annotate bgc_calib.xml @ 4:ae8078d85fc3 draft

planemo upload for repository https://github.com/galaxyecology/tools-ecology/tree/master/tools/ocean_data_view_manager commit bf6cd39b867fec1a8251ff9a4890f276799cdc0c

<tool id="tool_biogeochemical_calibration" name="BioGeoChemical calibration" version="@VERSION@+galaxy@VERSION_SUFFIX@" profile="22.05" license="MIT">
    <description>for sensors according to Argo recommendations</description>
    <macros>
        <token name="@VERSION@">1.0</token>
        <token name="@VERSION_SUFFIX@">2</token>
    </macros>
    <requirements>
        <container type="docker">easyqcvbgc/easy-qcv_calibration-methods_tool:@VERSION@</container>
    </requirements>
    <command detect_errors="exit_code"><![CDATA[
        export HOME=\$PWD &&
        mkdir -p /runtime/config &&
        mkdir -p '/runtime/data-in-raw/' &&
        mkdir -p '/runtime/data-in-ref/' &&
        mkdir -p '/runtime/data-out/' &&
        mkdir -p ./outputs &&

	#for $i, $infile in enumerate($input_raw):
            #set real_name = $infile.element_identifier
            #set extension = $real_name.split('.')[-1]
            #if  $extension == 'nc':
                cp '${infile.file_name}' "/runtime/data-in-raw/${infile.element_identifier}" &&
            #else
                cp '${infile.file_name}' "/runtime/data-in-raw/${infile.element_identifier}.nc" &&
            #end if
	#end for	
        #if $input_ref:
            #for $i, $infile in enumerate($input_ref):            
            	cp '$infile' '/runtime/data-in-ref/${infile.element_identifier}' &&
            #end for
        #end if
        
        #if $config.type == 'yes':
            ## Write YAML content
            cp '${config_file}' '/runtime/config/config_user_odv_tool.yml' &&
        #end if 

        #set $mode="R_APP_CN"
        /app/launchers/start-app.sh $mode &&

        cp -r '/runtime/data-out/' './outputs/'
    ]]></command>
    <configfiles>
        <configfile name="config_file">
        #if $config.type == 'yes':
            #for $i, $file in enumerate($config.section_config):
                default :
                    file_name_list_raw:
                        #for $i, $infile in enumerate($input_raw):
                            #set real_name = $infile.element_identifier
                            #set extension = $real_name.split('.')[-1]
                            #if  $extension == 'nc':
                                #set name_raw='${infile.element_identifier}'
                            #else:
                                #set name_raw='${infile.element_identifier}.nc'
                            #end if 
                            - ${name_raw}  
                        #end for
                    param : nitrate
                    param_qc :
                        bad : '${param_qc_bad}'
                        to_be_replaced : '${param_qc_toreplaced}'
                        replaced_by : '${param_qc_replacedby}'
                    calib_configs :
                        - number : '${i}'
                            method_raw : 
                                name : method1
                                options : 
                                    bypass_P : '${file.bypass.p_value}'
                                    P : '${file.bypass.p_value}'
                                    multiple_linear_regression : '${file.regression.reg_value}'
                                    single_favored : '${file.favored}'
                            method_ref : 
                                name : '${file.method_ref.method_ref_name}'
                                reference : 
                                    name : '${file.method_ref.reference}'
                                    #if '${file.method_ref.reference}' == 'woa':
                                        resolution : '${file.method_ref.resolution}'
                                    #end if
            #end for
        #end if
        </configfile>
    </configfiles>
    <inputs>
        <param name="input_raw" type="data" format="netcdf" multiple="true" label="Input harmonized netcdf data" help="This netcdf should come from the ODV history manager or the QCV harmonizer."/>
        <param name="input_ref" type="data" format="netcdf" optional="true" multiple="true" label="Input reference data" help="Comes from QCV harmonizer"/>
        <param name="param_qc_bad" type="text" optional="true" value="4" label="Bad data QC value"/>
        <param name="param_qc_toreplaced" type="text" optional="true" value="3" label="Nitrate QC value to be replaced"/>
        <param name="param_qc_replacedby" type="text" optional="true" value="1" label="QC value of replacement"/>
        <!--<param name="mode" type="select" label="Select in which mode you want to run the tool">
            <option value="R_APP_CN">
                QV that creates a odv collection with raw and ref files to qualify
            </option>
            <option value="DEMO">
                Valide with ODV demo that performes the demo
            </option>
        </param>-->
        <conditional name="config">
            <param name="type" type="select" label="Select if you want to write your own configuration file or not.">   
                <option value="no">No, I don't want to, I'll use the default one</option>
                <option value="yes">Yes, I to write my own configuration file</option>
            </param>
            <when value="yes">
                <repeat name="section_config" title="Configurations" min="0" max="10">
                    <!--<param name="method_name" type="select" label="Enter methode name">
                        <option value="method1">method1<option>
                    </param>-->
                    <conditional name="bypass">
                        <param name="bypass_value" type="select" label="Bypass">
                            <option value="0">The tool defines the P of reference for adjustment</option>
                            <option value="1">The user defines P and chooses P below</option>
                        </param>
                        <when value="0">
                            <param name="p_value" type="select" label="Fixed P_value">
                                <option value="NULL">In this case the P value is NULL (no option)</option>
                            </param>
                        </when>
                        <when value="1">
                            <param name="p_value" type="integer" value="0" min="0" max="20000" label="Enter P_value"/>
                        </when>
                    </conditional>
                    <conditional name="regression">
                        <param name="ml_regression" type="select" label="Enter multiple linear regression">
                            <option value="0">Automatic choice of N</option>
                            <option value="1">Single linear regression</option>
                            <option value="N>1">N linear regressions</option>
                        </param>
                        <when value="0">
                            <param name="reg_value" type="select" label="Regression choice">
                                <option value="0">Automatic choice of N</option>
                            </param>
                        </when>    
                        <when value="1">
                            <param name="reg_value" type="select" label="Regression choice">
                                <option value="1">Single linear regression</option>
                            </param>
                        </when>
                        <when value="N>1">
                            <param name="reg_value" type="integer" min="2" value="2" label="Regression choice, enter the N linear regressions"/>
                        </when> 
                    </conditional>                  
                    <param name="favored" type="boolean" label="Enter FALSE or TRUE to favored the single linear regression"/>
                    <conditional name="method_ref">
                        <param name="method_ref_name" type="select" label="Reference method">
                            <option value="model">Model</option>
                            <option value="clim">Climatology</option>
                        </param>
                        <when value="clim">
                            <param name="reference" type="select" label="Reference">
                                <option value="woa">WOA</option>
                            </param>
                            <param name="resolution" type="select" label="Resolution">
                                <option value="annual">Annual</option>
                                <option value="monthly">Monthly</option>
                                <option value="seasonal">Seasonal</option>
                            </param>
                        </when>
                        <when value="model">
                            <param name="reference" type="select" label="Reference">
                                <option value="canyon_b">CanyonB</option>
                                <option value="canyon_med">Canyon Med</option>
                            </param>
                        </when>
                    </conditional>
                </repeat>
            </when>
            <when value="no">
            </when>
        </conditional>
    </inputs>
    <outputs>
        <collection type="list" name="nc_files" label="BGC calibration netcdf collection">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)\.nc" directory="./outputs" recurse="true" format="netcdf"/>
        </collection>
        <collection type="list" name="png_files" label="BGC calibration figure collection">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)\.png" directory="./outputs" recurse="true" format="png"/>
        </collection>
    </outputs>
    <tests>
        <test expect_num_outputs="2">
            <param name="input_raw" location="https://gitlab.com/pokapok-projects/easy-qcv-bgc/qcv-bgc-tools/-/raw/3ae832154f55c76a7c95a5dc1e6bb19946917233/galaxy_tool/test-data/QCV_harmonizer_NetCDF_data_0_qced.nc"/>
            <conditional name="config">
                <param name="type" value="no"/>
            </conditional>
            <output_collection name="nc_files" type="list" count="1"/>
            <output_collection name="png_files" type="list" count="4"/>

        </test>
        <test expect_num_outputs="2">
            <param name="input_raw" location="https://gitlab.com/pokapok-projects/easy-qcv-bgc/qcv-bgc-tools/-/raw/3ae832154f55c76a7c95a5dc1e6bb19946917233/galaxy_tool/test-data/QCV_harmonizer_NetCDF_data_0_qced.nc"/>
            <conditional name="config">
                <param name="type" value="yes"/>
		<repeat name="section_config">
                    <conditional name="bypass">
                        <param name="bypass_value" value="0"/>
                    </conditional>
                    <conditional name="regression">
                        <param name="ml_regression" value="0"/>
                        <param name="reg_value" value="0"/>
                    </conditional>
                    <param name="favored" value="FALSE"/>
                    <conditional name="method_ref">
                        <param name="method_ref_name" value="model"/>
                        <param name="reference" value="canyon_b"/>
                    </conditional>  
		</repeat>
            </conditional>
            <output_collection name="nc_files" type="list" count="1"/>
            <output_collection name="png_files" type="list" count="4"/>
        </test>
    </tests>
    <help><![CDATA[
==========================
BioGeoChemical calibration
==========================

**What it does**

This tool calibrates BioGeoChemical (BGC) data measured by ARGO or GLIDER network. The BGC variables managed by the tool for the moment are :

- NITRATE using methods published in `archimer <https://archimer.ifremer.fr/doc/00350/46121/>`


**Input description**

- file_name_raw : nc file to be qualified
- param_qc$bad : qc values to be removed from param before calibration
- param_qc$to_be_replaced : raw qc values to be replaced for building the adjusted qc values
- param_qc$replaced_by : good qc value replacing raw qc values for building adjusted qc values
- method_raw$name which method/equation will be used for estimating the adjusted variable
- method_raw$options :
	- bypass_P : 0 (tool defines the P of reference for adjustment ) OR 1 (user defines P and chooses P below)
        - P : NULL (if bypass_P = 0) OR numeric_values (if bypass_P = 1)
        - multiple_linear_regression : 0 (automatic choice of N) or 1 (single linear regression) or N>1 (N linear regressions)
        - single_favored : FALSE or TRUE to favored the single linear regression
- method_ref$name : which kind of reference method (for example model or climatology) will be used by the method_raw for estimating the adjusted variable.
- reference : which reference method like canyonb or woa will be used by the method_raw for estimating the adjusted variable.

If the list of raw file name is empty, the tool stops If the user makes an error in the file extension, tool changes it by itself. If the user misses the yaml configuration, tool uses its default config and upload available files in data-in-raw

compatibility matrix

- variable / application mode / [method_raw.name, method_ref.name, reference]
- nitrate / R_APP_CN / [method1, model, canyon_b] & [method1, model, canyon_med] & [method1, clim, woa]
	    
**Output**

- an extended nc file including the _adjusted and _adjusted_qc
- figures helping to understand the adjustment

    ]]></help>
    <citations>
        <citation type="bibtex">
            @Manual{,
            title = {ODV tool},
            author = {Pokapok},
            year = {2024},
            note = {https://gitlab.com/pokapok-projects/easy-qcv-bgc/qcv-odv-tool}
        </citation>    
    </citations>
</tool>