connection. [%TUNNEL_PIDFILE%]
   --url URL                                           Connect to the local webserver at URL. (default: "http://localhost:8080") [%TUNNEL_URL%]
   --hello-world                                       Run Hello World Server (default: false) [%TUNNEL_HELLO_WORLD%]
   --socks5                                            specify if this tunnel is running as a SOCK5 Server (default: false) [%TUNNEL_SOCKS%]
   --proxy-connect-timeout value                       HTTP proxy timeout for establishing a new connection (default: 30s)
   --proxy-tls-timeout value                           HTTP proxy timeout for completing a TLS handshake (default: 10s)
   --proxy-tcp-keepalive value                         HTTP proxy TCP keepalive duration (default: 30s)
   --proxy-no-happy-eyeballs                           HTTP proxy should disable "happy eyeballs" for IPv4/v6 fallback (default: false)
   --proxy-keepalive-connections value                 HTTP proxy maximum keepalive connection pool size (default: 100)
   --proxy-keepalive-timeout value                     HTTP proxy timeout for closing an idle connection (default: 1m30s)
   --proxy-connection-timeout value                    HTTP proxy timeout for closing an idle connection (default: 1m30s)
   --proxy-expect-continue-timeout value               HTTP proxy timeout for closing an idle connection (default: 1m30s)
   --http-host-header value                            Sets the HTTP Host header for the local webserver. [%TUNNEL_HTTP_HOST_HEADER%]
   --origin-server-name value                          Hostname on the origin server certificate. [%TUNNEL_ORIGIN_SERVER_NAME%]
   --unix-socket value                                 Path to unix socket to use instead of --url [%TUNNEL_UNIX_SOCKET%]
   --origin-ca-pool value                              Path to the CA for the certificate of your origin. This option should be used only if your certificate is not signed by Cloudflare. [%TUNNEL_ORIGIN_CA_POOL%]
   --no-tls-verify                                     Disables TLS verification of the certificate presented by your origin. Will allow any certificate from the origin to be accepted. Note: The connection from your machine to Cloudflare's Edge is still encrypted. (default: false) [%NO_TLS_VERIFY%]
   --no-chunked-encoding                               Disables chunked transfer encoding; useful if you are running a WSGI server. (default: false) [%TUNNEL_NO_CHUNKED_ENCODING%]
   --bastion                                           Runs as jump host (default: false) [%TUNNEL_BASTION%]
   --proxy-address value                               Listen address for the proxy. (default: "127.0.0.1") [%TUNNEL_PROXY_ADDRESS%]
   --proxy-port value                                  Listen port for the proxy. (default: 0) [%TUNNEL_PROXY_PORT%]
   --loglevel value                                    Application logging level {fatal, error, info, debug}. At debug level, request URL, method, protocol, content legnth and header will be logged. Response status, content length and header will also be logged in debug level. (default: "info") [%TUNNEL_LOGLEVEL%]
   --transport-loglevel value, --proto-loglevel value  Transport logging level(previously called protocol logging level) {fatal, error, info, debug} (default: "fatal") [%TUNNEL_PROTO_LOGLEVEL%, %TUNNEL_TRANSPORT_LOGLEVEL%]
   --logfile value                                     Save application log to this file for reporting issues. [%TUNNEL_LOGFILE%]
   --log-directory value                               Save application log to this directory for reporting issues. [%TUNNEL_LOGDIRECTORY%]
   --trace-output value                                Name of trace output file, generated when cloudflared stops. [%TUNNEL_TRACE_OUTPUT%]
   --proxy-dns                                         Run a DNS over HTTPS proxy server. (default: false) [%TUNNEL_DNS%]
   --proxy-dns-port value                              Listen on given port for the DNS over HTTPS proxy server. (default: 53) [%TUNNEL_DNS_PORT%]
   --proxy-dns-address value                           Listen address for the DNS over HTTPS proxy server. (default: "localhost") [%TUNNEL_DNS_ADDRESS%]
   --proxy-dns-upstream value                          Upstream endpoint URL, you can specify multiple endpoints for redundancy. (default: "https://1.1.1.1/dns-query", "https://1.0.0.1/dns-query") [%TUNNEL_DNS_UPSTREAM%]
   --proxy-dns-bootstrap value                         bootstrap endpoint URL, you can specify multiple endpoints for redundancy. (default: "https://162.159.36.1/dns-query", "https://162.159.46.1/dns-query", "https://[2606:4700:4700::1111]/dns-query", "https://[2606:4700:4700::1001]/dns-query") [%TUNNEL_DNS_BOOTSTRAP%]
   --hostname value                                    Set a hostname on a Cloudflare zone to route traffic through this tunnel. [%TUNNEL_HOSTNAME%]
   --lb-pool value                                     The name of a (new/existing) load balancing pool to add this origin to. [%TUNNEL_LB_POOL%]
   --metrics-update-freq value                         Frequency to update tunnel metrics (default: 5s) [%TUNNEL_METRICS_UPDATE_FREQ%]
   --tag KEY=VALUE                                     Custom tags used to identify this tunnel, in format KEY=VALUE. Multiple tags may be specified [%TUNNEL_TAG%]
   --retries value                                     Maximum number of retries for connection/protocol errors. (default: 5) [%TUNNEL_RETRIES%]
   --compression-quality value                         (beta) Use cross-stream compression instead HTTP compression. 0-off, 1-low, 2-medium, >=3-high. (default: 0) [%TUNNEL_COMPRESSION_LEVEL%]
   --name value, -n value                              Stable name to identify the tunnel. Using this flag will create, route and run a tunnel. For production usage, execute each command separately [%TUNNEL_NAME%]
   --ui                                                Launch tunnel UI. Tunnel logs are scrollable via 'j', 'k', or arrow keys. (default: false)
   --help, -h                                          show help (default: false)
   --version, -v, -V                                   Print the version (default: false)

Hey,
I notices multiple Prefixes on the BYOIP Docs to be none Cloudflare ones.

Maybe update them to Documentation Prefixes or Cloudflare owned.

For example

• 162.211.156.0/24
• 1.2.3.4/24

wrangler contains a link to instructions to update the CLI, if we notice there's a new version. The link goes to
https://workers.cloudflare.com/docs/quickstart/updating-the-cli/

However, that now 404s. Can we make that redirect to https://developers.cloudflare.com/workers/cli-wrangler/install-update ?

See cloudflare/wrangler-legacy#1552 where this was fixed in wrangler – but users with old versions of wrangler will still be getting shown a broken link when prompted to upgrade.

"Last year, clientTcpRtt and edgeRequestKeepAliveStatus were added to the requests.cf object. We should expose this in developer docs! cc @ispivey"
cloudflare/workers-docs#784

"(Based on experience with other platforms) lot of customers expect downtime or some sort of reboot time or significant cold start when a new Worker is uploaded. We should explain (either in the How it Workers section), or under the wrangler publish command what happens to existing Workers, or inflight requests when you upload a new script, and hence why users shouldn't experience downtime."
cloudflare/workers-docs#654

Redirect all /warpclient/* to /warp-client/*

Unfortunately falls outside the scope of of what lilredirector can do given our siloed product set up, so will likely have to be something custom.

cloudflare/workers-docs#669 (comment)

https://developers.cloudflare.com/workers/templates/pages/graphql_server/

Clean up can be done either in place in the existing Template Gallery, or in a PR in this repo as a new piece of content (can put it anywhere temporarily).

We should use https://ruthless.design/mocks/workers-docs-redesign/reference/html-rewriter/ as the basis for improving what’s currently at https://developers.cloudflare.com/workers/runtime-apis/html-rewriter.

We should be able to show off the polyglot nature of Workers by having code samples in different languages. An MVP might be JavaScript and TypeScript, but we should consider how Rust would work. I wanted to create this issue to get some thoughts on the best way we could do this.

Should there be a language toggle per code sample? for the entire documentation page?

@ispivey please do what you can as discussed in chat, then assign to me for final touchups :)

cloudflare/workers-docs#659

"EG to the security article add how to add 2FA, watch results in various browsers, search results, etc.

Reasoning; comprised accounts can become spam bots:
https://blog.sucuri.net/2020/02/abused-cloudflare-workers-service-used-to-inject-korean-seo-spam.html?utm_source=twitter&utm_medium=blog&utm_campaign=wyatt"

The documentation for wrangler dev implies that the "server on `localhost" is "execut[ing] your Worker": https://developers.cloudflare.com/workers/cli-wrangler/commands#dev

We should update this to be clear that the Worker code is uploaded to Cloudflare's API and the Worker is being invoked at the edge, as described in more detail here: https://blog.cloudflare.com/announcing-wrangler-dev-the-edge-on-localhost/ .

This causes confusion for users who expect that wrangler dev's local server can make requests to locally-running services that are not exposed to the Internet, among other issues.

"I couldn't find this in the docs currently.

We should document the behavior and limitations of making a fetch() call to a URL that matches a route that runs a Worker.

If the fetch URL is on the same zone as the original request, the request will go to the origin, and will not be handled by any other Worker.
If the fetch URL is on a different zone, the request will be handled by the Worker matching the route." cloudflare/workers-docs#858

See cloudflare/cloudflare-docs-engine#181

OLD,NEW
/byoip/api/*,/byoip/dynamic-advertisement/*
/byoip/dynamic-advertisement/configure-prefixes,/byoip/dynamic-advertisement/configure-dynamic-advertisement
/byoip/irr/*,/byoip/irr-records/*
/byoip/about,/byoip/getting started
byoip/delegations,/byoip/prefix-delegations

*Edit I forgot to add the wildcard after routes that have pages underneath them

Add a new page explaining more about using and creating environment variables

The prior stand-alone Workers docs repo used to have incremental builds:

https://github.com/cloudflare/cloudflare-docs-engine/blob/19b0a89bd4b6cb37b8f965ed0213170c97945bad/.github/workflows/build.yml#L10-L21

https://github.com/cloudflare/cloudflare-docs-engine/blob/19b0a89bd4b6cb37b8f965ed0213170c97945bad/.github/workflows/deploy.yml#L15-L24

(The links are @cloudflare/cloudflare-docs-engine but this used to be the @cloudflare/workers-docs-engine repo.)

Should we re-implement that in this new Cloudflare docs repo?

Is it possible that this somehow come from https://github.com/cloudflare/cloudflare-docs-engine? I can’t think of a great way to handle this. Best I can think of is for the engine or its documentation to provide build.yml and deploy.yml templates...

We should document things like:

• CF-Connecting-IP (and the IP you get on Worker subrequests)
• CF-Worker
• CF-EW-Via

I found these two old articles

1. https://blog.cloudflare.com/unit-testing-workers-in-cloudflare-workers/
2. https://findwork.dev/blog/testing-cloudflare-workers/

And this issue cloudflare/wrangler-legacy#1026

Now the wrangler dev is out of beta, it would be nice to have an official page in docs on how to test workers locally and with CI tools.

The tutorial shows examples from and is based on an old version of the cloudflare terraform provider.

For example, the Hello World example uses the token parameter while that's been renamed to api_key.

A much more minor nitpick: there are several typos throughout the guide (eg: "As you configuration" on the Hello World guide).

Thanks for creating the Terraform provider and the cf-terraforming importer, they're very useful! It was just a bit harder to get started than it should have been since the docs were out of date.

Some of the properties in RequestInitCfProperties only work if the account owner has a certain minimum plan level, and some of the properties in IncomingRequestCfProperties are only present if the account owner has a certain minimum plan level.

We previously documented this, but in the content migration to the new docs engine, we dropped it. We should label properties on that page as appropriate.

"The Routes documentation is missing information about how to set routes and manage failure modes in the Workers dashboard.

With the release of free tier limits, rate limited users will be looking for how they manage their failure mode configuration."
https://developers.cloudflare.com/workers/platform/routes

cloudflare/workers-docs#454

On https://developers.cloudflare.com/workers/platform/sites/start-from-worker , the wrangler.toml block is unselectable. I think this is because it is using "sh" formatting instead of "toml" formatting -- "sh" formatting seems to do this for everything to the left of a "$" prompt.

https://wordpress.org/plugins/static-html-output-plugin/

"This plugin has been closed as of August 10, 2020 and is not available for download. This closure is permanent."

The Headers class is pretty straightforward, but its behavior is the source of some frequently asked questions. We should document its behavior in at least these regards:

• The comma-concatenation behavior of Headers.get(). If a Headers object is constructed from a message with multiple instances of the same header, Headers.get() will return a single string (not an array) with each of the multiple header values concatenated together with intervening ", ".
• The comma-concatenation described above is a valid transformation for any well-behaved header, with one infamous exception: Set-Cookie. For this reason, we implement the old, removed Headers.getAll() function, but only for the Set-Cookie header (it throws if you try to access anything else). You can read the saga behind this here: whatwg/fetch#973
• The Cloudflare Workers implementation of Headers.get() returns a USVString (or null, of course), not a ByteString as required by the spec. In English, this means we interpret the header bytes on the wire as UTF-8, decode them, and expose them as Unicode strings in JavaScript. We once tried to "fix" this by switching to ByteStrings, but ByteStrings are difficult to use correctly, and in practice it is common to set a header from the contents of URLSearchParams.get(), which returns a USVString, and would thus produce a corrupt header.
• Related to the above, we warn in the preview if a script does try to add a ByteString to a Headers object. Example: https://cloudflareworkers.com/#2b9b84913d7147de5592b6306404e128:https://tutorial.cloudflareworkers.com

/cc @ispivey

In "2. Install the Workers CLI", due to this issue, provide link to alternative installation via yarn or by building locally.

"The docs are not very clear on how to use npm dependencies.

The answer (I believe?) is to use webpack. There is a page on webpack:

https://developers.cloudflare.com/workers/tooling/wrangler/webpack/

But this page is also not very clear -- it seems to assume the reader is already deeply familiar with webpack, and only provides some advanced details.

We should improve the doc so that people new to webpack don't need to go learn webpack first, and highlight better than this is the way to use npm dependencies."

Reported on Community here, https://community.cloudflare.com/t/incoming-request-cf-properties/199908

I am finding out the correct plan for incoming request cf. I came from this article.

developers.cloudflare.com

Request · Cloudflare Workers docs
Documentation for Cloudflare Workers, a serverless execution environment that allows you to create entirely new applications or augment existing ones without configuring or maintaining infrastructure.

As for the free plan -
I can get

request.cf.asn;
request.cf.colo;

but not city,latitude,longitude,continent,postalCode,metroCode,region,regionCode,timezone,region .
May I know pro plan of cloudflare can give which of these . Thank you.

These fields were typically only available on Business plans and higher

https://web.archive.org/web/20200414104608/https://developers.cloudflare.com/workers/reference/apis/request/

The current documentation does not mention that any more, however the limitation was not changed, rather the documentation needs updating.

So, Business plans and higher.

https://developers.cloudflare.com/workers/cli-wrangler/configuration#environments

route should have it's own explanation

Both MDN and the spec state that the crypto.getRandomValues() function should return the passed ArrayBuffer with random values inside it, but the Cloudflare Workers docs says it returns a Promise<ArrayBuffer> instead.

I do hope this is just a typo.

OLD                          NEW
/images/about                /images
/images/responsive           /images/responsive-images
/images/worker               /images/resizing-with-workers
/images/overlays             /images/drawing-overlays
/images/errors               /images/troubleshooting
/images/protected-origin/    /images/controlling-origin-access

See 1f6f131

cc @dotjs

cloudflare/workers-docs#879

"In particular, the following won't work:

let response = new Response(...)
return response.redirect(...)
(it should just be Response.redirect())

See https://community.cloudflare.com/t/redirect-method-in-response-object-is-not-exists-but-the-documentation-says-it-exists/179953"

The old docs included documentation on Wrangler commands for sites. cloudflare/workers-docs@604bba9#diff-2aea78e7c8d86c0589ec9b3618e95fe0d7ed918a0f9fcd6a28eba9ffc22b7e98 This doesn't exist yet in the new docs.

https://developers.cloudflare.com/workers/sites/ this used to resolve and it doesnt anymore