Metadata-Version: 2.1
Name: gamayun-utils
Version: 0.2.1
Summary: Python utils for writing gamayun job scripts
Home-page: https://github.com/ivan-brko/GamayunPyUtils
Author: Ivan Brko
Author-email: ivan.brko@outlook.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: grpcio (==1.30.0)
Requires-Dist: protobuf (==3.13.0)

# Python utilities for Gamayun

This is a package containing python utilities for writing Gamayun jobs. 
Main Gamayun repository can be found [here](https://github.com/ivan-brko/Gamayun).

## Contents
* [Getting the package](#getting-package)
* [Usage](#usage)
  * [Reporting results and errors](#reporting-results)
  * [Executing the script logic](#executing-script-logic)

<a name="getting-package"></a>
## Getting the package

Install the package with the following command:
```bash
pip3 install gamayun-utils
```

This package depends on gprcio, which needs C/C++ compilers installed on the system.
Needed dependencies on Alpine are:
```sh
apk add --no-cache python3 \
&& apk add --no-cache gcc \
&& apk add --no-cache g++ \
&& apk add --no-cache libc-dev \
&& apk add --no-cache linux-headers \
&& apk add --no-cache python3-dev
```
Note that suggested way of using Gamayun is with Docker, so you don't have to install these dependencies locally. If you are using Docker, there is an image containing Gamayun and Gamayun Python utilities [here](https://hub.docker.com/repository/docker/ibrko/gamayun_py_utils). 

See [Gamayun main repo](https://github.com/ivan-brko/Gamayun) for more information and see example usage [here](https://github.com/ivan-brko/GamayunConfigurationSample).

<a name="usage"></a>
## Usage
This part documents the package API

<a name="reporting-results"></a>
### Reporting results and errors

Functions for reporting results or errors from the script are following:

```python
def report_result(results)
```
```python
def report_error(error)
```

```report_result``` receives a list of strings which represent results and ```report_error``` receives a string which represents the error.

Note that once result or error is reported for the script/gamayun-job, Gamayun will stop listening for results/errors for that job, so in entire job there should be only one call to ```report_result``` or ```report_error```.

<a name="executing-script-logic"></a>
### Executing the script logic

There is a function with the following signature in this package:
```python 
def run_gamayun_script_logic(callback)
```

All the script logic should be placed in some function and that function should be given to ```run_gamayun_script_logic``` as argument as this will ensure that all uncought exceptions are caught and reported as errors to Gamayun. 

So, a job script file should look something like this:
```python
from gamayun_utils import report_result
from gamayun_utils import report_error
from gamayun_utils import run_gamayun_script_logic

def script_logic():
    # place the script logic here
    if OK:
        report_result(results)
    else:
        report_error(error)

run_gamayun_script_logic(script_logic)
```

