{
"info": {
"author": "Gabriel Gazola Milan",
"author_email": "gabriel.gazola@poli.ufrj.br",
"bugtrack_url": null,
"classifiers": [
"Development Status :: 4 - Beta",
"Intended Audience :: Developers",
"License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
"Programming Language :: Python :: 3",
"Programming Language :: Python :: 3.4",
"Programming Language :: Python :: 3.5",
"Programming Language :: Python :: 3.6",
"Programming Language :: Python :: 3.7",
"Topic :: Software Development :: Libraries :: Python Modules"
],
"description": "# Akuanduba\n\n
\n \n
\n\nImage source: [link](https://my-bestiario.fandom.com/pt-br/wiki/Akuanduba)\n\nAkuanduba is a Python framework that eases manipulation of multiple running threads and shared resources. Its name was inspired by a god of the Brazilian mythology: Akuanduba, the god of order.\n\nThis framework's been based on the [Gaugi](https://gitlab.cern.ch/jodafons/gaugi) project, a framework that Jo\u00e3o, the author of Gaugi, presented to me.\n\n# Top 5 reasons for Akuanduba\n\nThis is really a TL;DR section:\n\n* Flexible;\n* Easy developing and debugging;\n* Robust structure;\n* Effortless threading and shared resources manipulation;\n* Smooth attaching and removing features from your script.\n\n# First steps\n\n* Clone this repository:\n\n```\ngit clone https://github.com/gabriel-milan/Akuanduba\n```\n\n* Install this library:\n\n```\npip3 install Akuanduba\n```\n\n* Go to the example script folder:\n\n```\ncd Akuanduba/examples/TimedCounterFileSaving\n```\n\n* Read the script, in order to understand what's supposed to happen when you run it:\n\n```\nvim SampleScript.py\n```\n\n* Run the example script:\n\n```\npython3 SampleScript.py\n```\n\n# The paradigm\n\n## Base classes\n\nUsing Akuanduba, you'll have three base classes you can use to attend your needs, as they relate to one another through a common class called *Context* (this will be explained really soon). The three classes are:\n\n* *AkuandubaTool*;\n* *AkuandubaService*;\n* *AkuandubaDataframe*.\n\n**AkuandubaDataframe** is a base class for data models that will be attached to the *Context*. In other words, these *data frames* can store anything you desire and be accessible from any *tool* or *service* running on your main script. **Example:** Use data frames for storing data acquired with a service.\n\n**AkuandubaTool** is a base class for methods that will run once on every call, processing data from any *data frame* attached to the *Context* and appending the results to another *data frame* (or even the same). *Tools* have one main method: the \"execute\", where all the calculations will occur every time they're called. **Example:** Use tools to process data acquired by a service and store it on a data frame.\n\n**AkuandubaService** is a base class for parallel threads. These *services* have two main methods: \"run\", which is a loop that will be running parallel to the whole framework and \"execute\", that's executed once in every call. **Example**: Use services to acquire data.\n\n## Other main concepts\n\nBesides these base classes, Akuanduba relies on few other main concepts:\n\n* *Context*;\n* *DataframeManager*;\n* *ToolManager*;\n* *ServiceManager*;\n* *AkuandubaTrigger*.\n\nThe **Context** is an abstraction that holds every single thing attached to the framework: *tools*, *services* and *data frames*. This way, everything is accessible from any other component attached to Akuanduba. Every *dataframe* you get from the context is locked to your *execute* method and released only after its execution.\n\nThe **Managers** (*DataframeManager*, *ToolManager* and *ServiceManager*) are all based on the *Manager* class, the only difference among them is that they manage different things. You can manage *data frames*, *tools* and *services*, respectively. In other words, you can attach them with the *\\_\\_add\\_\\_* operation and retrieve them with the *retrieve* method. The things you attach to these managers will run in the order you do it (execution will be better discussed later).\n\nThe **AkuandubaTrigger** is a class that inherits from *AkuandubaTool*. Its purpose is to build an object you can attach multiple hypothesis tests (that must inherit from *TriggerCondition*) and *tools* in order to trigger the *execute* method of all these *tools*. You can use three types of triggers: 'and', 'or' and 'xor', all of them self-explanatory.\n\n## Execution\n\nAkuanduba itself has three main methods: *initialize*, *execute* and *finalize*.\n\nThe *initialize* method follows these steps:\n\n* Insert the framework status into the *Context*;\n* Insert all the *data frames* on the *DataframeManager* into the *Context*;\n* Insert all the *services* on the *ServiceManager* into the *Context*, running them;\n* Insert all the *tools* on the *ToolManager* into the *Context*, running them;\n* Checks for error in the *Context* initialization.\n\nThe *execute* method is the main execution loop. Its tasks are:\n\n* Loop through the *services*, calling their *execute* methods;\n* Loop through the *tools*, calling their *execute* methods;\n\nAnd finally, the *finalize* method just tries to kill every thread created by the Akuanduba framework.\n\n### Example\n\nLet's say you have two *services* that acquire data from different sources and one single *tool* that process the whole data. You could build a script that attaches both *services* to the *ServiceManager* and the *tool* to the *ToolManager*. Let's say that, for this purpose, you've created two *data frames*, one for storing the raw data and one for the processed data. A diagram for your script's execution would be the following:\n\n
![](\"https://raw.githubusercontent.com/gabriel-milan/Akuanduba/master/img/Akuanduba.jpg\")
\n![](\"https://raw.githubusercontent.com/gabriel-milan/Akuanduba/master/img/execution.png\")
\n\n