how to slim the git bulk -q summary --oneline command [Request for ideas]

[guenhter]

Hi,

when you combine the commands git bulk and git summary --oneline you get a very nice overview of your all your current repos like this:

$ git bulk -q summary --oneline | column -t -s /                                                                                                                                                           
Core command "summary" accepted.
deno-data-handler-prototype          age: 7 weeks                last active: 5 weeks ago             active on 5 days      commits: 21      uncommitted: 0
udp-handler                          age: 7 weeks                last active: 7 weeks ago             active on 1 days      commits: 4       uncommitted: 0
azure-client                         age: 7 weeks                last active: 7 weeks ago             active on 1 days      commits: 3       uncommitted: 0
dev-infrastructure-automation        age: 6 weeks                last active: 2 hours ago             active on 13 days     commits: 23      uncommitted: 0
gitlab-project-settings-corrector    age: 3 months               last active: 2 days ago              active on 2 days      commits: 3       uncommitted: 0
free-for-all                         age: 2 days                 last active: 2 days ago              active on 1 days      commits: 2       uncommitted: 0
realm-initializer                    age: 3 months               last active: 2 days ago              active on 6 days      commits: 23      uncommitted: 0
gitlab-k8s-cleaner                   age: 3 months               last active: 3 months ago            active on 2 days      commits: 7       uncommitted: 0
...

Event though it is pretty nice, there is much redundant information (like the title of each command is contained in each line in each column) what makes it hard to just catch the relevant information.

What I'd have in mind is something like this:

$ git bulk -q summary ......
Core command "summary" accepted.

Project Name                         Age        Last Active    Active On   Commits Uncommited
deno-data-handler-prototype          7 weeks    5 weeks ago    5 days      21      0
udp-handler                          7 weeks    7 weeks ago    1 days      4       0
azure-client                         7 weeks    7 weeks ago    1 days      3       0
dev-infrastructure-automation        6 weeks    2 hours ago    13 days     23      0
gitlab-project-settings-corrector    3 months   2 days ago     2 days      3       0
free-for-all                         2 days     2 days ago     1 days      2       0
realm-initializer                    3 months   2 days ago     6 days      23      0
gitlab-k8s-cleaner                   3 months   3 months ago   2 days      7       0
....

This would be slimmer.

The Problem: Git summary doesn't know that it is used in a bulk command and therefore just prints the title on each field.
I'm more than willing to implement this, but I want to collect ideas how this can be archived the best.

[spacewander]

The (maybe) simplest one: use pipeline. For instance, git bulk -q summary ...... | a python script.

[guenhter]

I'd somehow favor to add this in some way to the tool so that others can use it as well. Having in in a python script on my end means I have to create some code what is only there for me.

[guenhter]

What do you think of this approaches:

1. The "git bulk summary" is specially treated and knows what do do. Then no --oneline | column -t -s / is needed any more because the command is doing the job
2. The git summary command is able to print out the information of the repo in a tabelar form: 1st row are the headers, 2nd row are the data. This is done for each repo and I simply ignore the headers then with git bulk -q summary --oneline | grep something-to-swallow-headers | column -t -s /| but I keep the very first header.

[spacewander]

I vote for the second one.