Specify output directory in `dx run table-exporter` Answered

Luke C Pilling

Hi,

When running table-exporter from the command line is it possible to specify the output directory? The default is to put the new file in the root directory of the project. 

I tried doing -ioutput="files/filename" but table-exporter ends up throwing an error (this from the log):

Log excerpt (last 10 lines) 
+ [[ 0 -eq 0 ]]
+ hdfs dfs -copyToLocal hdfs:///tmp/csv /tmp
+ mkdir -p /home/dnanexus/out/csv
++ echo TSV
++ awk '{print tolower($0)}'
+ file_format=tsv
+ for folder in /tmp/csv/output/*
+ ls '/tmp/csv/output/files/part*.csv'
+ ls -ail ''
ls: cannot access '': No such file or directory

I couldn't see any other output-related options in the documentation (https://documentation.dnanexus.com/developer/apps/developing-spark-apps/table-exporter-application) yet when executing the Tabel Exporter via the app, there is an option to change the output directory.

Thanks,

Luke

Comments

2 comments

Date Votes
  • Comment author
    Daisy V The helpers that keep the community running smoothly. UKB Community team Data Analyst

    Hi Luke,

    Yes, whenever you run any app using dx run you can set the directory for the outputs using the --destination flag. For more info please see Index of dx commands.

    Here is an example command:

    dx run table-exporter -idataset_or_cohort_or_dashboard=<dataset_id> -ifield_names=p31 -ientity=participant --destination test/

    Hope this helps!

    3
  • Comment author
    Luke C Pilling

    Fantastic, thank you. Can't believe I missed that.

    All the best,

    Luke

    0

Please sign in to leave a comment.