.. _GeoArmadillo-Pipeline:

GeoArmadillo Pipelines
######################

The GeoArmadillo *Pipeline* combines multiple *Pipeline Fragments* to perform tasks of any complexity.

All *Blocks* that are needed to build a *Pipeline* can be found in the *Pipeline* category of the toolbox:

.. image:: images/pipeline_category.png
	:alt: The green *Pipeline* category


Pipeline Fragments
==================

The *Pipeline Fragments* control what actually happens in the *Pipeline*. Each *Pipeline Fragment* has a number, the ID number of the respective *Pipeline Fragment*. In the example below it is number 2. Additionally, a description can be added to each *Pipeline Fragment*. A single or multiple stacked *Blocks* are inserted into the free space of the *Pipeline Fragment*. Those *Blocks* define the computations performed in the respective *Pipeline Fragment*.

.. image:: images/pipeline_fragment.png
	:alt: A an image of the *Pipeline Fragment* block.



The Pipeline
============

The actual *Pipeline* combines all *Pipeline Fragments* in any order and thus generates the main method in the output Python script. There must be exactly one *Pipeline* *Block* (this is currently not enforced or checked, so take care of it yourself please!). The order of the *Pipeline Fragments* defines the order in which the computations are executed.

.. image:: images/pipeline.png
	:alt: The *Pipeline* block that gets compiles into the main method

.. image:: images/pipeline_shadow.png
	:alt: The *Pipeline Fragment* shadow used to sort the execution order in the main method.

.. tip::
    Running first *Pipeline* Fragement 2 and in continuation *Pipeline Fragment* 1 looks like this:
    
    .. image:: images/pipeline_example.png

    Variables defined in previously executed *Pipeline Fragments* can be used in subsequent *Pipeline Fragments*. However, there is no future access. If *Pipeline Fragment* 1 computes a variable required in *Pipeline Fragment* 2, *Pipeline Fragment* 1 has to be listed above *Pipeline Fragment* 2 in the *Pipeline*.