These scripts serve the purpose of retrieving valuable information from git repositories.
Prints state of git repository at the current commit. Let's assume your repository consists of only single file exact to the following.
using System;
namespace Foo
{
public class Bar
{
public int MyProperty { get; set; }
private int _myField;
public Bar()
{
}
}
}
bash snapshot.sh -r | tee file.csv
Running this command will print the following output in a comma-separated format. When put into an Excel spreadsheet it will look like this:
| File | Lines_count | Commit_count_within_time_span | Leading_spaces |
|---|---|---|---|
| src/file.cs | 17 | 1 | 52 |
- File represents path to file under git version control
- Lines_count simply calculates all lines within file
- Commit_count_within_time_span prints count of commits within a year since now
- Leading_spaces counts all whitespaces appearing before any other character in code (this should more or less represent complexity - the more indentation the more complex the code)
Similar result can be achieved with this oneliner:
git ls-files | xargs -I{} sh -c 'printf "{}\t" ; cat "{}" | printf "`wc -l`\t" ; git log --after="1 year ago" --oneline "{}" | printf "`wc -l`\t" ; grep -o '^[[:blank:]]*' "{}" | tr -d "\n" | wc -c'
Visit git mining plus see some oneliners below:
Prints all files prefixed with number of lines, ordered
git ls-files -z | xargs -0 wc -l | sort -g -r
Prints all meaningful lines, i.e., lines that have at least one number or letter
git ls-files -z | xargs -0 grep '[[:alnum:]]' | wc -l
Print tsv, where first column is file path and second column is meaningful lines count in this file
git ls-files -z | xargs -0 -n 1 sh -c 'printf "$0\t" ; grep "[[:alnum:]]" "$0" | wc -l'
Fast count leading spaces
git ls-files -z | xargs -0 grep -ho '^[[:blank:]]*' | tr -d '\n' | wc -c
Count all tokens
git ls-files -z | xargs -0 grep -h '[[:alnum:]]' | tr -s '[:space:]' '\n' | wc -l