ddev-spidergram

• What is ddev-spidergram?
• Installation
• Basic usage
• How to access ArangoDB
• Backup and restore ArangoDB
• Behind the scenes
• TODO
• Contributing

What is ddev-spidergram?

This repository provides an addon to use @Spidergram within DDEV.

Spidergram is a customizable toolkit for crawling and analyzing complicated web properties. While it can be used to crawl any website, we (the folks at Autogram) designed it specifically for “ten websites in a trench coat” scenarios where a web property encompasses multiple CMSs, multiple domains, and multiple design systems, maintained by multiple teams.

</blockquote>

Installation

1. Make sure Docker and DDEV are installed.

2. Create a new directory and move into it. For simplicity reasons I am using the name spidercrawl across this readme. You are able to choose any other name here instead.

mkdir spidercrawl && cd spidercrawl

1. Initialize your DDEV project. By using the defaults the project name will be equal to the directory name.

ddev config --auto

1. Download the ddev-spidergram-addon.

ddev add-on get rpkoller/ddev-spidergram

1. Start DDEV and wait a few minutes until the DDEV and ArangoDB images, Spidergram, as well as Playwright are downloaded and installed.

ddev start

Basic usage

1. Run an initial status check that everything is set up correctly.

ddev spidergram status

The resulting output should look like that:

$> ddev spidergram status

SPIDERGRAM CONFIG
Config file: /var/www/html/spidergram.config.yaml

ARANGODB
Status:   online
URL:      https://spidercrawl.ddev.site:8529
Database: db

1. Crawl and analyze your first site.

ddev spidergram go https://ddev.com

1. For more details see the Spidergram documentation. All configuration changes are applied to the spidergram.config.yaml file.

How to access ArangoDB

1. The ArangoDB web interface could be reached in the web browser via the the URL shown for ddev spidergram status. The port :8529 is appended to the project’s URL (https://spidercrawl.ddev.site:8529).

1. You have to click _SYSTEM in the upper right corner of the screen. In the select form on the next screen you have to click _system again and then choose the option db and confirm.

Backup and restore ArangoDB

1. To backup your ArangoDB database and delete your project:

ddev arangodump
ddev delete spidergram --omit-snapshot

The database is saved in .ddev/arangodb-backup. After the successful dump ddev delete spidergram --omit-snapshot deletes the project’s containers, images and volumes. The project files as well as the DDEV config files in .ddev, including the ArangoDB database dump, remain untouched. That saves disk space and enables you to re-add the project at a later point as described in the second step.

1. To restore your project:

ddev config
ddev start
ddev arangorestore

That way you re-register the existing project in DDEV, start it up and restore the database you have previously used in ArangeoDB.

1. In case you want to use arangodump not to have a final backup before you delete your project but save one or more backups in your daily usage it has to be noted that with the current implementation it is not possible. By running arangodump the previous dump gets overwritten! Providing a more flexible and convinient solution is planned for the future.

Behind the scenes

1. Adds a docker-compose file (docker-compose.arangodb.yaml) for ArangoDB. The Spidegram database and password are set to db to be in line with DDEV’s standards. The only difference is that the default username was left at root since it wasn’t changeable in ArangoDB. The ArangoDB container was set to not require any authentication, which is in line with the Spidergram docker-compose file.
2. Adds a dockerfile (Dockerfile.spidergram) to the web-build folder. It runs a npm install --global spidergram, npx playwright install, and a npx playwright install-deps when the addon is installed.
3. Adds a spidergram web command. For example you only have to type ddev spidergram status instead of ddev exec spidergram status.
4. Adds a spidergram.config.yaml to the project root. The Yaml file with that exact file name is mandatory for Spidergram to run.
5. The config.ddev-spidergram.yaml file ensures that Node.js is set to version 18. In a post-start-hook it is also taken care that the URL set in spidergram.config.yaml is in line with the overall project settings. The project name, based on $DDEV_PROJECT, and the TLD, based on $DDEV_TLD, is getting replaced by a regex statement on every start. That way, if the project name or the TLD changes at a later point, Spidergram still just keeps working.
6. Adds a arangodump web command. The database dump is written to a fixed destination .ddev/arangodb-backup/. Currently arangodumpis intended to be used to backup the database before a project is getting removed from DDEV.
7. Adds a arangorestore web command. Make sure that your folder with the database backup is available at .ddev/arangodb-backup/ within your project folder before you run ddev config && ddev start.
8. The .ddev/arangodb-backup/ directory is created with the -p option in a post_install_action and a .gitignore file is being added to the directory excluding everything within.

TODO

• Figure out the best approach how to upgrade Spidergram and it’s dependencies for an already existing Spidergram DDEV instance and update the README accordingly (I have to wait for the next Spidergram release being able to test that).
• Expand the number of available settings in the spidergram.config.yaml. At the moment I am only using the default values from an old template found at https://github.com/autogram-is/create-spidergram/tree/main/templates.
• Further expand the interaction capabilities ddev-spidergram provides for ArangoDB.

Contributing

Any feedback in regard to bugs and potential improvements is welcome.

Contributed and maintained by @rpkoller based on the original ddev-addon-template