README.markdown

git-number

git-number is a perl script that increases my command-line git productivity (with some help from another two perl scripts).

Usage Examples

Here's how it increase my productivity (it might increase yours too):

$ alias gn='git number'
$ alias ga='git number add'

$ gn
# On branch master
# Untracked files:
#   (use "git add <file>..." to include in what will be committed)
#
#1      .README.swp
#2      README
$

Does the output look familiar? Notice the numbers before the filenames? Those are their ids.

If you prefer the short status format then you can use the -s option. This will run git status with the --short option.

$ git number -s
1 ?? .README.swp
2 ?? README

Now look at this:

$ ga 2
git add  README  # <- It does this in the background

$ gn
# On branch master
# Changes to be committed:
#   (use "git reset HEAD <file>..." to unstage)
#
#1      new file:   README
#
# Untracked files:
#   (use "git add <file>..." to include in what will be committed)
#
#2      .README.swp

When run without arguments, 'git number' runs 'git status' and attach a unique number for each line of filename printed by 'git status', and it will 'remember' this number-to-filename association. When run with arguments, like this:

$ git number <any git command> [one or more numbers or git options/args]

'git number' will run that <any git command> and subtitute all the numbers to their equivalent filenames. Non-numeric argument are passed intact to git.

It accepts multiple args and ranges too:

$ ga 2-4 6 10

Which is the same as writing

$ ga 2 3 4 6 10

You can also ask git-number to run arbitrary command instead of git on the given arguments using the -c option:

$ gn -c rm 1

This will run the command "rm README"

The kind of fun that this gives you include the following:

$ alias vn='git number -c vi'
$ vn 1

This will run "vi README"

What's included

1. git-number: Show or operate on files by their 'ids'

2. git-list: List filenames from given ids

3. git-id: Generate and show the file ids

   for example to show the second file run:

    $ git list 2
   

   or to show the first three files, and the 9th and 13th:

    $ git list 1-3 9 13
   

What's not included

Batteries.

How it works

'git-id' is a perl script that does two things:

1. Runs "git status" and inserts a number before each file reported by "git status"
2. Show and save a copy of the output to a file (.git/gitids.txt)

(If you're pedantic then it does four things)

'git-list' is a perl script that converts numbers and ranges to their equivalent filenames from the previous run of 'git-id'.

'git-number' uses 'git-list' to convert all its numbers and ranges arguments to filenames and passes them down to git.

Caveat

1. For a file that is marked as conflicting, the ansi closing color escape sequence printed by git comes after the final newline, which breaks this script a little. This seems to be fixed in latest git.

2. git-number depends on the output of git-status, which is a porcelain. Caveat emptor.

3. It does not work for renames:

    $ git mv a.txt b.txt
    $ gn
    # On branch b
    # Changes to be committed:
    #   (use "git reset HEAD ..." to unstage)
    #
    #1      renamed:    a.txt -> b.txt
    #
    $ gn reset 1  # this will NOT do what you want it to do!

I'm sure there are a few more. Send me a patch :)

Installation

Copy (or make a symbolic link to) 'git-number', 'git-list', 'git-id' into your $HOME/bin directory, or wherever you prefer to put them.

See also

scm_breeze by Nathan Broadbent - similar in spirit with git-number, has more features, requires either bash or zsh.