path: root/git-coverage

#!/usr/bin/env python

import fnmatch
import getopt
import os
import re
import subprocess
import sys

COVERAGE_EXTENSIONS = [".c", ".cpp", ".cc"]

# Portions of the code:
#
# Copyright (C) 2005-2010 Aaron Bentley, Canonical Ltd
# <aaron.bentley@utoronto.ca>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
#

BINARY_FILES_RE = re.compile('Binary files (.*) and (.*) differ\n')

class MalformedHunkHeader(Exception):
    def __init__(self, desc, line):
        self.desc = desc
        self.line = line

    def __str__(self):
        return "Malformed hunk header.  %(desc)s\n%(line)r" % self.__dict__

class BadPatch(Exception):
    def __init(self, message):
        self.message = message

    def __str__(self):
        return self.message


def get_patch_names(iter_lines):
    try:
        line = iter_lines.next()
        if not line.startswith("--- "):
            raise BadPatch("No orig name: %s" % line)
        else:
            orig_name = line[4:].rstrip("\n")
    except StopIteration:
        raise BadPatch("No orig line")
    try:
        line = iter_lines.next()
        if not line.startswith("+++ "):
            raise BadPatch("No mod name")
        else:
            mod_name = line[4:].rstrip("\n")
    except StopIteration:
        raise BadPatch("No mod line")
    return (orig_name, mod_name)


def parse_range(textrange):
    tmp = textrange.split(',')
    if len(tmp) == 1:
        pos = tmp[0]
        range = "1"
    else:
        (pos, range) = tmp
    pos = int(pos)
    range = int(range)
    return (pos, range)


def hunk_from_header(line):
    matches = re.match(r'\@\@ ([^@]*) \@\@( (.*))?\n', line)
    if matches is None:
        raise MalformedHunkHeader("Does not match format.", line)
    try:
        (orig, mod) = matches.group(1).split(" ")
    except (ValueError, IndexError), e:
        raise MalformedHunkHeader(str(e), line)
    if not orig.startswith('-') or not mod.startswith('+'):
        raise MalformedHunkHeader("Positions don't start with + or -.", line)
    try:
        (orig_pos, orig_range) = parse_range(orig[1:])
        (mod_pos, mod_range) = parse_range(mod[1:])
    except (ValueError, IndexError), e:
        raise MalformedHunkHeader(str(e), line)
    if mod_range < 0 or orig_range < 0:
        raise MalformedHunkHeader("Hunk range is negative", line)
    tail = matches.group(3)
    return Hunk(orig_pos, orig_range, mod_pos, mod_range, tail)


class HunkLine:
    def __init__(self, contents):
        self.contents = contents


class ContextLine(HunkLine):
    def __init__(self, contents):
        HunkLine.__init__(self, contents)


class InsertLine(HunkLine):
    def __init__(self, contents):
        HunkLine.__init__(self, contents)


class RemoveLine(HunkLine):
    def __init__(self, contents):
        HunkLine.__init__(self, contents)


def parse_line(line):
    if line.startswith("\n"):
        return ContextLine(line)
    elif line.startswith(" "):
        return ContextLine(line[1:])
    elif line.startswith("+"):
        return InsertLine(line[1:])
    elif line.startswith("-"):
        return RemoveLine(line[1:])
    else:
        raise BadPatch("Unknown line type" % line)


class Hunk:
    def __init__(self, orig_pos, orig_range, mod_pos, mod_range, tail=None):
        self.orig_pos = orig_pos
        self.orig_range = orig_range
        self.mod_pos = mod_pos
        self.mod_range = mod_range
        self.tail = tail
        self.lines = []


def iter_hunks(iter_lines, allow_dirty=False):
    hunk = None
    for line in iter_lines:
        if line == "\n":
            if hunk is not None:
                yield hunk
                hunk = None
            continue
        if hunk is not None:
            yield hunk
        try:
            hunk = hunk_from_header(line)
        except MalformedHunkHeader:
            if allow_dirty:
                # If the line isn't a hunk header, then we've reached the end
                # of this patch and there's "junk" at the end.  Ignore the
                # rest of this patch.
                return
            raise
        orig_size = 0
        mod_size = 0
        while orig_size < hunk.orig_range or mod_size < hunk.mod_range:
            hunk_line = parse_line(iter_lines.next())
            hunk.lines.append(hunk_line)
            if isinstance(hunk_line, (RemoveLine, ContextLine)):
                orig_size += 1
            if isinstance(hunk_line, (InsertLine, ContextLine)):
                mod_size += 1
    if hunk is not None:
        yield hunk


class Patch(object):
    def __init__(self, oldname, newname):
        self.oldname = oldname
        self.newname = newname
        self.hunks = []


def parse_patch(iter_lines, allow_dirty=False):
    (orig_name, mod_name) = get_patch_names(iter_lines)
    patch = Patch(orig_name, mod_name)
    for hunk in iter_hunks(iter_lines, allow_dirty):
        patch.hunks.append(hunk)
    return patch


def iter_file_patch(iter_lines, allow_dirty=False):
    saved_lines = []
    orig_range = 0
    beginning = True
    for line in iter_lines:
        if BINARY_FILES_RE.match(line):
            continue
        if line.startswith('=== ') or line.startswith('*** '):
            continue
        if line.startswith('#'):
            continue
        elif orig_range > 0:
            if line.startswith('-') or line.startswith(' '):
                orig_range -= 1
        elif line.startswith('--- '):
            if allow_dirty and beginning:
                # Patches can have "junk" at the beginning
                # Stripping junk from the end of patches is handled when we
                # parse the patch
                beginning = False
            elif len(saved_lines) > 0:
                yield saved_lines
            saved_lines = []
        elif line.startswith('@@'):
            hunk = hunk_from_header(line)
            orig_range = hunk.orig_range
        saved_lines.append(line)
    if len(saved_lines) > 0:
        yield saved_lines


#
# End of code from bzr
# ---------------------------------------------------------------------------

def iter_process_lines(argv):
    proc = subprocess.Popen(argv, stdout=subprocess.PIPE)
    while True:
        line = proc.stdout.readline()
        if line != "":
            yield line
        else:
            return

def iter_desired_coverage(iter_lines):
    for fil in iter_file_patch(iter_lines, allow_dirty=True):
        patch = parse_patch(fil.__iter__(), allow_dirty=True)
        filename = os.path.normpath(patch.newname.split("/", 1)[1])
        (root, ext) = os.path.splitext(filename)
        if ext not in COVERAGE_EXTENSIONS:
            continue
        for hunk in patch.hunks:
            offset = hunk.mod_pos
            for line in hunk.lines:
                if isinstance(line, RemoveLine):
                    yield (filename, offset, line.contents)
                elif isinstance(line, InsertLine):
                    yield (filename, offset, line.contents)
                    offset += 1
                else:
                    offset += 1

def find_all_gcno_files(directory):
    paths = []
    def visit(paths, dirname, names):
        for name in names:
            path = os.path.normpath(os.path.join(dirname, name))
            if os.path.isdir(path):
                continue
            if not fnmatch.fnmatch(name, "*.gcno"):
                continue
            (base, ext) = os.path.splitext(path)
            if not os.path.exists(base + ".gcda"):
                continue
            paths.append(path)
    os.path.walk(".", visit, paths)
    return paths


def match_gcno_files(filename, gcno_cache):
    matches = []
    (directory, base) = os.path.split(filename)
    (base, ext) = os.path.splitext(base)
    match = "%s/*%s.gcno" % (directory, base)
    for gcno in gcno_cache:
        if fnmatch.fnmatch(gcno, match):
            matches.append(gcno)
    return matches

CREATING_RE = re.compile(".*'(.+\.gcov)'.*")

def gcov_lines_for_files(filename, gcno_cache):
    gcno = match_gcno_files(filename, gcno_cache)
    if not gcno:
        return

    # gcov wants to be in the directory with the source files
    # so we make all the gcno paths absolute and change to that
    # directory

    absgcno = [os.path.abspath(path) for path in gcno]
    (directory, base) = os.path.split(filename)

    oldpwd = os.getcwd()
    os.chdir(directory)

    # We scrape the output of the command for the names of the
    # gcov files created, which we process, and then remove
    gcovs = []

    cmd = ['gcov', '--preserve-paths', '--relative-only']
    for line in iter_process_lines(cmd + absgcno):
        match = CREATING_RE.match(line)
        if not match:
            continue
        gcov = match.group(1)
        if os.path.exists(gcov):
            gcovs.append(os.path.abspath(gcov))

    # Because we change the directory, we have to take care not
    # to yield while the current directory is changed

    os.chdir(oldpwd)

    for gcov in gcovs:
        with open(gcov, 'r') as f:
            for l in f:
                yield l
            os.unlink(gcov)


def gcov_coverage_lines(gcov_iter):
    coverage = { }
    for line in gcov_iter:

        # Each gcov coverage output line looks something like this
        #  coverage:   lineno:  remainder is actual line content
        parts = line.split(':', 2)
        if len(parts) != 3:
            continue

        covered = parts[0].strip()
        try:
            no = int(parts[1].strip())
            if covered == '-':
                count = 0
            else:
                count = int(covered)
            coverage[no] = parts[2]
        except ValueError:
            pass
    return coverage


def main(argv):
    try:
        opts, args = getopt.getopt(argv[1:], "h", ["help"])
    except getopt.GetoptError, err:
        print str(err)
        return 2
    for o, a in opts:
        if o in ("-h", "--help"):
            usage()
            return 0
        else:
            assert False, "unhandled option"

    if len(args) == 0:
        cmd = ['git', 'diff', 'HEAD']
    else:
        cmd = ['git', 'show'] + args

    # First process all the lines that we need
    needed_coverage = { }
    iterator = iter_process_lines(cmd)
    for (filename, no, content) in iter_desired_coverage(iterator):
        if filename not in needed_coverage:
            needed_coverage[filename] = { }
        needed_coverage[filename][no] = content

    gcno_cache = find_all_gcno_files(".")

    for (filename, needed) in needed_coverage.items():
        gcov = gcov_lines_for_files(filename, gcno_cache)
        coverage = gcov_coverage_lines(gcov)

        for (no, content) in needed.items():
            if no not in coverage:
                print filename, no, content,

if __name__ == '__main__':
    sys.exit(main(sys.argv))