/ Check-in [a611c750]
Login
SQLite training in Houston TX on 2019-11-05 (details)
Part of the 2019 Tcl Conference

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:More test cases for the REGEXP operator. Fix minor bugs uncovered by these test cases.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: a611c75061c8e821cb266fcb09759100d4a646b0
User & Date: drh 2012-12-31 20:16:35
Context
2013-01-01
13:55
Enhance the table_info pragma so that the pk column shows the order of the columns in a multi-column primary key. check-in: 3076a890 user: drh tags: trunk
2012-12-31
20:16
More test cases for the REGEXP operator. Fix minor bugs uncovered by these test cases. check-in: a611c750 user: drh tags: trunk
19:18
Add the test_regexp.c module containing a cross-platform implementation of the REGEXP operator. check-in: 46c8c01b user: drh tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to src/shell.c.

1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
      exit(1);
    }
#ifndef SQLITE_OMIT_LOAD_EXTENSION
    sqlite3_enable_load_extension(p->db, 1);
#endif
#ifdef SQLITE_ENABLE_REGEXP
    {
      extern sqlite3_add_regexp_func(sqlite3*);
      sqlite3_add_regexp_func(db);
    }
#endif
  }
}

/*







|







1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
      exit(1);
    }
#ifndef SQLITE_OMIT_LOAD_EXTENSION
    sqlite3_enable_load_extension(p->db, 1);
#endif
#ifdef SQLITE_ENABLE_REGEXP
    {
      extern int sqlite3_add_regexp_func(sqlite3*);
      sqlite3_add_regexp_func(db);
    }
#endif
  }
}

/*

Changes to src/test_regexp.c.

338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
...
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
** If c is a hex digit, also set *pV = (*pV)*16 + valueof(c).  If
** c is not a hex digit *pV is unchanged.
*/
static int re_hex(int c, int *pV){
  if( c>='0' && c<='9' ){
    c -= '0';
  }else if( c>='a' && c<='f' ){
    c -= 'a' + 10;
  }else if( c>='A' && c<='F' ){
    c -= 'A' + 10;
  }else{
    return 0;
  }
  *pV = (*pV)*16 + c;
  return 1;
}

/* A backslash character has been seen, read the next character and
** return its intepretation.
*/
static unsigned re_esc_char(ReCompiled *p){
  static const char zEsc[] = "afnrtv\\()*.+?[$^{|";
  static const char zTrans[] = "\a\f\n\r\t\v";
  int i, v = 0;
  char c = p->zIn[0];
  if( c=='u' ){
    v = 0;
    if( re_hex(p->zIn[1],&v)
     && re_hex(p->zIn[2],&v)
................................................................................
    if( i>1 ){
      p->zIn += i;
      return v;
    }
  }
  for(i=0; zEsc[i] && zEsc[i]!=c; i++){}
  if( zEsc[i] ){
    if( c<6 ) c = zTrans[i];
    p->zIn++;
  }else{
    p->zErr = "unknown \\ escape";
  }
  return c;
}








|

|



|







|







 







|







338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
...
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
** If c is a hex digit, also set *pV = (*pV)*16 + valueof(c).  If
** c is not a hex digit *pV is unchanged.
*/
static int re_hex(int c, int *pV){
  if( c>='0' && c<='9' ){
    c -= '0';
  }else if( c>='a' && c<='f' ){
    c -= 'a' - 10;
  }else if( c>='A' && c<='F' ){
    c -= 'A' - 10;
  }else{
    return 0;
  }
  *pV = (*pV)*16 + (c & 0xff);
  return 1;
}

/* A backslash character has been seen, read the next character and
** return its intepretation.
*/
static unsigned re_esc_char(ReCompiled *p){
  static const char zEsc[] = "afnrtv\\()*.+?[$^{|}]";
  static const char zTrans[] = "\a\f\n\r\t\v";
  int i, v = 0;
  char c = p->zIn[0];
  if( c=='u' ){
    v = 0;
    if( re_hex(p->zIn[1],&v)
     && re_hex(p->zIn[2],&v)
................................................................................
    if( i>1 ){
      p->zIn += i;
      return v;
    }
  }
  for(i=0; zEsc[i] && zEsc[i]!=c; i++){}
  if( zEsc[i] ){
    if( i<6 ) c = zTrans[i];
    p->zIn++;
  }else{
    p->zErr = "unknown \\ escape";
  }
  return c;
}

Changes to test/regexp1.test.

75
76
77
78
79
80
81






























82




































83





























































84
} {}
do_execsql_test regexp1-1.17 {
  SELECT x FROM t1 WHERE y REGEXP 'alive.$' ORDER BY x;
} {4}
do_execsql_test regexp1-1.18 {
  SELECT x FROM t1 WHERE y REGEXP 'alive\.$' ORDER BY x;
} {4}

































































































































finish_test







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

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
} {}
do_execsql_test regexp1-1.17 {
  SELECT x FROM t1 WHERE y REGEXP 'alive.$' ORDER BY x;
} {4}
do_execsql_test regexp1-1.18 {
  SELECT x FROM t1 WHERE y REGEXP 'alive\.$' ORDER BY x;
} {4}
do_execsql_test regexp1-1.19 {
  SELECT x FROM t1 WHERE y REGEXP 'ma[nd]' ORDER BY x;
} {1 2 4}
do_execsql_test regexp1-1.20 {
  SELECT x FROM t1 WHERE y REGEXP '\bma[nd]' ORDER BY x;
} {1 2 4}
do_execsql_test regexp1-1.21 {
  SELECT x FROM t1 WHERE y REGEXP 'ma[nd]\b' ORDER BY x;
} {1 2}
do_execsql_test regexp1-1.22 {
  SELECT x FROM t1 WHERE y REGEXP 'ma\w' ORDER BY x;
} {1 2 4}
do_execsql_test regexp1-1.23 {
  SELECT x FROM t1 WHERE y REGEXP 'ma\W' ORDER BY x;
} {}
do_execsql_test regexp1-1.24 {
  SELECT x FROM t1 WHERE y REGEXP '\sma\w' ORDER BY x;
} {1 2 4}
do_execsql_test regexp1-1.25 {
  SELECT x FROM t1 WHERE y REGEXP '\Sma\w' ORDER BY x;
} {}
do_execsql_test regexp1-1.26 {
  SELECT x FROM t1 WHERE y REGEXP 'alive\S$' ORDER BY x;
} {4}
do_execsql_test regexp1-1.27 {
  SELECT x FROM t1 WHERE y REGEXP
          '\b(unto|us|son|given|his|name|called|' ||
          'wonderful|councelor|mighty|god|everlasting|father|' ||
          'prince|peace|alive)\b';
} {4}

do_execsql_test regexp1-2.1 {
  SELECT 'aaaabbbbcccc' REGEXP 'ab*c', 
         'aaaacccc' REGEXP 'ab*c';
} {1 1}
do_execsql_test regexp1-2.2 {
  SELECT 'aaaabbbbcccc' REGEXP 'ab+c',
         'aaaacccc' REGEXP 'ab+c';
} {1 0}
do_execsql_test regexp1-2.3 {
  SELECT 'aaaabbbbcccc' REGEXP 'ab?c',
         'aaaacccc' REGEXP 'ab?c';
} {0 1}
do_execsql_test regexp1-2.4 {
  SELECT 'aaaabbbbbbcccc' REGEXP 'ab{3,5}c',
         'aaaabbbbbcccc' REGEXP 'ab{3,5}c',
         'aaaabbbbcccc' REGEXP 'ab{3,5}c',
         'aaaabbbcccc' REGEXP 'ab{3,5}c',
         'aaaabbcccc' REGEXP 'ab{3,5}c',
         'aaaabcccc' REGEXP 'ab{3,5}c'
} {0 1 1 1 0 0}
do_execsql_test regexp1-2.5 {
  SELECT 'aaaabbbbcccc' REGEXP 'a(a|b|c)+c',
         'aaaabbbbcccc' REGEXP '^a(a|b|c){11}c$',
         'aaaabbbbcccc' REGEXP '^a(a|b|c){10}c$',
         'aaaabbbbcccc' REGEXP '^a(a|b|c){9}c$'
} {1 0 1 0}
do_execsql_test regexp1-2.6 {
  SELECT 'aaaabbbbcccc' REGEXP '^a(a|bb|c)+c$',
         'aaaabbbbcccc' REGEXP '^a(a|bbb|c)+c$',
         'aaaabbbbcccc' REGEXP '^a(a|bbbb|c)+c$'
} {1 0 1}
do_execsql_test regexp1-2.7 {
  SELECT 'aaaabbbbcccc' REGEXP '^a([ac]+|bb){3}c$',
         'aaaabbbbcccc' REGEXP '^a([ac]+|bb){4}c$',
         'aaaabbbbcccc' REGEXP '^a([ac]+|bb){5}c$'
} {0 1 1}

do_execsql_test regexp1-2.8 {
  SELECT 'abc*def+ghi.jkl[mno]pqr' REGEXP 'c.d',
         'abc*def+ghi.jkl[mno]pqr' REGEXP 'c\*d',
         'abc*def+ghi.jkl[mno]pqr' REGEXP 'f\+g',
         'abc*def+ghi.jkl[mno]pqr' REGEXP 'i\.j',
         'abc*def+ghi.jkl[mno]pqr' REGEXP 'l\[mno\]p'
} {1 1 1 1 1}

do_test regexp1-2.9 {
  set v1 "abc\ndef"
  db eval {SELECT $v1 REGEXP '^abc\ndef$'}
} {1}
do_test regexp1-2.10 {
  set v1 "abc\adef"
  db eval {SELECT $v1 REGEXP '^abc\adef$'}
} {1}
do_test regexp1-2.11 {
  set v1 "abc\tdef"
  db eval {SELECT $v1 REGEXP '^abc\tdef$'}
} {1}
do_test regexp1-2.12 {
  set v1 "abc\rdef"
  db eval {SELECT $v1 REGEXP '^abc\rdef$'}
} {1}
do_test regexp1-2.13 {
  set v1 "abc\fdef"
  db eval {SELECT $v1 REGEXP '^abc\fdef$'}
} {1}
do_test regexp1-2.14 {
  set v1 "abc\vdef"
  db eval {SELECT $v1 REGEXP '^abc\vdef$'}
} {1}
do_execsql_test regexp1-2.15 {
  SELECT 'abc\def' REGEXP '^abc\\def',
         'abc(def' REGEXP '^abc\(def',
         'abc)def' REGEXP '^abc\)def',
         'abc*def' REGEXP '^abc\*def',
         'abc.def' REGEXP '^abc\.def',
         'abc+def' REGEXP '^abc\+def',
         'abc?def' REGEXP '^abc\?def',
         'abc[def' REGEXP '^abc\[def',
         'abc$def' REGEXP '^abc\$',
         '^def'    REGEXP '\^def',
         'abc{4}x' REGEXP '^abc\{4\}x$',
         'abc|def' REGEXP '^abc\|def$'
} {1 1 1 1 1 1 1 1 1 1 1 1}

do_execsql_test regexp1-2.20 {
  SELECT 'abc$¢€xyz' REGEXP '^abc\u0024\u00a2\u20acxyz$',
         'abc$¢€xyz' REGEXP '^abc\u0024\u00A2\u20ACxyz$',
         'abc$¢€xyz' REGEXP '^abc\x24\xa2\x20acxyz$'
} {1 1 1}
do_execsql_test regexp1-2.21 {
  SELECT 'abc$¢€xyz' REGEXP '^abc[\u0024][\u00a2][\u20ac]xyz$',
         'abc$¢€xyz' REGEXP '^abc[\u0024\u00A2\u20AC]{3}xyz$',
         'abc$¢€xyz' REGEXP '^abc[\x24][\xa2\x20ac]+xyz$'
} {1 1 1}
do_execsql_test regexp1-2.22 {
  SELECT 'abc$¢€xyz' REGEXP '^abc[^\u0025-X][^ -\u007f][^\u20ab]xyz$'
} {1}

finish_test