view README.md @ 378:88aa790d1437

Do not assume existance of a specific database in setup In case the invoking database user has no database of the same name, we can just use the database we want to create anyhow.
author Tom Gottfried <tom@intevation.de>
date Fri, 10 Aug 2018 13:42:51 +0200
parents fd04bccae6ca
children 0f78b13f4a9a
line wrap: on
line source

# Quick Start

## Build

- To build all components of gemma, simply type `make` on the top
  level directory of the repro.

- To only (re)build the back end you can use `make gemma`.

- To only build the SPA-Client you can use `make client`.

For further details see [docs/DEVELOPMENT](docs/DEVELOPMENT.md),


## Setup Database

- You will need a PostgreSQL cluster with PostGIS.

- To install the **gemma** schema and roles use the script
  `./schema/install-db.sh`.

- `./schema/install-db.sh --help` shows you available options.
  Per default the script will create a database named "gemma" and all
  necessary roles in the postgres default cluster (listening on port
  5432).

- The script must be run as a user with PostgreSQL super user rights.
  By convention this is the "postgres" on most systems.

- Standard case: as user "postgres", in the top level directory of the
  repo run:

## Running gemma

- Best is to create a configuration file.  Copy the example from
  `cmd/gemma/gemma.toml.example` to get started:
  ```
  cp cmd/gemma/gemma.toml.example gemma.toml
  ```

- Edit `gemma.toml`, some parameters you propably want to change:

  * `host` and `port` to make gemma listen on a public interface
  * `service-password` to match the password for "gemma_service" user
    of your database.

- `./cmd/gemma/gemma -h` gives you an overview of more available
  options.

- Then start gemma:
  ```

  ```