This program started as a hack, bending and twisting pylint.py by Tim Peters. At that time I was interested on code metrics and thought let's do something quick and dirty. pycount then and today scans a Python file and reports back various cate- gories of lines it thinks it has found, like comments, doc strings, blank lines and, sometimes, real code as well.

The output can be either as a table with only the number of lines found for each file or as a listing of the file(s) prefixed with some running numbers and classification for each line.

The former is useful to scan a whole project e.g. when you need to know if the project is documented well or at all and where this info can be found. The latter is at least a nice new view to your own sources or that of others if nothing else!

There are a couple of minor known bugs with pycount like: Doc strings must be tripple-quoted ones otherwise they are classified as normal source code. Continuation lines ending with a backslash are not treated at all. Complex regular ex- pressions (as in pycount itself) can knock the parser down, quickly. There is a built-in quick-and-dirty solution to this which might work whenever the problem is on one line only. But in "most cases" it works...

Usage:

pycount.py [-v] <file1> [<file2> ...]
pycount.py [-v] <expr>
pycount.py [-F] <linetypes> <file>
pycount.py [-R] <expr>

where <fileN>     is a Python file (usually ending in .py),
      <linetypes> is a comma-seperated list of python line
                  type codes (code, comment, doc string, blank)
                  e.g. '###' or 'DOC,###,---'
      <expr>      is a shell expression with meta-characters
                  (note that for -R you must quote it)
      -v          verbose flag, listing the source
      -F          filter flag, listing the filtered source
      -R          apply recursively on subdirectories