Question: HTML output directory question
0
gravatar for danielfortin86
23 months ago by
United States
danielfortin86100 wrote:

I'm trying to write an xml wrapper for a script which generates an html file ("report.html") in an output directory specified with the -o parameter. The script correctly executes but when I click on the "information" icon, the command line shows me the path for "$html_file.files_path" yet the directory does not exist. Can anyone give me any suggestions? Relevant lines of xml are below:

 <command>python script.py -o "$html_file.files_path" </command>

 

  <outputs>
    <data name="html_file" format="html" from_work_dir="report.html" label="Output (HTML) on ${on_string}"/>
  </outputs>

when I look at the command line, the output directory is specified as: 

/remote/home/user/galaxy/database/job_working_directory/000/576/dataset_687_files

but when I search for the output file, it is found here:

/remote/home/user/galaxy/database/files/000/dataset_687_files/

Thanks!

 

tool html custom • 817 views
ADD COMMENTlink modified 23 months ago • written 23 months ago by danielfortin86100

This appears to be the same as this issue:

Galaxy Passing An Incorrect Directory To A Script...

However, if I use the "extra_files_path" i still get the same result in that the data is placed in "files/000/dataset_xxx_files/" rather than  "job_working_directory/000/XXX/dataset_xxx_files" 

ADD REPLYlink modified 23 months ago • written 23 months ago by danielfortin86100

ok - so that old difference between extra_files_path and files_path has been resolved - that's good!

At completion of a job, galaxy moves things from the transient job working directory to permanent storage as you've noticed. That means tools can happily spew out temporary messy stuff to that transient working directory but at the end, only the tool code specified and user-ready outputs are available for viewing in a history. I'm not sure what you were expecting or why this is a problem? 

ADD REPLYlink written 23 months ago by fubar1.1k

Sorry, i just realized I switched the directory names, I've since changed the comment..

In the <command> I'm passing the temporary directory, the results get copied to the final spot. However, the html file I receive from the galaxy interface is empty (the file report.html). 

The instructions for doing this seem to suggest that I'm doing everything correctly:

https://wiki.galaxyproject.org/Admin/Tools/Multiple%20Output%20Files

Any suggestions would be greatly appreciated

ADD REPLYlink modified 23 months ago • written 23 months ago by danielfortin86100

Maybe this is a bug, but it works fine if I add this to the <command>:

; cp $html_file.files_path/report.html $html_file

ADD REPLYlink written 23 months ago by danielfortin86100
0
gravatar for fubar
23 months ago by
fubar1.1k
Australia
fubar1.1k wrote:

Try $html_file.extra_files_path in that command line ? 

ADD COMMENTlink written 23 months ago by fubar1.1k

Thanks for the suggestion, but I got the same result

ADD REPLYlink written 23 months ago by danielfortin86100
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 115 users visited in the last hour