Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/oliveroneill/hanserver

An image collection server for displaying recent and local images to users
https://github.com/oliveroneill/hanserver

geolocation golang instagram mongodb

Last synced: 1 day ago
JSON representation

An image collection server for displaying recent and local images to users

Awesome Lists containing this project

README 

        
# han server

[![Build Status](https://travis-ci.org/oliveroneill/hanserver.svg?branch=master)](https://travis-ci.org/oliveroneill/hanserver)



A server that stores images from arbitrary sources and can return them based

on location and recency. Han is short for 'here and now'. This server is used

with the Han app, which can be installed [here](https://itunes.apple.com/au/app/han-here-and-now/id1205247234?mt=8&ign-mpt=uo%3D4)



## Status

Unfortunately Instagram is [deprecating the API](https://www.instagram.com/developer/changelog/)

that was used to collect a majority of the content populating Han. For Han

to continue to be a useful project, I'll need to work out another source

of images that has enough open data. Until then, this project won't

continue development.



## Dependencies

* [Docker](https://docs.docker.com/engine/installation/)

* [docker-compose](https://docs.docker.com/compose/install/)



## Usage

There are two separate components that are part of han:

* Image population (hancollector) - this will use a variety of collector

implementations to retrieve images from different sources and store them

in a unified format.

* Web server (hanhttpserver) - this retrieves the images for a client using HTTP.



Both of these are started through `hanhttpserver`, which can be started by

simply calling:

```bash

docker-compose build && docker-compose up

```

from the base directory. Configuration is required for image population to

work, see the [Configuration section](#configuration) below.

Alternatively they can be started individually by calling that same

command from within either `hanhttpserver` or `hancollector`.

Note that `hanhttpserver` automatically starts `hancollector` within the same

process, this is used to keep track of API calls between the server and the

collector. `hanhttpserver` can be started without `hancollector` by using the

`--no-collection` option.



### Slack logging

Errors can be logged through Slack by passing in the `--slacktoken` argument

into `hanhttpserver`. This is logged to the "hanserver" channel but can be

changed in `hanapi/reporting/reporting.go`.



### Configuration

Before calling `docker-compose up` you will need to copy `default_config.json`

and set the required fields to configure the collectors. Copying this json

and calling it `config.json` is recommended since the `gitignore` includes

this file.

`hanhttpserver` will throw an error if no collectors are enabled.

If you don't want to use the implemented collectors, just set `enabled` to

`false`. You must then implement your own collector, see

`hancollector/README.md` for more info.



The `hanapi` directory contains common classes between these two components.



There's an additional README in both `hanhttpserver` and `hancollector` that

discusses their development.



## Testing

All tests can be run using the command `go test ./...`



## TODO

This is a list of features or issues I'd like to work on in the future.

* Deployment - the two Dockerfiles contain the same dependencies and should use

the same base image

* Regioning - to make this project scalable, locations are broken up into

regions, these regions are used to avoid populating the whole world with images.

This should aim to keep the database size down by choosing the most recently

used locations.

* Cleaning up images - images that have been deleted from their original

source need to be taken down, there needs to be a neat way of doing this

without periodically making loads of requests to check the response code