CLOCBear: Add CLOCBear

This bear gives statistics about number of lines,
number of comment lines, and total number of lines.

Closes coala#1577

.ci/deps.sh

@@ -13,7 +13,7 @@ esac
 # apt-get commands
 export DEBIAN_FRONTEND=noninteractive
 
-deps="libclang1-3.4 indent mono-mcs chktex r-base julia golang-go luarocks verilator cppcheck flawfinder devscripts"
+deps="libclang1-3.4 indent mono-mcs chktex r-base julia golang-go luarocks verilator cppcheck flawfinder devscripts cloc"
 deps_infer="m4 opam"
 
 case $CIRCLE_BUILD_IMAGE in

.travis.yml

@@ -43,6 +43,7 @@ addons:
       - opam
       - php-codesniffer
       - verilator
+      - cloc
 
 cache:
   pip: true

bears/general/CLOCBear.py

@@ -0,0 +1,58 @@
+import re
+
+from coalib.bearlib.abstractions.Linter import linter
+from coalib.results.Result import Result
+from dependency_management.requirements.DistributionRequirement import (
+    DistributionRequirement)
+
+
+@linter(executable='cloc',
+        use_stdout=True,
+        use_stderr=False)
+class CLOCBear:
+    """
+    Summarises file of or number of files in directory structure
+    with total lines, number comment lines, actual code lines using tool cloc
+    """
+    LANGUAGES = {'All'}
+    REQUIREMENTS = {DistributionRequirement(apt_get='cloc')}
+    AUTHORS = {'The coala developers'}
+    AUTHORS_EMAILS = {'[email protected]'}
+    LICENSE = 'AGPL-3.0'
+    CAN_DETECT = {'Comment lines','Empty lines'}
+
+    @staticmethod
+    def create_arguments(filename, file, config_file):
+        return ('', filename)
+
+    def process_output(self, output, filename, file):
+        out = output
+        lines = out.split('\n')
+        files = lines[1][:lines[1].find('unique')-1]
+        if(files == 0):
+            msg = 'No valid files. File should belong to valid programming language.'
+            yield Result.from_values(origin = self,
+                                     message = msg,
+                                     file=filename)
+        else:
+            regex = r"(^\S+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)"
+            for row in lines[8:-2]:
+                match = re.search(regex, row)
+                lang = match.group(1)
+                nfiles = match.group(2)
+                blank = (int)(match.group(3))
+                comment = (int)(match.group(4))
+                code = (int)(match.group(5))
+                total = blank + comment + code
+                report = \
+                """Language: {0}
+                Total files: {1}
+                Total lines: {2}
+                Code lines: {3}({4:.2f}%)
+                Comment lines: {5}({6:.2f}%)
+                Blank lines: {7}({8:.2f}%)
+                """\
+                .format(lang, nfiles, total, code, code*100./total, comment, comment*100./total, blank, blank*100./total)
+                yield Result.from_values(origin = self,
+                                         message = report,
+                                         file = filename)

tests/general/CLOCBearTest.py

@@ -0,0 +1,12 @@
+from bears.general.CLOCBear import CLOCBear
+from coalib.testing.LocalBearTestHelper import verify_local_bear
+
+test_file = """
+int main() {
+    return 0;
+}
+"""
+
+CLOCBearTest = verify_local_bear(CLOCBear,
+                                 valid_files=(),
+                                 invalid_files=(test_file,))