repo_metrics.erl

-module(repo_metrics).





-export([

    file_as_string/1,

    nonws_lines/1,
    nonws_linecount_string/1,
    nonws_linecount_file/1,

    git_churn/0,
      git_churn/1,
      git_churn/2,

    histograph/1,

    to_daterange/1,

    churn/0,
      churn/1,
      churn/2

]).



file_as_string(Filename) ->

    {ok, FaS} = file:read_file(Filename),
    binary_to_list(FaS).



nonws_lines(String)            -> string:tokens(String, "\r\n").
nonws_linecount_string(String) -> length(nonws_lines(String)).
nonws_linecount_file(Filename) -> nonws_linecount_string(file_as_string(Filename)).



to_daterange(none) -> "";
to_daterange(DR)   -> " --since='" ++ DR ++ "'".



to_dir(none) -> "";
to_dir(Dir)  -> " -C " ++ Dir.



git_churn()               -> git_churn(none, none).
git_churn(DateRange)      -> git_churn(none, DateRange).
git_churn(Dir, DateRange) ->

    nonws_lines(os:cmd(
        "git" ++ to_dir(Dir) ++ " log --all -M --name-only --format='format:'" ++ to_daterange(DateRange) ++ " | grep -v '^$'"
    )).



histograph([])                      -> [];
histograph(List) when is_list(List) ->

    [Head|Tail] = lists:sort(List),
    histo_count(Tail, Head, 1, []).



histo_count( [],             Current, Count, Work) -> lists:reverse([{Current,Count}]++Work);
histo_count( [Current|Tail], Current, Count, Work) -> histo_count(Tail, Current, Count+1, Work);
histo_count( [New|Tail],     Current, Count, Work) -> histo_count(Tail, New,     1,       [{Current,Count}] ++ Work).




churn_row(File, ChurnCt) ->

    NLF = nonws_linecount_file(File),
    {File, ChurnCt, NLF, ChurnCt / NLF}.



churn()               -> churn(none, none).
churn(DateRange)      -> churn(none, DateRange).
churn(Dir, DateRange) ->

    [ churn_row(File, ChurnCt) || {File, ChurnCt} <- histograph(git_churn(Dir, DateRange)) ].