How do I exclude directories when using git log?

Question

Aim

I would like to know the amount of lines changed in a particular repository over a period of time and print this into a simple text file. However, when I run the code below I receive an error that I'm not sure how to solve.

The code

git log --stat --since="2015-01-01" --until="2015-11-16" `find . \( ! -path "*[PATH]*" -a ! -path "*[PATH2]*" \)` | awk -F',' "/files? changed/ {
    files += \$1
    insertions += \$2
    deletions += \$3
    destination = \"$DESTINATION\"}

END {
    print \"Files Changed: \" files > destination
    print \"Insertions: \" insertions > destination
    print \"Deletions: \" deletions > destination
    print \"Lines changed: \" insertions + deletions > destination}"

The error

fatal: ambiguous argument '[PATH]': unknown revision or path not in the working tree.
Use '--' to separate paths from revisions
awk: cmd. line:7: fatal: expression for `>' redirection has null string value

Other useful information

• The version of Git used is 1.7.1. Therefore I cannot use this: --oneline --format=%s -- . ":(exclude)[PATH]"
• I know this code works when I do not try to exclude directories.
• I also know that the "find" works correctly when run separately.

Question

Could someone please help me understand what I've done wrong or provide a better way of excluding multiple directories?

Any help would be much appreciated.

Answer 1

The error message written by git tells how to fix the error. The problem is that, sometimes, git is unable to tell if the first argument it is given is a path or a revision. The -- can be used to disambiguate arguments.

Asides from this, there is a few other minor problems in your code. The shell also understands simple quotes, where variables are not expanded. You definitely should write your awk script in simple quotes, as it prevents clumsy escape sequences to spam your code.

git log --stat --since='2015-01-01' --until='2015-11-16' $(find . \( ! -path '*[PATH]*' -a ! -path '*[PATH2]*' \)) | awk -F',' '
/files? changed/ {
  files += $1
  insertions += $2
  deletions += $3
}

END {
  print "Files Changed: " files
  print "Insertions: " insertions
  print "Deletions: " deletions
  print "Lines changed: " insertions + deletions
}' > "${DESTINATION}"

Also, if you are writing output to only one file, it is easier to let awk write its report to stdout and redirect it to the file you want, rather than doing output redirections in awk itself.