How to automatically track the work log time?

[gudata]

Hello,
Is there a known way in which i do something like that

jira worklog start ISSUE

...work

jira worklog end ISSUE 

and jira logs the endtime - startime ?

It can be done with additional scripting but wondering how you doi t?

[colton22]

So I already had something similar to this setup.. Very caveman style but re-flavor it as you wish.

12:28:42 colton22:~ > cat bin/worklog 
#!/bin/bash

if [ "$1" == "" ] && [ "$2" == "" ]; then
    echo -e "usage: worklog ticket action\n  -action <start|stop>";
    exit;
fi

act=$(echo "$2" |tr '[:upper:]' '[:lower:]');
tkt=$1;

if [ "$act" == "reset" ]; then
    rm ~/.worklog/$tkt;
    echo "$tkt reset!";
    exit;
fi
if [ "$act" != "start" ] && [ "$act" != "stop" ]; then
    echo -e "Action must be start or stop!";
    exit;
fi

if [ "$(tail -n 1 ~/.worklog/$tkt 2>/dev/null |awk -F '|' '{print $1}')" == "$act" ]; then
    echo "$tkt already ${act}ed!";
    exit;
fi
echo "$act|$(date +%s)" >> ~/.worklog/$tkt

if [ "$act" == "start" ]; then
    echo "Logged Start Time in ~/.worklog/$tkt";
else
    #use tail -r on bsd
    log=$(tac ~/.worklog/$tkt);
    sec=0;
    lstop=0;
    bifs=$IFS;
    IFS=$'\n';
    for i in $(echo -e "$log");do
        a=$(echo "$i" |awk -F '|' '{print $1}');
        v=$(echo "$i" |awk -F '|' '{print $2}');
        if [ "$a" == "stop" ]; then
            lstop=$v;
        else
            dif=$(( $lstop - $v ));
            sec=$(( $sec + $dif ));
        fi
    done
    echo "You have worked on $tkt for ${sec}s";
fi
12:28:46 colton22:~ > 

I know this does not answer your question but might be a workaround for now. Note you will have to mkdir ~/.worklog to make it work.

[coryb]

I think you can create a few simple aliases in $HOME/.jira.d/config.yml for startwork and endwork:

custom-commands:
  - name: startwork
    help: mark the start of work on an issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
    script: |-
      mkdir -p $HOME/.jira.d/work
      touch -p $HOME/.jira.d/work/{{args.ISSUE}}

  - name: endwork
     help: submit worklog for issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
    script: |-
      starttime=$(stat -f "%Sm" -t "%Y-%m-%dT%H:%M:%S.000%z" $HOME.jira.d/work/{{args.ISSUE}})
      {{jira}} worklog add {{args.ISSUE}} -S "$starttime"

This effectively allows for jira startwork ISSUE-123 and jira endwork ISSUE-123. The startwork command really just touches a file. The endwork command will then read the last modified time for the previously touched file and call jira worklog add ....

I don't use worklogs myself, so I am not 100% sure this will work, but it is where I would start.

[dimaosipa]

@coryb thank you for pointing out to the right direction. I fixed a couple of errors – now it works for me.

Here's my version of @coryb's script:

custom-commands:
  - name: startwork
    help: mark the start of work on an issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
    script: |-
      mkdir -p $HOME/.jira.d/work
      touch $HOME/.jira.d/work/{{args.ISSUE}}

  - name: endwork
    help: submit worklog for issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
    script: |-
      if [ ! -f $HOME/.jira.d/work/{{args.ISSUE}} ]; then
        echo "You must execute jira startwork ISSUE first"
        exit 1
      fi
      startdate=$(stat -f "%Sm" -t "%Y-%m-%dT%H:%M:%S.000%z" $HOME/.jira.d/work/{{args.ISSUE}})
      starttime=$(stat -f "%Sm" -t "%s" $HOME/.jira.d/work/{{args.ISSUE}})
      endtime=$(date +%s)
      diff=$((($endtime-$starttime)/60))
      if (( $diff > 0 )); then
        {{jira}} worklog add --noedit -S "$startdate" -T "$diff"m {{args.ISSUE}}
        rm $HOME/.jira.d/work/{{args.ISSUE}}
      fi

[gueldenstone]

Thank @bomjkolyadun! Just wanted to note, that this script only works with the macOS version of stat!

This one should work with linux:

custom-commands:
  - name: startwork
    help: mark the start of work on an issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
    script: |-
      mkdir -p $HOME/.jira.d/work
      touch $HOME/.jira.d/work/{{args.ISSUE}}

  - name: endwork
    help: submit worklog for issue
    args:
      - name: ISSUE
        help: issue to mark work being started on
        required: true
      - name: comment
        help: comment for the worklog
        required: true
    script: |-
      if [ ! -f $HOME/.jira.d/work/{{args.ISSUE}} ]; then
        echo "You must execute jira startwork ISSUE first"
        exit 1
      fi
      startdate=$(echo -n @$(stat -c "%X" $HOME/.jira.d/work/{{args.ISSUE}}) | xargs date +%Y-%m-%dT%H:%M:%S.000%z -d)
      starttime=$(stat -c "%X" $HOME/.jira.d/work/{{args.ISSUE}})
      endtime=$(date +%s)
      diff=$((($endtime-$starttime)/60))
      if [ "$diff" -gt 0 ]; then
        echo "logegd $diff m"
        {{jira}} worklog add --noedit -S "$startdate" -T "$diff"m --comment "{{args.comment}}" {{args.ISSUE}}
        rm $HOME/.jira.d/work/{{args.ISSUE}}
      fi