Count number of code lines user wrote in current repo state

[valeriupredoi]

If anyone needs this for their HR/organizational contributions/show-off etc, there is a nice way to get the number of lines each GitHub author contributed to the current state of the repository:

git ls-files | while read f; do git blame -w --line-porcelain -- "$f" | grep -I '^author '; done | sort -f | uniq -ic | sort -n

This might also help us identify users' contributions (not just by using the GH stats measured in commits) and promoting select users/praising them etc. Fun command, don't start bragging now 😁

In ESMValTool we have the top 10 liners as:

   6194 author Bettina Gier
   7234 author Lee de Mora
   7727 author Ruth Lorenz
   7787 author jhardenberg
  13637 author Axel Lauer
  16807 author Valeriu Predoi
  17094 author Bouwe Andela
  22632 author Mattia Righi
  29716 author Björn Brötz
  48391 author Manuel Schlund

@schlunma will officially be known as Ernest Hemingway from now on 😆

[zklaus]

I agree, it's certainly important to show that you don't need to spend many lines of code and that your code reaches old age without needing too many modifications. Cf https://stackoverflow.com/questions/184071/when-if-ever-is-number-of-lines-of-code-a-useful-metric etc.

[valeriupredoi]

Nice one, K-man! The problem is that some institutions/individuals still use this metric as a standard, if I had a penny for how many times I got asked "and how many lines of code did you contribute to project X or Y?" during job interviews or assessment panels, I'd be Bill Gates rich by now 😁 On the other hand, we could use such a metric from time to time, to see if we have significantly grown our diagnostics base (unlike in Core), we are interested in volume in Tool 😛

[zklaus]

From previous discussions (difficult testing, maintenance of old diagnostics, duplication in preprocessors, etc.) it certainly seems that our problems are rather too many LOCs. That doesn't mean we need to reduce functionality. Standard approaches, like relying on libraries, increasing code reuse, pruning deprecated or outdated features, all help to keep the lines of code down without compromising functionality.

I have to confess that I have never been asked how many lines of code I have contributed, neither when working in the private sector nor in the public sector. But maybe that is exceptional? 🤷‍♂️

[valeriupredoi]

It happened to me a lot in the academic sector, hopefully things have changed by now there too, given that I left academia five years ago and I hear they've aligned themselves to many other modern things (including using GitHub 😆 )

[jvegreg]

I knew of a company that used code lines as a metric of your daily!! production.

Programmers just shuffle blank lines and splitted one liners to meet the expected goal...