/usr/share/pyshared/PyMetrics/halstead.py is in pymetrics 0.8.1-6.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 | """ Compute HalsteadMetric Metrics.
HalsteadMetric metrics, created by Maurice H. HalsteadMetric in 1977, consist
of a number of measures, including:
Program length (N): N = N1 + N2
Program vocabulary (n): n = n1 + n2
Volume (V): V = N * LOG2(n)
Difficulty (D): D = (n1/2) * (N2/n2)
Effort (E): E = D * V
Average Volume (avgV) avgV = sum(V)/m
Average Effort (avgE) avgE = sum(E)/m
where:
n1 = number of distinct operands
n2 = number of distinct operators
N1 = total number of operands
N2 = total number of operators
m = number of modules
What constitues an operand or operator is often open to
interpretation. In this implementation for the Python language:
operators are of type OP, INDENT, DEDENT, or NEWLINE since these
serve the same purpose as braces and semicolon in C/C++, etc.
operands are not operators or whitespace or comments
(this means operands include keywords)
$Id: halstead.py,v 1.3 2005/09/17 04:28:12 rcharney Exp $
"""
__version__ = "$Revision: 1.3 $"[11:-2]
__author__ = 'Reg. Charney <pymetrics@charneyday.com>'
import math
import time
from metricbase import MetricBase
from globals import *
class HalsteadMetric( MetricBase ):
""" Compute various HalsteadMetric metrics. """
totalV = 0
totalE = 0
numModules = 0
def __init__( self, context, runMetrics, metrics, pa, *args, **kwds ):
""" Initialization for the HalsteadMetric metrics."""
self.inFile = context['inFile']
self.context = context
self.runMetrics = runMetrics
self.metrics = metrics
self.pa = pa
self.inFile = context['inFile']
self.numOperators = 0
self.numOperands = 0
self.uniqueOperators = {}
self.uniqueOperands = {}
HalsteadMetric.numModules += 1
# initialize category accummulators as dictionaries
self.hsDict = {}
for t in ['token','stmt','block','function','class','module','run']:
self.uniqueOperators[t] = {}
self.uniqueOperands[t] = {}
#for v in ['N','N1','N2','n','n1','n2','V','D','E','avgV','avgE']:
# self.hsDict[(t,v)] = 0
def processToken( self, currentFcn, currentClass, tok, *args, **kwds ):
""" Collect token data for Halstead metrics."""
if tok.type in [WS, EMPTY, ENDMARKER, NEWLINE, EMPTY, COMMENT]:
pass
elif tok.type in [OP, INDENT, DEDENT]:
self.numOperators += 1
self.uniqueOperators['token'][tok.text] = self.uniqueOperators['token'].get(tok.text, 0) + 1
else:
self.numOperands += 1
sDict = self.context.__repr__()
k = (sDict,tok.text)
self.uniqueOperands['token'][k] = self.uniqueOperands['token'].get(tok.text, 0) + 1
def processStmt( self, currentFcn, currentClass, stmt, *args, **kwds ):
""" Collect statement data for Halstead metrics."""
result = None
# the two lines following this comment would compute the Halstead
# metrics for each statement in the run, However, it is
# normally overkill, so these lines are commented out.
#lineNum = stmt[0].row
#result = self.computeCategory( 'stmt', lineNum, stmt )
return result
def processBlock( self, currentFcn, currentClass, block, *args, **kwds ):
""" Collect block data for Halstead metrics."""
result = None
# the two lines following this comment would compute the Halstead
# metrics for each statement in the run, However, it is
# normally overkill, so the two lines are commented out.
#blockNum = self.context['blockNum']
#result = self.computeCategory( 'block', blockNum, block )
return result
def processFunction( self, currentFcn, currentClass, fcn, *args, **kwds ):
""" Collect function data for Halstead metrics."""
result = self.computeCategory( 'function', currentFcn, fcn )
return result
def processClass( self, currentFcn, currentClass, cls, *args, **kwds ):
""" Collect class data for Halstead metrics."""
result = self.computeCategory( 'class', currentClass, cls )
return result
def processModule( self, moduleName, mod, *args, **kwds ):
""" Collect module data for Halstead metrics."""
result = self.computeCategory( 'module', moduleName, mod )
return result
def processRun( self, run, *args, **kwds ):
""" Collect run data for Halstead metrics."""
datestamp = time.strftime("%Y-%m-%d.%H:%m%Z",time.localtime())
result = self.computeCategory( 'run', datestamp, run )
return result
def __LOGb( self, x, b ):
""" convert to LOGb(x) from natural logs."""
try:
result = math.log( x ) / math.log ( b )
except OverflowError:
result = 1.0
return result
def computeIncr( self, cat, tok, uniqueOperators, uniqueOperands ):
""" Compute increment for token depending on which category it falls into."""
operatorIncr = operandIncr = 0
if tok.type in [WS, EMPTY, ENDMARKER, NEWLINE, EMPTY, COMMENT]:
return (operatorIncr,operandIncr)
if tok.type in [OP, INDENT, DEDENT]:
operatorIncr = 1
uniqueOperators[tok.text] = uniqueOperators.get(tok.text, 0) + 1
else:
operandIncr = 1
uniqueOperands[tok.text] = uniqueOperands.get(tok.text,0) + 1
return (operatorIncr,operandIncr)
def computeCategory( self, cat, mod, lst ):
""" Collection data for cat of code."""
modID= id( mod )
numOperators = numOperands = 0
for tok in lst:
result = self.computeIncr( cat, tok, self.uniqueOperators[cat], self.uniqueOperands[cat] )
numOperators += result[0]
numOperands += result[1]
result = self.compute( cat, modID, numOperators, numOperands, self.uniqueOperators[cat], self.uniqueOperands[cat] )
return result
def compute( self, cat, modID, numOperators, numOperands, uniqueOperators, uniqueOperands, *args, **kwds ):
""" Do actual calculations here."""
n1 = len( uniqueOperands )
n2 = len( uniqueOperators )
N1 = numOperands
N2 = numOperators
N = N1 + N2
n = n1 + n2
V = float(N) * self.__LOGb( n, 2 )
try:
D = (float(n1)/2.0) * (float(N2)/float(n2))
except ZeroDivisionError:
D = 0.0
E = D * V
HalsteadMetric.totalV += V
HalsteadMetric.totalE += E
avgV = HalsteadMetric.totalV / HalsteadMetric.numModules
avgE = HalsteadMetric.totalE / HalsteadMetric.numModules
self.hsDict[(cat,modID,'n1')] = n1
self.hsDict[(cat,modID,'n2')] = n2
self.hsDict[(cat,modID,'N1')] = N1
self.hsDict[(cat,modID,'N2')] = N2
self.hsDict[(cat,modID,'N')] = N
self.hsDict[(cat,modID,'n')] = n
self.hsDict[(cat,modID,'V')] = V
self.hsDict[(cat,modID,'D')] = D
self.hsDict[(cat,modID,'E')] = E
self.hsDict[(cat,modID,'numModules')] = HalsteadMetric.numModules
self.hsDict[(cat,modID,'avgV')] = avgV
self.hsDict[(cat,modID,'avgE')] = avgE
return self.hsDict
def display( self, cat=None ):
""" Display the computed Halstead Metrics."""
if self.pa.quietSw:
return self.hsDict
hdr = "\nHalstead Metrics for %s" % self.inFile
print hdr
print "-"*len(hdr) + '\n'
if len( self.hsDict ) == 0:
print "%-8s %-30s " % ('**N/A**','All Halstead metrics are zero')
return self.hsDict
keyList = self.hsDict.keys()
keyList.sort()
if 0:
for k,i,v in keyList:
if cat:
if k!=cat:
continue
print "%14.2f %s %s %s" % (self.hsDict[(k,i,v)],k,i,v)
print
hdr1 = "Category Identifier D E N N1 N2 V avgE avgV n n1 n2"
hdr2 = "-------- ---------------------------------- -------- -------- ----- ---- ---- -------- -------- -------- ----- ---- ----"
# 12345678 123456789012345678901234567890 12345678 12345678 12345 1234 1234 12345678 12345678 12345678 12345 1234 1234
fmt1 = "%-8s %-33s "
fmt2 = "%8.2e %8.2e %5d %4d %4d %8.2e %8.2e %8.2e %5d %4d %4d"
# this loop uses the Main Line Standards break logic. It does this to convert the
# normal vertical output to a horizontal format. The control variables are the
# category name and the identifier value.
oldK = oldI = None
vDict = {}
vList = []
hdrSw = True # output header for first time thru
for k,i,v in keyList:
# only print data for the category we want
if cat:
if k != cat:
continue
if v == "numModules": # ignore this value for now
continue
if (oldK,oldI) != (k,i): # change in category/id
if oldK and oldI: # this is not first time thru
#t = tuple([self.hsDict[(k,i,v)] for v in vList])
t = tuple([vDict[v] for v in vList])
print fmt1 % (k,i),
print fmt2 % t
# initialize for next set of category/id
vDict = {}
vDict[v] = self.hsDict[(k,i,v)]
vList = []
vList.append( v )
oldK = k
oldI = i
if hdrSw:
print hdr1
print hdr2
hdrSw = False
else: # we are still in the same category/id
vDict[v] = self.hsDict[(k,i,v)]
vList.append( v )
print
return self.hsDict
|