influx_si v7.0.1_1

[sgsokol]

FOR CONTRIBUTOR:

• - I have read the CONTRIBUTING.md document and this tool is appropriate for the tools-iuc repo.
• - License permits unrestricted use (educational + commercial)
• - This PR adds a new tool or tool collection
• - This PR updates an existing tool or tool collection
• - This PR does something else (explain below)

[sgsokol]

Please review and deploy if OK.

[sgsokol]

FOR CONTRIBUTOR:

* [ ]  - I have read the [CONTRIBUTING.md](https://github.com/galaxyproject/tools-metabolomics/blob/master/CONTRIBUTING.md) document and this tool is appropriate for the tools-iuc repo.

The URL for CONTRIBUTING.md is broken. Probaly it should be https://github.com/workflow4metabolomics/tools-metabolomics/blob/master/CONTRIBUTING.md

[sgsokol]

All issues seem to be resolved to me. Do you see any left out?

[MathsCell]

Please merge if no further action is required on my part.

[lecorguille]

Thank you @bgruening and @bernt-matthias for your reviews! You rocks!

Thank you @sgsokol for your contribution and to have apply the advices and requests.

[bernt-matthias]

Suggested change

	<tool id="influx_si" name="influx_si" version="@TOOL_VERSION@+galaxy1" profile="21.09">
	<tool id="influx_si" name="influx_si" version="@TOOL_VERSION@+galaxy0" profile="21.09">

[bernt-matthias]

Labels should be reverted, like so:

Suggested change

	<param argument="--tblimit" type="integer" min="0" value="0" label="--tblimit" optional="true" help="developer option: set trace back limit for python error messages" />
	<param argument="--tblimit" type="integer" min="0" value="0" label="Python traceback limit" optional="true" help="developer option: set trace back limit for python error messages" />

for all parameters. Or if you added new parameters the label should be a short description of the parameter. (The argument is already automatically added to the help. So its clear which Galaxy parameter corresponds to which command line argument.)

[bernt-matthias]

You might want to clarify what the .opt file is and how to specify it.

[bernt-matthias]

I would still suggest to create multiple output collections. One for each output type.

[sgsokol]

Interesting. Do you have examples to point me to?

[bernt-matthias]

The planemo doc cover this a bit. In general the outputs section can contain an arbitrary number of collection and dataset outputs.

Something like this:

<collection name="influx_si_output" type="list:list" label="influx_${si.s_i}_on_${on_string}: log and error files">
    <discover_datasets match_relative_path="true" recurse="true" pattern="(?P&lt;identifier_0&gt;[^/]+)_res/(?P&lt;identifier_1&gt;[^/]+)\.err" ext="txt"/>
    <discover_datasets match_relative_path="true" recurse="true" pattern="(?P&lt;identifier_0&gt;[^/]+)_res/(?P&lt;identifier_1&gt;[^/]+)\.log" ext="txt"/>
</collection>
<collection name="influx_si_output" type="list:list" label="influx_${si.s_i}_on_${on_string}: sim and stats files">
    <discover_datasets match_relative_path="true" recurse="true" pattern="(?P&lt;identifier_0&gt;[^/]+)_res/(?P&lt;identifier_1&gt;[^/]+)\.sim" ext="txt"/>
    <discover_datasets match_relative_path="true" recurse="true" pattern="(?P&lt;identifier_0&gt;[^/]+)_res/(?P&lt;identifier_1&gt;[^/]+)\.stats" ext="txt"/>
</collection>

The first collection contains the log and error files and the second the sim and stats files. Bu using the ext attribute in discover_datasets we set the extension and do not need to move.

If you like to have the extensions in the indentifiers then you could use a discovery rule like (i.e. make the file extension part of the identifier_1 pattern):

    <discover_datasets match_relative_path="true" recurse="true" pattern="(?P&lt;identifier_0&gt;[^/]+)_res/(?P&lt;identifier_1&gt;[^/]+\.stats)" ext="tsv"/>

For really large numbers of possible outputs filters are also a nice option.

[bernt-matthias]

The number of CPU cores must not be under the users control. Use

Suggested change

	#if $opt.np:
	--np='$opt.np'
	--np='$opt.np'
	#end if
	--np='\${GALAXY_SLOTS:-1}'

Then Galaxy will set it automatically

[sgsokol]

And what if user wants to run it sequentially? Or does not want to occupy all cores but only half of them? Or only 2 (as more cores does not necessarily accelerate the calculation but only wastes the ressources)?

[bernt-matthias]

That's a thing a user should not be bothered with. For Galaxy the user does not even know where the code runs. Some Galaxy servers user HPC systems scattered over different countries.

Bottom line is that this should be configured by the admin/galaxy.

There is something called ressource parameters that admins can enable for some tools/users .. then the number of cores would be controllable by the user.

[sgsokol]

Thank you @bgruening and @bernt-matthias for your reviews! You rocks!

Thank you @sgsokol for your contribution and to have apply the advices and requests.

Thanks to all for your advices and corrections.