>>> import sys, platform
>>> print '%s %s, Python %s' %(platform.system(), platform.release(), platform.python_version())
Windows XP, Python 2.7.11
>>> sys.version
'2.7.11 (v2.7.11:6d1b6a68f775, Dec 5 2015, 20:32:19) [MSC v.1500 32 bit (Intel)]' 

import os, sys
rawCountInfo = [0, 0, 0, 0, 0]
detailCountInfo = [] 

 def CalcLinesCh(line, isBlockComment):
 lineType, lineLen = 0, len(line)
 if not lineLen:
  return lineType

 line = line + '\n' #添加一个字符防止iChar+1时越界
 iChar, isLineComment = 0, False
 while iChar < lineLen:
  if line[iChar] == ' ' or line[iChar] == '\t': #空白字符
   iChar += 1; continue
  elif line[iChar] == '/' and line[iChar+1] == '/': #行注释
   isLineComment = True
   lineType |= 2; iChar += 1 #跳过'/'
  elif line[iChar] == '/' and line[iChar+1] == '*': #块注释开始符
   isBlockComment[0] = True
   lineType |= 2; iChar += 1
  elif line[iChar] == '*' and line[iChar+1] == '/': #块注释结束符
   isBlockComment[0] = False
   lineType |= 2; iChar += 1
  else:
   if isLineComment or isBlockComment[0]:
    lineType |= 2
   else:
    lineType |= 1
  iChar += 1

 return lineType #Bitmap：0空行，1代码，2注释，3代码和注释

def CalcLinesPy(line, isBlockComment):
 #isBlockComment[single quotes, double quotes]
 lineType, lineLen = 0, len(line)
 if not lineLen:
  return lineType

 line = line + '\n\n' #添加两个字符防止iChar+2时越界
 iChar, isLineComment = 0, False
 while iChar < lineLen:
  if line[iChar] == ' ' or line[iChar] == '\t': #空白字符
   iChar += 1; continue
  elif line[iChar] == '#':   #行注释
   isLineComment = True
   lineType |= 2
  elif line[iChar:iChar+3] == "'''": #单引号块注释
   if isBlockComment[0] or isBlockComment[1]:
    isBlockComment[0] = False
   else:
    isBlockComment[0] = True
   lineType |= 2; iChar += 2
  elif line[iChar:iChar+3] == '"""': #双引号块注释
   if isBlockComment[0] or isBlockComment[1]:
    isBlockComment[1] = False
   else:
    isBlockComment[1] = True
   lineType |= 2; iChar += 2
  else:
   if isLineComment or isBlockComment[0] or isBlockComment[1]:
    lineType |= 2
   else:
    lineType |= 1
  iChar += 1

 return lineType #Bitmap：0空行，1代码，2注释，3代码和注释 

 from ctypes import c_uint, c_ubyte, CDLL
CFuncObj = None
def LoadCExtLib():
 try:
  global CFuncObj
  CFuncObj = CDLL('CalcLines.dll')
 except Exception: #不捕获系统退出(SystemExit)和键盘中断(KeyboardInterrupt)异常
  pass

def CalcLines(fileType, line, isBlockComment):
 try:
  #不可将CDLL('CalcLines.dll')放于本函数内，否则可能严重拖慢执行速度
  bCmmtArr = (c_ubyte * len(isBlockComment))(*isBlockComment)
  CFuncObj.CalcLinesCh.restype = c_uint
  if fileType is 'ch': #is(同一性运算符)判断对象标识(id)是否相同，较==更快
   lineType = CFuncObj.CalcLinesCh(line, bCmmtArr)
  else:
   lineType = CFuncObj.CalcLinesPy(line, bCmmtArr)

  isBlockComment[0] = True if bCmmtArr[0] else False
  isBlockComment[1] = True if bCmmtArr[1] else False
  #不能采用以下写法，否则本函数返回后isBlockComment列表内容仍为原值
  #isBlockComment = [True if i else False for i in bCmmtArr]
 except Exception, e:
  #print e
  if fileType is 'ch':
   lineType = CalcLinesCh(line, isBlockComment)
  else:
   lineType = CalcLinesPy(line, isBlockComment)

 return lineType 

from cffi import FFI
CFuncObj, ffiBuilder = None, FFI()
def LoadCExtLib():
 try:
  global CFuncObj
  ffiBuilder.cdef('''
  unsigned int CalcLinesCh(char *line, unsigned char isBlockComment[2]);
  unsigned int CalcLinesPy(char *line, unsigned char isBlockComment[2]);
  ''')
  CFuncObj = ffiBuilder.dlopen('CalcLines.dll')
 except Exception: #不捕获系统退出(SystemExit)和键盘中断(KeyboardInterrupt)异常
  pass

def CalcLines(fileType, line, isBlockComment):
 try:
  bCmmtArr = ffiBuilder.new('unsigned char[2]', isBlockComment)
  if fileType is 'ch': #is(同一性运算符)判断对象标识(id)是否相同，较==更快
   lineType = CFuncObj.CalcLinesCh(line, bCmmtArr)
  else:
   lineType = CFuncObj.CalcLinesPy(line, bCmmtArr)

  isBlockComment[0] = True if bCmmtArr[0] else False
  isBlockComment[1] = True if bCmmtArr[1] else False
  #不能采用以下写法，否则本函数返回后isBlockComment列表内容仍为原值
  #isBlockComment = [True if i else False for i in bCmmtArr]
 except Exception, e:
  #print e
  if fileType is 'ch':
   lineType = CalcLinesCh(line, isBlockComment)
  else:
   lineType = CalcLinesPy(line, isBlockComment)

 return lineType 

def SafeDiv(dividend, divisor):
 if divisor: return float(dividend)/divisor
 elif dividend:  return -1
 else:    return 0

gProcFileNum = 0
def CountFileLines(filePath, isRawReport=True, isShortName=False):
 fileExt = os.path.splitext(filePath)
 if fileExt[1] == '.c' or fileExt[1] == '.h':
  fileType = 'ch'
 elif fileExt[1] == '.py': #==(比较运算符)判断对象值(value)是否相同
  fileType = 'py'
 else:
  return

 global gProcFileNum; gProcFileNum += 1
 sys.stderr.write('%d files processed...\r'%gProcFileNum)

 isBlockComment = [False]*2 #或定义为全局变量，以保存上次值
 lineCountInfo = [0]*5  #[代码总行数, 代码行数, 注释行数, 空白行数, 注释率]
 with open(filePath, 'r') as file:
  for line in file:
   lineType = CalcLines(fileType, line.strip(), isBlockComment)
   lineCountInfo[0] += 1
   if lineType == 0: lineCountInfo[3] += 1
   elif lineType == 1: lineCountInfo[1] += 1
   elif lineType == 2: lineCountInfo[2] += 1
   elif lineType == 3: lineCountInfo[1] += 1; lineCountInfo[2] += 1
   else:
    assert False, 'Unexpected lineType: %d(0~3)!' %lineType

 if isRawReport:
  global rawCountInfo
  rawCountInfo[:-1] = [x+y for x,y in zip(rawCountInfo[:-1], lineCountInfo[:-1])]
  rawCountInfo[-1] += 1
 elif isShortName:
  lineCountInfo[4] = SafeDiv(lineCountInfo[2], lineCountInfo[2]+lineCountInfo[1])
  detailCountInfo.append([os.path.basename(filePath), lineCountInfo])
 else:
  lineCountInfo[4] = SafeDiv(lineCountInfo[2], lineCountInfo[2]+lineCountInfo[1])
  detailCountInfo.append([filePath, lineCountInfo]) 

SORT_ORDER = (lambda x:x[0], False)
def SetSortArg(sortArg=None):
 global SORT_ORDER
 if not sortArg:
  return
 if any(s in sortArg for s in ('file', '0')): #条件宽松些
 #if sortArg in ('rfile', 'file', 'r0', '0'):
  keyFunc = lambda x:x[1][0]
 elif any(s in sortArg for s in ('code', '1')):
  keyFunc = lambda x:x[1][1]
 elif any(s in sortArg for s in ('cmmt', '2')):
  keyFunc = lambda x:x[1][2]
 elif any(s in sortArg for s in ('blan', '3')):
  keyFunc = lambda x:x[1][3]
 elif any(s in sortArg for s in ('ctpr', '4')):
  keyFunc = lambda x:x[1][4]
 elif any(s in sortArg for s in ('name', '5')):
  keyFunc = lambda x:x[0]
 else: #因argparse内已限制排序参数范围，此处也可用assert
  print >>sys.stderr, 'Unsupported sort order(%s)!' %sortArg
  return

 isReverse = sortArg[0]=='r' #False:升序(ascending); True:降序(decending)
 SORT_ORDER = (keyFunc, isReverse)

def ReportCounterInfo(isRawReport=True, stream=sys.stdout):
  #代码注释率 = 注释行 / (注释行+有效代码行)
 print >>stream, 'FileLines CodeLines CommentLines BlankLines CommentPercent %s'\
   %(not isRawReport and 'FileName' or '')

 if isRawReport:
  print >>stream, '%-11d%-11d%-14d%-12d%-16.2f<Total:%d Code Files>' %(rawCountInfo[0],\
    rawCountInfo[1], rawCountInfo[2], rawCountInfo[3], \
    SafeDiv(rawCountInfo[2], rawCountInfo[2]+rawCountInfo[1]), rawCountInfo[4])
  return

 total = [0, 0, 0, 0]
 #对detailCountInfo排序。缺省按第一列元素(文件名)升序排序，以提高输出可读性。
 detailCountInfo.sort(key=SORT_ORDER[0], reverse=SORT_ORDER[1])
 for item in detailCountInfo:
  print >>stream, '%-11d%-11d%-14d%-12d%-16.2f%s' %(item[1][0], item[1][1], item[1][2], \
    item[1][3], item[1][4], item[0])
  total[0] += item[1][0]; total[1] += item[1][1]
  total[2] += item[1][2]; total[3] += item[1][3]
 print >>stream, '-' * 90 #输出90个负号(minus)或连字号(hyphen)
 print >>stream, '%-11d%-11d%-14d%-12d%-16.2f<Total:%d Code Files>' \
   %(total[0], total[1], total[2], total[3], \
   SafeDiv(total[2], total[2]+total[1]), len(detailCountInfo)) 

def ParseTargetList(targetList):
 fileList, dirList = [], []
 if targetList == []:
  targetList.append(os.getcwd())
 for item in targetList:
  if os.path.isfile(item):
   fileList.append(os.path.abspath(item))
  elif os.path.isdir(item):
   dirList.append(os.path.abspath(item))
  else:
   print >>sys.stderr, "'%s' is neither a file nor a directory!" %item
 return [fileList, dirList] 

def CountDir(dirList, isKeep=False, isRawReport=True, isShortName=False):
 for dir in dirList:
  if isKeep:
   for file in os.listdir(dir):
    CountFileLines(os.path.join(dir, file), isRawReport, isShortName)
  else:
   for root, dirs, files in os.walk(dir):
    for file in files:
     CountFileLines(os.path.join(root, file), isRawReport, isShortName)

def CountFile(fileList, isRawReport=True, isShortName=False):
 for file in fileList:
  CountFileLines(file, isRawReport, isShortName)

def LineCounter(isKeep=False, isRawReport=True, isShortName=False, targetList=[]):
 fileList, dirList = ParseTargetList(targetList)
 if fileList != []:
  CountFile(fileList, isRawReport, isShortName)
 if dirList != []:
  CountDir(dirList, isKeep, isRawReport, isShortName) 

import argparse
def ParseCmdArgs(argv=sys.argv):
 parser = argparse.ArgumentParser(usage='%(prog)s [options] target',
      description='Count lines in code files.')
 parser.add_argument('target', nargs='*',
   help='space-separated list of directories AND/OR files')
 parser.add_argument('-k', '--keep', action='store_true',
   help='do not walk down subdirectories')
 parser.add_argument('-d', '--detail', action='store_true',
   help='report counting result in detail')
 parser.add_argument('-b', '--basename', action='store_true',
   help='do not show file\'s full path')
## sortWords = ['0', '1', '2', '3', '4', '5', 'file', 'code', 'cmmt', 'blan', 'ctpr', 'name']
## parser.add_argument('-s', '--sort',
##  choices=[x+y for x in ['','r'] for y in sortWords],
##  help='sort order: {0,1,2,3,4,5} or {file,code,cmmt,blan,ctpr,name},' \
##    "prefix 'r' means sorting in reverse order")
 parser.add_argument('-s', '--sort',
   help='sort order: {0,1,2,3,4,5} or {file,code,cmmt,blan,ctpr,name}, ' \
    "prefix 'r' means sorting in reverse order")
 parser.add_argument('-o', '--out',
   help='save counting result in OUT')
 parser.add_argument('-c', '--cache', action='store_true',
   help='use cache to count faster(unreliable when files are modified)')
 parser.add_argument('-v', '--version', action='version',
   version='%(prog)s 3.0 by xywang')

 args = parser.parse_args()
 return (args.keep, args.detail, args.basename, args.sort, args.out, args.cache, args.target) 

CACHE_FILE = 'Counter.dump'
CACHE_DUMPER, CACHE_GEN = None, None

from json import dump, JSONDecoder
def CounterDump(data):
 global CACHE_DUMPER
 if CACHE_DUMPER == None:
  CACHE_DUMPER = open(CACHE_FILE, 'w')
 dump(data, CACHE_DUMPER)

def ParseJson(jsonData):
 endPos = 0
 while True:
  jsonData = jsonData[endPos:].lstrip()
  try:
   pyObj, endPos = JSONDecoder().raw_decode(jsonData)
   yield pyObj
  except ValueError:
   break

def CounterLoad():
 global CACHE_GEN
 if CACHE_GEN == None:
  CACHE_GEN = ParseJson(open(CACHE_FILE, 'r').read())

 try:
  return next(CACHE_GEN)
 except StopIteration, e:
  return []

def shouldUseCache(keep, detail, basename, cache, target):
 if not cache: #未指定启用缓存
  return False

 try:
  (_keep, _detail, _basename, _target) = CounterLoad()
 except (IOError, EOFError, ValueError): #缓存文件不存在或内容为空或不合法
  return False

 if keep == _keep and detail == _detail and basename == _basename \
  and sorted(target) == sorted(_target):
  return True
 else:
  return False 

def main():
 global gIsStdout, rawCountInfo, detailCountInfo
 (keep, detail, basename, sort, out, cache, target) = ParseCmdArgs()
 stream = sys.stdout if not out else open(out, 'w')
 SetSortArg(sort); LoadCExtLib()
 cacheUsed = shouldUseCache(keep, detail, basename, cache, target)
 if cacheUsed:
  try:
   (rawCountInfo, detailCountInfo) = CounterLoad()
  except (EOFError, ValueError), e: #不太可能出现
   print >>sys.stderr, 'Unexpected Cache Corruption(%s), Try Counting Directly.'%e
   LineCounter(keep, not detail, basename, target)
 else:
  LineCounter(keep, not detail, basename, target)

 ReportCounterInfo(not detail, stream)
 CounterDump((keep, detail, basename, target))
 CounterDump((rawCountInfo, detailCountInfo)) 

if __name__ == '__main__':
 from time import clock
 startTime = clock()
 main()
 endTime = clock()
 print >>sys.stderr, 'Time Elasped: %.2f sec.' %(endTime-startTime) 

#include <stdio.h>
#include <string.h>
#define TRUE 1
#define FALSE 0

unsigned int CalcLinesCh(char *line, unsigned char isBlockComment[2]) {
 unsigned int lineType = 0;
 unsigned int lineLen = strlen(line);
 if(!lineLen)
  return lineType;

 char *expandLine = calloc(lineLen + 1/*\n*/, 1);
 if(NULL == expandLine)
  return lineType;
 memmove(expandLine, line, lineLen);
 expandLine[lineLen] = '\n'; //添加一个字符防止iChar+1时越界

 unsigned int iChar = 0;
 unsigned char isLineComment = FALSE;
 while(iChar < lineLen) {
  if(expandLine[iChar] == ' ' || expandLine[iChar] == '\t') { //空白字符
   iChar += 1; continue;
  }
  else if(expandLine[iChar] == '/' && expandLine[iChar+1] == '/') { //行注释
   isLineComment = TRUE;
   lineType |= 2; iChar += 1; //跳过'/'
  }
  else if(expandLine[iChar] == '/' && expandLine[iChar+1] == '*') { //块注释开始符
   isBlockComment[0] = TRUE;
   lineType |= 2; iChar += 1;
  }
  else if(expandLine[iChar] == '*' && expandLine[iChar+1] == '/') { //块注释结束符
   isBlockComment[0] = FALSE;
   lineType |= 2; iChar += 1;
  }
  else {
   if(isLineComment || isBlockComment[0])
    lineType |= 2;
   else
    lineType |= 1;
  }
  iChar += 1;
 }

 free(expandLine);
 return lineType; //Bitmap：0空行，1代码，2注释，3代码和注释
}

unsigned int CalcLinesPy(char *line, unsigned char isBlockComment[2]) {
 //isBlockComment[single quotes, double quotes]
 unsigned int lineType = 0;
 unsigned int lineLen = strlen(line);
 if(!lineLen)
  return lineType;

 char *expandLine = calloc(lineLen + 2/*\n\n*/, 1);
 if(NULL == expandLine)
  return lineType;
 memmove(expandLine, line, lineLen);
 //添加两个字符防止iChar+2时越界
 expandLine[lineLen] = '\n'; expandLine[lineLen+1] = '\n';

 unsigned int iChar = 0;
 unsigned char isLineComment = FALSE;
 while(iChar < lineLen) {
  if(expandLine[iChar] == ' ' || expandLine[iChar] == '\t') { //空白字符
   iChar += 1; continue;
  }
  else if(expandLine[iChar] == '#') { //行注释
   isLineComment = TRUE;
   lineType |= 2;
  }
  else if(expandLine[iChar] == '\'' && expandLine[iChar+1] == '\''
    && expandLine[iChar+2] == '\'') { //单引号块注释
   if(isBlockComment[0] || isBlockComment[1])
    isBlockComment[0] = FALSE;
   else
    isBlockComment[0] = TRUE;
   lineType |= 2; iChar += 2;
  }
  else if(expandLine[iChar] == '"' && expandLine[iChar+1] == '"'
    && expandLine[iChar+2] == '"') { //双引号块注释
   if(isBlockComment[0] || isBlockComment[1])
    isBlockComment[1] = FALSE;
   else
    isBlockComment[1] = TRUE;
   lineType |= 2; iChar += 2;
  }
  else {
   if(isLineComment || isBlockComment[0] || isBlockComment[1])
    lineType |= 2;
   else
    lineType |= 1;
  }
  iChar += 1;
 }

 free(expandLine);
 return lineType; //Bitmap：0空行，1代码，2注释，3代码和注释
} 

 #define TRUE 1
#define FALSE 0
unsigned int CalcLinesCh(char *line, unsigned char isBlockComment[2]) {
 unsigned int lineType = 0;

 unsigned int iChar = 0;
 unsigned char isLineComment = FALSE;
 while(line[iChar] != '\0') {
  if(line[iChar] == ' ' || line[iChar] == '\t') { //空白字符
   iChar += 1; continue;
  }
  else if(line[iChar] == '/' && line[iChar+1] == '/') { //行注释
   isLineComment = TRUE;
   lineType |= 2; iChar += 1; //跳过'/'
  }
  else if(line[iChar] == '/' && line[iChar+1] == '*') { //块注释开始符
   isBlockComment[0] = TRUE;
   lineType |= 2; iChar += 1;
  }
  else if(line[iChar] == '*' && line[iChar+1] == '/') { //块注释结束符
   isBlockComment[0] = FALSE;
   lineType |= 2; iChar += 1;
  }
  else {
   if(isLineComment || isBlockComment[0])
    lineType |= 2;
   else
    lineType |= 1;
  }
  iChar += 1;
 }

 return lineType; //Bitmap：0空行，1代码，2注释，3代码和注释
}

unsigned int CalcLinesPy(char *line, unsigned char isBlockComment[2]) {
 //isBlockComment[single quotes, double quotes]
 unsigned int lineType = 0;

 unsigned int iChar = 0;
 unsigned char isLineComment = FALSE;
 while(line[iChar] != '\0') {
  if(line[iChar] == ' ' || line[iChar] == '\t') { //空白字符
   iChar += 1; continue;
  }
  else if(line[iChar] == '#') { //行注释
   isLineComment = TRUE;
   lineType |= 2;
  }
  else if(line[iChar] == '\'' && line[iChar+1] == '\''
    && line[iChar+2] == '\'') { //单引号块注释
   if(isBlockComment[0] || isBlockComment[1])
    isBlockComment[0] = FALSE;
   else
    isBlockComment[0] = TRUE;
   lineType |= 2; iChar += 2;
  }
  else if(line[iChar] == '"' && line[iChar+1] == '"'
    && line[iChar+2] == '"') { //双引号块注释
   if(isBlockComment[0] || isBlockComment[1])
    isBlockComment[1] = FALSE;
   else
    isBlockComment[1] = TRUE;
   lineType |= 2; iChar += 2;
  }
  else {
   if(isLineComment || isBlockComment[0] || isBlockComment[1])
    lineType |= 2;
   else
    lineType |= 1;
  }
  iChar += 1;
 }

 return lineType; //Bitmap：0空行，1代码，2注释，3代码和注释
} 

E:\PyTest>pypy
Python 2.7.10 (b0a649e90b66, Apr 28 2016, 13:11:00)
[PyPy 5.1.1 with MSC v.1500 32 bit] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>>> import cffi
>>>> cffi.__version__
'1.6.0' 

 D:\pytest>CPLineCounter -d lctest -s code
FileLines CodeLines CommentLines BlankLines CommentPercent FileName
6   3   4    0   0.57   D:\pytest\lctest\hard.c
27   7   15   5   0.68   D:\pytest\lctest\file27_code7_cmmt15_blank5.py
33   19   15   4   0.44   D:\pytest\lctest\line.c
44   34   3    7   0.08   D:\pytest\lctest\test.c
44   34   3    7   0.08   D:\pytest\lctest\subdir\test.c
243  162  26   60   0.14   D:\pytest\lctest\subdir\CLineCounter.py
------------------------------------------------------------------------------------------
397  259  66   83   0.20   <Total:6 Code Files>
Time Elasped: 0.04 sec. 

 E:\PyTest>kernprof -l -v CPLineCounter.py source -d > out.txt
140872  93736  32106   16938  0.26   <Total:82 Code Files>
Wrote profile results to CPLineCounter.py.lprof
Timer unit: 2.79365e-07 s

Total time: 5.81981 s
File: CPLineCounter.py
Function: CountFileLines at line 143

Line #  Hits   Time Per Hit % Time Line Contents
==============================================================
 143           @profile
 144           def CountFileLines(filePath, isRawReport=True, isShortName=False):
... ... ... ... ... ... ... ...
 162  82  7083200 86380.5  34.0  with open(filePath, 'r') as file:
 163 140954  1851877  13.1  8.9   for line in file:
 164 140872  6437774  45.7  30.9    lineType = CalcLines(fileType, line.strip(), isBlockComment)
 165 140872  1761864  12.5  8.5    lineCountInfo[0] += 1
 166 140872  1662583  11.8  8.0    if lineType == 0: lineCountInfo[3] += 1
 167 123934  1499176  12.1  7.2    elif lineType == 1: lineCountInfo[1] += 1
 168  32106  406931  12.7  2.0    elif lineType == 2: lineCountInfo[2] += 1
 169  1908  27634  14.5  0.1    elif lineType == 3: lineCountInfo[1] += 1; lineCountInfo[2] += 1
... ... ... ... ... ... ... ...