https://github.com/gmemstr/daily-servo

https://github.com/gmemstr/daily-servo

Last synced: 6 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/gmemstr/daily-servo
• Owner: gmemstr
• License: agpl-3.0
• Created: 2024-08-01T14:26:42.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-09-04T14:58:21.000Z (over 1 year ago)
• Last Synced: 2025-03-19T08:19:14.590Z (10 months ago)
• Language: JavaScript
• Size: 62.5 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
Daily Servo

===

Despite the name, this doesn't (technically) render Servo images daily (at least, not automatically)

Still very much a work in progress but is generally at a point where you can fork and use it.

About

---

Various tools for rendering and sharing snapshots of websites using Servo, primarily to track [my own site](https://servo.arch.dog)

over time.

`snapshot.sh` is a wrapper around Docker for rendering a screenshot then producing a hash of the image. This can help

indicate when Servo's rendering has changed. If you get inconsistant results with the same version and site, see

[this GitHub issue](https://github.com/servo/servo/issues/32771).

`worker/` is a small Cloudflare Worker for collecting and serving the images. Currently requires manual uploading.

snapshot.sh

---

Generating snapshots is easy but requires a few commands to be run in sequence. `snapshot.sh` wraps:

* Pulling the latest container image

* Running the container image with a volume mount to produce a snapshot

* Hashing the image

* Uploading the image to a Worker endpoint (see below)

Uploading the image is optional, provided `--dry-run` is passed.

Worker

---

The main purpose of the worker is to serve as a repository for snapshots over time. It stores images in Cloudflare R2 and

pointers to the image in KV, with a dedicated `LATEST` key pointing to the latest snapshot. The hash of the image is used as

the filename (it could technically be anything, but it *should* be the hash).

### Setup

You'll need a KV namespace and R2 bucket, which you can swap out in `worker/wrangler.toml`. You'll

also want to set an API token with `npx wrangler secret put API_TOKEN`.

You can then POST to it!

```

 curl -X POST -H "Authorization: API_TOKEN" \

   https://daily-servo.your-name.workers.dev/new \

   -F date=(date -I) -F hash=098f6bcd4621d373cade4e832627b4f6 \

   -F file=@file.png

```

The hash is md5 since that's "good enough" for detecting duplicate files at the moment, but it could easily be swapped out.

### Endpoints

A few endpoints are exposed

| endpoint | about |

|----------|-------|

|`/`|Basic HTML page|

|`/latest`|Will `302` redirect to the latest snapshot's image|

|`/latest.json`|Latest snapshot info with the hash, file URL and date|

|`/list.json`|List of all snapshots (dates and hashes), along with a `latest` pointer|

|`/new`|`POST` endpoint for uploading a new snapshot. Authorized endpoint|

|`/`|Specific snapshot in JSON format|

#### `/latest.json` | `/`

```json

{

  "hash": string,

  "file": string,

  "date": ISO-8601 string

}

```

#### `/list.json`

```json

{

  "latest": string,

  "list": [

    {

      "date": ISO-8601 string,

      "hash": string

    }

  ]

}

```