The Tasking Manager is composed of two parts:
- Frontend: A user interface built using React.
- Backend: A database and API built using Python.
The two parts can be developed independently of each other.
The Tasking Manager uses OAuth2 with OSM to authenticate users.
In order to use the frontend, you may need to create keys for OSM:
-
Login to OSM (If you do not have an account yet, click the signup button at the top navigation bar to create one).
Click the drop down arrow on the top right of the navigation bar and select My Settings.
-
Register your Tasking Manager instance to OAuth 2 applications.
Put your login redirect url as
http://127.0.0.1:3000/authorized/
Note:
127.0.0.1
is required for debugging instead oflocalhost
due to OSM restrictions. -
Permissions required:
- Read user preferences (read_prefs).
- Modify the map (write_api).
-
Now save your Client ID and Client Secret for the next step.
-
Copy the
example.env
totasking-manager.env
.cp example.env tasking-manager.env
-
Update the following variables
TM_CLIENT_ID=from-previous-step TM_CLIENT_SECRET=from-previous-step
If you are a frontend developer and do not wish to configure the backend, you can use our staging server API.
Update the variable:
TM_APP_API_URL='https://tasking-manager-staging-api.hotosm.org'
before running the
yarn start
command.Be aware that the staging API can be offline while we are deploying newer versions to the staging server and that you'll not have access to some management views due to permissions. Check the configuration section to learn more about how to configure Tasking Manager.
For more details see the configuration section.
The easiest option to get started with all components may be using Docker.
Docker Engine must be available locally.
Once you have the docke engine running, Quickly generate an environment file from an existing example.env
.
cp example.env tasking-manager.env
Now you can proceed with starting the services.
docker compose pull
docker compose build
docker compose up --detach
Tasking Manager should be available from: http://127.0.0.1:3000
You change the default port from 3000
to any other port.
However, you must change your OAuth redirect URL to reflect this, in addition to any variables including a port, e.g. TM_APP_BASE_URL.
The default dotenv file can also be changed.
TM_DEV_PORT=9000 ENV_FILE=.env docker compose up --detach
docker compose build
docker compose up --detach
If you want to add custom configuration for the docker services. You can make a copy of docker-compose.override.sample.yml
which you can edit as per your need.
Create an override file from sample.
cp docker-compose.override.sample.yml docker-compose.override.yml
If you want to use your local postgresql server or some other exter database service.
Find these sets of environment variables in tasking-manager.env
POSTGRES_DB=tasking-manager
POSTGRES_USER=tm
POSTGRES_PASSWORD=tm
POSTGRES_ENDPOINT=<replace-with-your-database-endpoint>
POSTGRES_PORT=5432
NOTE: If database server is self managed on your local machine, Use your machine's ip address. Also make sure it can be reachable from
tm-backend
container.
Once Updated, recreate containers with
docker compose up -d
If you are looking to deploy only Frontend service with docker, You will need to make sure the following env vars are corrent in tasking-manager.env
TM_APP_API_URL=http://127.0.0.1:8000
This refers to the backend service that you are going to consume, If you don't have a Tasking Manager backend instance you can use the staging server hosted by hotosm.
TM_APP_API_URL=https://tasking-manager-staging-api.hotosm.org
Then proceed with starting only frontend service with docker.
docker compose up -d tm-frontend
Check server logs with
docker logs tasking-manager-main-tm-frontend-1 -f
> [email protected] patch-rapid
> bash -c "cp patch/rapid-imagery.min.json public/static/rapid/data/imagery.min.json"
ℹ 「wds」: Project is running at http://172.22.0.2/
ℹ 「wds」: webpack output is served from
ℹ 「wds」: Content not from webpack is served from /usr/src/app/public
ℹ 「wds」: 404s will fallback to /
Starting the development server...
Compiled successfully!
You can now view TaskingManager-frontend in the browser.
Local: http://localhost:3000
On Your Network: http://172.22.0.2:3000
Note that the development build is not optimized.
To create a production build, use yarn build.
For OSM related CLIENT_ID
and SECRETS
check OSM AUTH section.
The client is the front-end user interface of the Tasking Manager. It is based on the React framework and you can find all files in the frontend
directory.
The following dependencies must be available globally on your system:
- Download and install NodeJS LTS v12+ and yarn
- Go into the
frontend
directory and executeyarn
.
In the project directory, you can run:
Runs the app in the development mode.
Open http://localhost:3000 to view it in the browser.
The page will reload if you make edits.
You will also see any lint errors in the console.
Launches the test runner in the interactive watch mode.
See the section about running tests for more information.
Builds the app for production to the build
folder.
It correctly bundles React in production mode and optimizes the build for the best performance.
The build is minified and the filenames include the hashes.
Your app is ready to be deployed!
See the section about deployment for more information.
You can learn more in the Create React App documentation.
To learn React, check out the React documentation.
The backend is made up of a postgres database and an associated API that calls various end points to create tasks, manage task state, and produce analytics.
- Python 3.7+
- Python 3.7 is what HOT uses in production. You can use Python 3.8 too.
- PostgreSQL with PostGIS
- pip
- libgeos-dev
You can check the Dockerfile to have a reference of how to install it in a Debian/Ubuntu system.
There are two ways to configure Tasking Manager. You can set some
environment variables on your shell or you can define the
configuration in the tasking-manager.env
file on the repository root
directory. To use that last option, follow the below instructions:
- Copy the example configuration file to start your own configuration:
cp example.env tasking-manager.env
. - Adjust the
tasking-manager.env
configuration file to fit your configuration. - Make sure that the following variables are set correctly in the
tasking-manager.env
configuration file:TM_APP_BASE_URL
=web-server-endpointPOSTGRES_DB
=tasking-manager-database-namePOSTGRES_USER
=database-user-namePOSTGRES_PASSWORD
=database-user-passwordPOSTGRES_ENDPOINT
=database-endpoint-can-be-localhostPOSTGRES_PORT
=database-portTM_SECRET
=define-freely-any-number-and-letter-combinationTM_CLIENT_ID
=oauth-client-id-from-openstreetmapTM_CLIENT_SECRET
=oauth-client-secret-key-from-openstreetmapTM_REDIRECT_URI
=oauth-client-redirect_uriTM_SCOPE
=oauth-client-scopesTM_LOG_DIR=logs
In order to send email correctly, set these variables as well:
TM_SMTP_HOST
TM_SMTP_PORT
TM_SMTP_USER
TM_SMTP_PASSWORD
TM_SMTP_USE_TLS=0
TM_SMTP_USE_SSL=1
(Either TLS or SSL can be set to 1 but not both)
- Install project dependencies:
- First ensure the Python version in
pyproject.toml:requires-python
is installed on your system. pip install --upgrade pdm
pdm install
- First ensure the Python version in
The project includes a suite of Unit and Integration tests that you should run after any changes.
python3 -m unittest discover tests/backend
or
pdm run test
cd frontend && yarn build-locales
You need to delete all the versions in ./migrations/version. Then, import the new model into the file ./backend/init.py Finally, enter inside the migration container and run:
python manage.py db migrate
and
python manage.py db upgrade
We use Flask-Migrate to create the database from the migrations directory. Check the instructions on how to setup a PostGIS database with docker or on your local system. Then you can execute the following command to apply the migrations:
flask db upgrade
or
pdm run upgrade
To be able to create projects and have full permissions as an admin user inside TM, login to the TM with your OSM account to populate your user information in the database, then execute the following command on your terminal (with the OS user that is the owner of the database):
psql -d <your_database> -c "UPDATE users set role = 1 where username = '<your_osm_username>'"
If you plan to only work on the API you only have to build the backend architecture. Install the backend dependencies, and run the server:
# Install dependencies
pdm install
# Run (Option 1)
pdm run start
# Run (Option 2)
pdm run flask run --debug --reload
You can access the API documentation on http://localhost:8000/api-docs, it also allows you to execute requests on your local TM instance. The API docs is also available on our production and staging instances.
In order to authenticate on the API, you need to have an Authorization Token.
- Run the command line
manage.py
viaflask
with thegen_token
option and-u <OSM_User_ID_number>
. The command line can be run in any shell session as long as you are in the tasking-manager directory.
flask gen_token -u 99999999
This will generate a line that looks like this:
Your base64 encoded session token: b'SWpFaS5EaEoxRlEubHRVC1DSTVJZ2hfalMc0xlalu3KRk5BUGk0'
- In the Swagger UI, where it says
Token sessionTokenHere==
replace sessionTokenHere==
with the string of characters between the
apostrophes (' ') above so you end up with something that looks like
this in that field:
Token SWpFaS5EaEoxRlEubHRVC1DSTVJZ2hfalMc0xlalu3KRk5BUGk0
Your user must have logged in to the local testing instance once of course and have the needed permissions for the API call.
You can get your OSM user id number either by finding it in your local
testing/dev database select * from users
or from OSM by viewing the
edit history of your user, selecting a changeset from the list, and
then at the bottom link Changeset XML
and it will be in the uid
field of the XML returned.
To get your token on the production or staging Tasking Manager instances, sign in in the browser and then either:
- go to the user profile page, enable Expert mode in the settings, and copy the token from the API Key section.
- inspect a network request and search for the
Authorization
field in the request headers section.
If you're not able to connect to an existing tasking-manager DB, we have a Dockerfile that will allow you to run PostGIS locally as follows.
- From the root of the project:
docker build -t tasking-manager-db ./scripts/docker/postgis
- The image should be downloaded and build locally. Once complete you should see it listed, with
docker images
- You can now run the image (this will run PostGIS in a docker container, with port 5432 mapped to localhost):
docker run -d -p 5432:5432 tasking-manager-db
- Confirm the image is running successfully:
docker ps
- Finally you can set your env variable to point at your containerised DB:
export TM_DB=postgresql://hottm:hottm@localhost/tasking-manager
- Refer to the rest of the instructions in the README to setup the DB and run the app.
It may be the case you would like to set up the database without using Docker for one reason or another. This provides you with a set of commands to create the database and export the database address to allow you to dive into backend development.
First, ensure that Postgresql and PostGIS are installed and running on your computer.
Assuming you have sudo access and the unix Postgresql owner is postgres
:
$ sudo -u postgres psql
$ CREATE USER "hottm" PASSWORD 'hottm';
$ CREATE DATABASE "tasking-manager" OWNER "hottm";
$ \c "tasking-manager";
$ CREATE EXTENSION postgis;
Finally, add the environmental variable to access the database:
export TM_DB=postgresql://hottm:hottm@localhost/tasking-manager
It is possible to install and run the Tasking Manager using Docker and Docker Compose.
Clone the Tasking Manager repository and use docker-compose up
to
get a working version of the API running.