Verification phase
=====================

We check that the titles are all different.


During execution
-------------------

This phase checks the coherence of the arguments between them. It concerns all occurrences of the keyword factor ACTION. There are 2 groups of checks.

The **first** group is common to the CHAM_GD and RESULTAT keyword arguments (which are mutually exclusive) and is reduced to the following checks:


* acceptability of components:

it is verified that the components requested are indeed present in the catalog describing the quantity to be treated,

The **second** group is specific to the RESULTAT keyword, and is reduced to checks:


* verification of the acceptability of the symbolic field:

we make sure that the symbolic field argument of the NOM_CHAM keyword exists for the type of concept result argument of the keyword RESULTAT,


* existence of at least one data structure for the symbolic field to be processed.

At the end of the verification phase, in the execution phase, the following alternative is presented:


* all occurrences of ACTION are correct and the operations are started,


* at least one occurrence of ACTION is incorrect, so a fatal error message is produced with the command stopped. Information provides information about the nature of the errors and how to correct them.


During treatment
------------------

A new check filter is presented but never causes the command to stop. This filter is limited to the following checks:


* existence of the data structure designated by CHAM_GD or RESULTAT and one of the keywords allowing access in a RESULTAT data structure.

In case of failure a message is sent and the next occurrence is processed,


* in the case of a calculation of tensor invariants, we check that the object of the treatment does indeed designate a tensor of order 2,


* in the case of an average calculation or a request to extract values, it is verified that the components to be treated have been calculated at the post-treatment site. If this check fails, a message details the components that are not available based on the meshes or nodes at the post-processing location.

If this last filter does not detect any impossibility of calculation, post-processing is started and the results are inserted into the table.