LostTime

Orienteering utilities on the web. Visit http://LostTimeOrienteering.com to simplify your workflow and recover some lost time.

The current goal of LostTime is not to replace event management software, but to make tasks before and after an event easier. Customization on a per-club basis is expected.

Usage

Entries

Input: any number of .csv files representing entries
Output: single .csv file ready for import into SportSoftware, or a PDF check-in list ready to print for the registration table. For the input .csv files, if the header line contains any of the following strings, its data will be included in the output: bib, first, last, club or school, class, sex or gen, punch or card, rent, and nc. Checking is loose: a column titled First Name will match first. (Special case, the word emergency does not match gen or nc.)

Event Results

Input: single IOF v3 .xml <ResultList> file
Output: an html page, results by class, sorted, and scored. Can also include a second html page with team results Scoring methods include: Time (no score), Ratio to winner (1000pts), World Cup (100, 95, 92, 90, ...), or Alphabetical sorting (no placement). Team scoring methods are added as requested. Currently there is one: WIOL (Washington Interscholastic Orienteering League) teams are the top three finishers from a school in a given class, except middle school where the girls and boys races are combined in a single Middle School Team class.

Series Results

Input: existing event results uploaded to Lost Time Orienteering Output: html page, series results by class, calculated and sorted. Series can be configured to count the best N scores out of M races. Individuals must run in the same class, and are matched on name+club.

Backlog

Entries

make the default output an IOF xml <EntryList> file, for greater inter-operability

Series

ability to add blank events to fill out remainder of a series
better detect input file types (for example, Sport Software ScoreO .csv)

Technical Notes

Architecture

This is a Flask application with a SQLite backend. There is currently no front-end framework; data is passed to the Jinja2 template engine that comes bundled with Flask.

Development and Testing

You'll need python 2.x and virtualenvwrapper installed on a linux machine (or windows subsystem for linux):

Fork the project, then clone to a folder on your machine

$ git clone https://github.com/<username>/LostTime.git <Path/To/Project/Folder>

Set up a python virtual environment with all dependencies. (The -a flag adds a virtual environment to an existing project directory; that's what we want to do.)
```
$ mkvirtualenv losttime -a <ProjectFolder> -r requirements.txt
```
It's likely that the lxml package will fail to install as it has some lower level C dependencies. Check out the lxml [http://lxml.de/installation.html](installation instructions) for help. I needed the following on my Ubuntu 14.04 in Windows Subsystem for Linux:
```
(losttime)$ sudo apt-get install libxml2-dev libxslt-dev python-dev zlib1g-dev
(losttime)$ pip install lxml #success!
```

add an instance folder and instanceconfig.py Mine looks like this:

# instance/instanceconfig.py
 from os.path import abspath, dirname, join
 _cwd = dirname(abspath(__file__))

 DEBUG = True
 SQLALCHEMY_DATABASE_URI = 'sqlite:///' + join(_cwd, 'dev.db') # puts the db in the instance folder
 SQLALCHEMY_ECHO = False
 SECRET_KEY = 'generate-a-secret-key-with-os.urandom(24)-and-paste-here'

 SEND_EMAILS = False # app will call unix sendmail program.
 FLASH_EMAILS = True # app will display emails in the UI as a flashed message.

run the development server
```
python rundevserver.py
```
OR
```
export FLASK_APP=losttime
export FLASK_DEBUG=1
python -m flask run
```
There are some bugs with how the flask command works in virtual environments, but running it from python -m works.

Database Management

Using the Flask-Migrations package which leverages the Alembic package for managing database schema changes. Changes in models.py are detected and automatically create a migration script. After making changes, run the following:

python -m flask db migrate -m 'message'
# review the new migration script in migrations/versions, then
python -m flask db upgrade

Deployment

It's easy! Open an ssh client and git pull and then restart the daemon via TERM in the NFS GUI. If there are new libraries, make sure requirements.txt is updated and pip install -r requirements.txt before restarting. You can upgrade to a new db schema with python -m flask db upgrade, though that line is also in the startup script so it's not necessary.

Initial

First Deployment on a fresh nearlyfreespeech.net server. There was a lot of debugging. Here's what worked in the end.

clone the repository to the protected folder:

$ cd /home/protected
$ git clone https://github.com/ercjns/LostTime.git .

create the virtual environment, install dependencies.

$ virtualenv -p /usr/local/bin/python2 venv //prevent python3 from installing
$ source venv/bin/activate
$ pip install -r requirements.txt //removed wsgiref from the list, it was causing issues

create instance file

$ mkdir instance
$ touch instanceconfig.py
$ emacs instanceconfig.py
$ // type in instance config (see above)
$ // save with C-x C-s, exit with C-x C-c

initialize the databse

$ export FLASK_APP=losttime
$ python -m flask db update

test the startup script

$ source venv/bin/activate
$ gunicorn -p app.pid losttime:app

Well, actually, the gunicorn command was still reading from the global gunicorn, which caused it to use python 3 and miss my packages. So the second line in the startup script is usr/local/bin/gunicorn ...

configure the daemon with the startup script in the NFS UI.

DanGrabski/LostTime

LostTime

Usage

Entries

Event Results

Series Results

Backlog

Entries

Series

Technical Notes

Architecture

Development and Testing

Database Management

Deployment

Initial