-
Notifications
You must be signed in to change notification settings - Fork 2k
Expand file tree
/
Copy pathCodeDuplication.qll
More file actions
281 lines (252 loc) · 10.9 KB
/
CodeDuplication.qll
File metadata and controls
281 lines (252 loc) · 10.9 KB
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
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
/** Provides classes for detecting duplicate or similar code. */
import python
/** Gets the relative path of `file`, with backslashes replaced by forward slashes. */
private
string relativePath(File file) {
result = file.getRelativePath().replaceAll("\\", "/")
}
/**
* Holds if the `index`-th token of block `copy` is in file `file`, spanning
* column `sc` of line `sl` to column `ec` of line `el`.
*
* For more information, see [LGTM locations](https://lgtm.com/help/ql/locations).
*/
pragma[noinline, nomagic]
private predicate tokenLocation(File file, int sl, int sc, int ec, int el, Copy copy, int index) {
file = copy.sourceFile() and
tokens(copy, index, sl, sc, ec, el)
}
/** A token block used for detection of duplicate and similar code. */
class Copy extends @duplication_or_similarity
{
private
int lastToken() {
result = max(int i | tokens(this, i, _, _, _, _) | i)
}
/** Gets the index of the token in this block starting at the location `loc`, if any. */
int tokenStartingAt(Location loc) {
tokenLocation(loc.getFile(), loc.getStartLine(), loc.getStartColumn(),
_, _, this, result)
}
/** Gets the index of the token in this block ending at the location `loc`, if any. */
int tokenEndingAt(Location loc) {
tokenLocation(loc.getFile(), _, _,
loc.getEndLine(), loc.getEndColumn(), this, result)
}
/** Gets the line on which the first token in this block starts. */
int sourceStartLine() {
tokens(this, 0, result, _, _, _)
}
/** Gets the column on which the first token in this block starts. */
int sourceStartColumn() {
tokens(this, 0, _, result, _, _)
}
/** Gets the line on which the last token in this block ends. */
int sourceEndLine() {
tokens(this, this.lastToken(), _, _, result, _)
}
/** Gets the column on which the last token in this block ends. */
int sourceEndColumn() {
tokens(this, this.lastToken(), _, _, _, result)
}
/** Gets the number of lines containing at least (part of) one token in this block. */
int sourceLines() {
result = this.sourceEndLine() + 1 - this.sourceStartLine()
}
/** Gets an opaque identifier for the equivalence class of this block. */
int getEquivalenceClass() {
duplicateCode(this, _, result) or similarCode(this, _, result)
}
/** Gets the source file in which this block appears. */
File sourceFile() {
exists(string name |
duplicateCode(this, name, _) or similarCode(this, name, _) |
name.replaceAll("\\", "/") = relativePath(result))
}
/**
* Holds if this element is at the specified location.
* The location spans column `startcolumn` of line `startline` to
* column `endcolumn` of line `endline` in file `filepath`.
* For more information, see
* [LGTM locations](https://lgtm.com/help/ql/locations).
*/
predicate hasLocationInfo(string filepath, int startline, int startcolumn, int endline, int endcolumn) {
sourceFile().getName() = filepath and
startline = sourceStartLine() and
startcolumn = sourceStartColumn() and
endline = sourceEndLine() and
endcolumn = sourceEndColumn()
}
/** Gets a textual representation of this element. */
string toString() { none() }
/**
* Gets a block that extends this one, that is, its first token is also
* covered by this block, but they are not the same block.
*/
Copy extendingBlock() {
exists(File file, int sl, int sc, int ec, int el |
tokenLocation(file, sl, sc, ec, el, this, _) and
tokenLocation(file, sl, sc, ec, el, result, 0)) and
this != result
}
}
/**
* Holds if there is a sequence of `SimilarBlock`s `start1, ..., end1` and another sequence
* `start2, ..., end2` such that each block extends the previous one and corresponding blocks
* have the same equivalence class, with `start` being the equivalence class of `start1` and
* `start2`, and `end` the equivalence class of `end1` and `end2`.
*/
predicate similar_extension(SimilarBlock start1, SimilarBlock start2, SimilarBlock ext1, SimilarBlock ext2, int start, int ext) {
start1.getEquivalenceClass() = start and
start2.getEquivalenceClass() = start and
ext1.getEquivalenceClass() = ext and
ext2.getEquivalenceClass() = ext and
start1 != start2 and
(ext1 = start1 and ext2 = start2 or
similar_extension(start1.extendingBlock(), start2.extendingBlock(), ext1, ext2, _, ext)
)
}
/**
* Holds if there is a sequence of `DuplicateBlock`s `start1, ..., end1` and another sequence
* `start2, ..., end2` such that each block extends the previous one and corresponding blocks
* have the same equivalence class, with `start` being the equivalence class of `start1` and
* `start2`, and `end` the equivalence class of `end1` and `end2`.
*/
predicate duplicate_extension(DuplicateBlock start1, DuplicateBlock start2, DuplicateBlock ext1, DuplicateBlock ext2, int start, int ext) {
start1.getEquivalenceClass() = start and
start2.getEquivalenceClass() = start and
ext1.getEquivalenceClass() = ext and
ext2.getEquivalenceClass() = ext and
start1 != start2 and
(ext1 = start1 and ext2 = start2 or
duplicate_extension(start1.extendingBlock(), start2.extendingBlock(), ext1, ext2, _, ext)
)
}
/** A block of duplicated code. */
class DuplicateBlock extends Copy, @duplication
{
override string toString() {
result = "Duplicate code: " + sourceLines() + " duplicated lines."
}
}
/** A block of similar code. */
class SimilarBlock extends Copy, @similarity
{
override string toString() {
result = "Similar code: " + sourceLines() + " almost duplicated lines."
}
}
/**
* Holds if `stmt1` and `stmt2` are duplicate statements in function or toplevel `sc1` and `sc2`,
* respectively, where `scope1` and `scope2` are not the same.
*/
predicate duplicateStatement(Scope scope1, Scope scope2, Stmt stmt1, Stmt stmt2) {
exists(int equivstart, int equivend, int first, int last |
scope1.contains(stmt1) and
scope2.contains(stmt2) and
duplicateCoversStatement(equivstart, equivend, first, last, stmt1) and
duplicateCoversStatement(equivstart, equivend, first, last, stmt2) and
stmt1 != stmt2 and scope1 != scope2
)
}
/**
* Holds if statement `stmt` is covered by a sequence of `DuplicateBlock`s, where `first`
* is the index of the token in the first block that starts at the beginning of `stmt`,
* while `last` is the index of the token in the last block that ends at the end of `stmt`,
* and `equivstart` and `equivend` are the equivalence classes of the first and the last
* block, respectively.
*/
private
predicate duplicateCoversStatement(int equivstart, int equivend, int first, int last, Stmt stmt) {
exists(DuplicateBlock b1, DuplicateBlock b2, Location startloc, Location endloc |
stmt.getLocation() = startloc and
stmt.getLastStatement().getLocation() = endloc and
first = b1.tokenStartingAt(startloc) and
last = b2.tokenEndingAt(endloc) and
b1.getEquivalenceClass() = equivstart and
b2.getEquivalenceClass() = equivend and
duplicate_extension(b1, _, b2, _, equivstart, equivend)
)
}
/**
* Holds if `sc1` is a function or toplevel with `total` lines, and `scope2` is a function or
* toplevel that has `duplicate` lines in common with `scope1`.
*/
predicate duplicateStatements(Scope scope1, Scope scope2, int duplicate, int total) {
duplicate = strictcount(Stmt stmt | duplicateStatement(scope1, scope2, stmt, _)) and
total = strictcount(Stmt stmt | scope1.contains(stmt))
}
/**
* Find pairs of scopes that are identical or almost identical
*/
predicate duplicateScopes(Scope s, Scope other, float percent, string message) {
exists(int total, int duplicate |
duplicateStatements(s, other, duplicate, total) |
percent = 100.0 * duplicate / total and percent >= 80.0 and
if duplicate = total then
message = "All " + total + " statements in " + s.getName() + " are identical in $@."
else
message = duplicate + " out of " + total + " statements in " + s.getName() + " are duplicated in $@."
)
}
/**
* Holds if `stmt1` and `stmt2` are similar statements in function or toplevel `scope1` and `scope2`,
* respectively, where `scope1` and `scope2` are not the same.
*/
private predicate similarStatement(Scope scope1, Scope scope2, Stmt stmt1, Stmt stmt2) {
exists(int start, int end, int first, int last |
scope1.contains(stmt1) and
scope2.contains(stmt2) and
similarCoversStatement(start, end, first, last, stmt1) and
similarCoversStatement(start, end, first, last, stmt2) and
stmt1 != stmt2 and scope1 != scope2
)
}
/**
* Holds if statement `stmt` is covered by a sequence of `SimilarBlock`s, where `first`
* is the index of the token in the first block that starts at the beginning of `stmt`,
* while `last` is the index of the token in the last block that ends at the end of `stmt`,
* and `equivstart` and `equivend` are the equivalence classes of the first and the last
* block, respectively.
*/
private predicate similarCoversStatement(int equivstart, int equivend, int first, int last, Stmt stmt) {
exists(SimilarBlock b1, SimilarBlock b2, Location startloc, Location endloc |
stmt.getLocation() = startloc and
stmt.getLastStatement().getLocation() = endloc and
first = b1.tokenStartingAt(startloc) and
last = b2.tokenEndingAt(endloc) and
b1.getEquivalenceClass() = equivstart and
b2.getEquivalenceClass() = equivend and
similar_extension(b1, _, b2, _, equivstart, equivend)
)
}
/**
* Holds if `sc1` is a function or toplevel with `total` lines, and `scope2` is a function or
* toplevel that has `similar` similar lines to `scope1`.
*/
private predicate similarStatements(Scope scope1, Scope scope2, int similar, int total) {
similar = strictcount(Stmt stmt | similarStatement(scope1, scope2, stmt, _)) and
total = strictcount(Stmt stmt | scope1.contains(stmt))
}
/**
* Find pairs of scopes that are similar
*/
predicate similarScopes(Scope s, Scope other, float percent, string message) {
exists(int total, int similar |
similarStatements(s, other, similar, total) |
percent = 100.0 * similar / total and percent >= 80.0 and
if similar = total then
message = "All statements in " + s.getName() + " are similar in $@."
else
message = similar + " out of " + total + " statements in " + s.getName() + " are similar in $@."
)
}
/**
* Holds if the line is acceptable as a duplicate.
* This is true for blocks of import statements.
*/
predicate whitelistedLineForDuplication(File f, int line) {
exists(ImportingStmt i |
i.getLocation().getFile() = f and i.getLocation().getStartLine() = line
)
}