File: cccc_tok.cc

package info (click to toggle)
cccc 3.pre81-2
  • links: PTS
  • area: main
  • in suites: sarge
  • size: 3,820 kB
  • ctags: 4,972
  • sloc: ansic: 33,244; cpp: 10,691; java: 618; makefile: 165; sh: 11
file content (165 lines) | stat: -rw-r--r-- 4,685 bytes parent folder | download | duplicates (2)
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
/*
 * cccc_tok.C
 * implementation of a token class for the cccc project
 *
 */

#include "cccc.h"
#include "cccc_tok.h"

/* static variables */
int ANTLRToken::RunningNesting=0;
int ANTLRToken::bCodeLine=0;
int ANTLRToken::numAllocated=0;
int toks_alloc1=0, toks_alloc2=0, toks_alloc3=0, toks_freed=0;

ANTLRToken currentLexerToken;

/* 
** Token objects are used to count the occurences of states which 
** our analyser is interested in within the text.  Any metric which
** can be reduced to lexical counting on the text can be recorded
** this way.
**
** This implementation counts the following features:
**   tokens
**   comment lines
**   lines containing at least one token of code
**  
** It also makes a lexical count for the following tokens, each of which
** is expected to increase McCabe's cyclomatic complexity (Vg) for the 
** section of code by one unit:
**  IF FOR WHILE SWITCH BREAK RETURN ? && ||
**
** Note that && and || create additional paths through the code due to C/C++ 
** short circuit evaluation of logical expressions.
**
** Also note the way SWITCH constructs are counted: the desired increment
** in Vg is equal to the number of cases provided for, including the 
** default case, whether or not an action is defined for it.  This is acheived
** by counting the SWITCH at the head of the construct as a surrogate for 
** the default case, and counting BREAKs as surrogates for the individual
** cases.  This approach yields the correct results provided that the
** coding style in use ensures the use of BREAK after all non-default
** cases, and forbids 'drop through' from one case to another other than
** in the case where two or more values of the switch variable require
** identical actions, and no executable code is defined between the 
** case gates (as in the switch statement in ANTLRToken::CountToken() below).
*/

/* default constructor */
ANTLRToken::ANTLRToken() : ANTLRCommonToken() { 
  toks_alloc1++;
  CurrentNesting=-99;
}

/* 
** constructor used by makeToken below 
*/
ANTLRToken::ANTLRToken(ANTLRTokenType t, ANTLRChar *s) : 
  ANTLRCommonToken(t,s) {
  setType(t);
  setText(s);
  CountToken();

  toks_alloc2++;
}

/* copy constructor */
ANTLRToken::ANTLRToken(ANTLRToken& copyTok) {
  setType(copyTok.getType());
  setText(copyTok.getText());
  setLine(copyTok.getLine());
  CurrentNesting=copyTok.CurrentNesting;
  toks_alloc3++;
}

/* 
** the virtual pseudo-constructor 
** This is required because the PCCTS support code does not know the
** exact nature of the token which will be created by the user's code, 
** and indeed does not forbid the user creating more than one kind of
** token, so long as ANTLRToken is defined and all token classes are
** subclassed from ANTLRAbstractToken
*/
ANTLRAbstractToken *ANTLRToken::makeToken(
					  ANTLRTokenType tt, ANTLRChar *txt, int line
					  ) {
  
  ANTLRToken *new_t = new ANTLRToken(tt,txt);
  if(new_t==0) { 
    cerr << "Memory overflow in "
      "ANTLRToken::makeToken(" << static_cast<int>(tt) << "," 
	 << txt << "," << line << ")" << endl;
    exit(2);
  }
  new_t->setLine(line);

  DbgMsg(
	 LEXER,cerr, 
	 "makeToken(tt=>" << static_cast<int>(tt) << 
	 ", txt=>" << txt << 
	 ",line=>" << line << 
	 ")" << endl
	 );

  return new_t;
}

/* the destructor */
ANTLRToken::~ANTLRToken() {
  toks_freed++;
  DbgMsg(MEMORY,cerr,"freeing token " << getText()
	 << " on line " << getLine()
	 << " c1:" << toks_alloc1 << " c2:" << toks_alloc2 
	 << " c3:" << toks_alloc3 << " freed:" << toks_freed << endl);
}

/* the assignment operator */
ANTLRToken& ANTLRToken::operator=(ANTLRToken& copyTok) {
  setType(copyTok.getType());
  setText(copyTok.getText());
  setLine(copyTok.getLine());
  CurrentNesting=copyTok.CurrentNesting;
  return *this;
}

/*
** ANTLRToken::CountToken performs counting of features which are traced
** back to individual tokens created up by the lexer, i.e. the token count 
** and McCabes VG.  Code lines and comment lines are both identified during
** the processing of text which the lexer will (usually) skip, so the code
** to increment these counts is in the relevant lexer rules in the file 
** cccc.g
*/
void ANTLRToken::CountToken()
{
  // we have seen a non-skippable pattern => this line counts toward LOC  
  bCodeLine=1;
  CurrentNesting=RunningNesting;
  DbgMsg(COUNTER,cerr,*this);
}

char *ANTLRToken::getTokenTypeName() { return ""; }

/*
** structured output method for token objects
*/
ostream& operator << (ostream& out, ANTLRToken& t) {
  int i;

  out << "TOK: " << t.getTokenTypeName() 
      << " " << t.getText() 
      << " " << t.getLine()
      << " " << t.getNestingLevel();

  out << endl;
  return out;
}