crawlnicle

My personalized news and blog aggregator. Taking back control over the algorithm. Pining for the days of Google Reader. An excuse to write more Rust.

Development Instructions

Prerequisites

Install these requirements to get started developing crawlnicle.

• rust

• postgres

• sqlx-cli

  • Only postgres needed. Install with:

  cargo install sqlx-cli --no-default-features --features native-tls,postgres
  

• just

• (optional) cargo-watch for auto-recompiling the server in development

• (optional) mold for faster builds

First-time setup

1. Create postgres user and database:

   createuser crawlnicle
   createdb crawlnicle
   sudo -u postgres -i psql
   postgres=# ALTER DATABASE crawlnicle OWNER TO crawlnicle;
   \password crawlnicle
   
   # Or, on Windows in PowerShell:
   
   & 'C:\Program Files\PostgreSQL\13\bin\createuser.exe' -U postgres crawlnicle
   & 'C:\Program Files\PostgreSQL\13\bin\createdb.exe' -U postgres crawlnicle
   & 'C:\Program Files\PostgreSQL\13\bin\psql.exe' -U postgres
   postgres=# ALTER DATABASE crawlnicle OWNER TO crawlnicle;
   \password crawlnicle
   

2. Save password somewhere safe and then and add a .env file to the project directory with the contents:

   RUST_LOG=crawlnicle=debug,cli=debug,lib=debug,tower_http=debug,sqlx=debug
   HOST=127.0.0.1
   PORT=3000
   DATABASE_URL=postgresql://crawlnicle:<password>@localhost/crawlnicle
   DATABASE_MAX_CONNECTIONS=5
   TITLE=crawlnicle
   MAX_MEM_LOG_SIZE=1000000
   

3. Run just migrate (or sqlx migrate run) which will run all the database migrations.

Running in Development

Run just watch to build and run the server while watching the source-files for changes and triggering a recompilation when modifications are made.

The server also triggers the browser to reload the page when the server binary is updated and the server is restarted.

It also separately watches the files in frontend/ which will trigger a transpilation with bun and then rebuild the server binary so that it includes the new JS bundle names.

Alternatively, you can just run cargo run.

Building for Production

You can also build the binary in release mode for running in production with the just build command (or cargo build --release).

Using the CLI

This project also comes with a CLI binary which allows you to manipulate the database directly without needing to go through the REST API server. Run cli --help to see all of the available commands.

Running jobs

To periodically fetch new items from all of the feeds execute the cli crawl command in a cronjob.