1
0
mirror of https://github.com/sqlite/sqlite.git synced 2025-11-18 10:21:03 +03:00

Merge the unicode61 tokenizer and the shared-cache-memory database changes

into the sessions branch.

FossilOrigin-Name: df817e70afc3f41e680d8f84dfa5772d5b3ae4d9
This commit is contained in:
drh
2012-05-28 18:22:41 +00:00
29 changed files with 27253 additions and 68 deletions

View File

@@ -254,6 +254,8 @@ foreach {tn uri error} "
4 {file:test.db?mode=Ro} {no such access mode: Ro}
5 {file:test.db?mode=Rw} {no such access mode: Rw}
6 {file:test.db?mode=Rwc} {no such access mode: Rwc}
7 {file:test.db?mode=memory} {not an error}
8 {file:test.db?mode=MEMORY} {no such access mode: MEMORY}
" {
do_test 7.$tn { open_uri_error $uri } $error
}

View File

@@ -131,4 +131,28 @@ do_faultsim_test 4.1 -prep {
faultsim_test_result {0 {}}
}
ifcapable fts3_unicode {
do_test 5.0 {
faultsim_delete_and_reopen
execsql {
CREATE VIRTUAL TABLE ft USING fts4(a, tokenize=unicode61);
}
faultsim_save_and_close
} {}
do_faultsim_test 5.1 -faults oom* -prep {
faultsim_restore_and_reopen
db eval {SELECT * FROM sqlite_master}
} -body {
execsql { INSERT INTO ft VALUES('the quick brown fox'); }
execsql { INSERT INTO ft VALUES(
'theunusuallylongtokenthatjustdragsonandonandonandthendragsonsomemoreeof'
);
}
execsql { SELECT docid FROM ft WHERE ft MATCH 'th*' }
} -test {
faultsim_test_result {0 {1 2}}
}
}
finish_test

227
test/fts4unicode.test Normal file
View File

@@ -0,0 +1,227 @@
# 2012 May 25
#
# The author disclaims copyright to this source code. In place of
# a legal notice, here is a blessing:
#
# May you do good and not evil.
# May you find forgiveness for yourself and forgive others.
# May you share freely, never taking more than you give.
#
#*************************************************************************
#
# The tests in this file focus on testing the "unicode" FTS tokenizer.
#
set testdir [file dirname $argv0]
source $testdir/tester.tcl
ifcapable !fts3_unicode { finish_test ; return }
set ::testprefix fts4unicode
proc do_unicode_token_test {tn input res} {
set input [string map {' ''} $input]
uplevel [list do_execsql_test $tn "
SELECT fts3_tokenizer_test('unicode61', '$input');
" [list [list {*}$res]]]
}
do_unicode_token_test 1.0 {a B c D} {0 a a 1 b B 2 c c 3 d D}
do_unicode_token_test 1.1 {<7B> <20> <20>} {0 <20> <20> 1 <20> <20> 2 <20> <20>}
do_unicode_token_test 1.2 {x<>x x<>x x<>x} {0 x<>x x<>x 1 x<>x x<>x 2 x<>x x<>x}
# 0x00DF is a small "sharp s". 0x1E9E is a capital sharp s.
do_unicode_token_test 1.3 "\uDF" "0 \uDF \uDF"
do_unicode_token_test 1.4 "\u1E9E" "0 <20> \u1E9E"
do_unicode_token_test 1.5 "\u1E9E" "0 \uDF \u1E9E"
do_unicode_token_test 1.6 "The quick brown fox" {
0 the The 1 quick quick 2 brown brown 3 fox fox
}
do_unicode_token_test 1.7 "The\u00bfquick\u224ebrown\u2263fox" {
0 the The 1 quick quick 2 brown brown 3 fox fox
}
#-------------------------------------------------------------------------
#
set docs [list {
Enhance the INSERT syntax to allow multiple rows to be inserted via the
VALUES clause.
} {
Enhance the CREATE VIRTUAL TABLE command to support the IF NOT EXISTS clause.
} {
Added the sqlite3_stricmp() interface as a counterpart to sqlite3_strnicmp().
} {
Added the sqlite3_db_readonly() interface.
} {
Added the SQLITE_FCNTL_PRAGMA file control, giving VFS implementations the
ability to add new PRAGMA statements or to override built-in PRAGMAs.
} {
Queries of the form: "SELECT max(x), y FROM table" returns the value of y on
the same row that contains the maximum x value.
} {
Added support for the FTS4 languageid option.
} {
Documented support for the FTS4 content option. This feature has actually
been in the code since version 3.7.9 but is only now considered to be
officially supported.
} {
Pending statements no longer block ROLLBACK. Instead, the pending statement
will return SQLITE_ABORT upon next access after the ROLLBACK.
} {
Improvements to the handling of CSV inputs in the command-line shell
} {
Fix a bug introduced in version 3.7.10 that might cause a LEFT JOIN to be
incorrectly converted into an INNER JOIN if the WHERE clause indexable terms
connected by OR.
}]
set map(a) [list "\u00C4" "\u00E4"] ; # LATIN LETTER A WITH DIAERESIS
set map(e) [list "\u00CB" "\u00EB"] ; # LATIN LETTER E WITH DIAERESIS
set map(i) [list "\u00CF" "\u00EF"] ; # LATIN LETTER I WITH DIAERESIS
set map(o) [list "\u00D6" "\u00F6"] ; # LATIN LETTER O WITH DIAERESIS
set map(u) [list "\u00DC" "\u00FC"] ; # LATIN LETTER U WITH DIAERESIS
set map(y) [list "\u0178" "\u00FF"] ; # LATIN LETTER Y WITH DIAERESIS
set map(h) [list "\u1E26" "\u1E27"] ; # LATIN LETTER H WITH DIAERESIS
set map(w) [list "\u1E84" "\u1E85"] ; # LATIN LETTER W WITH DIAERESIS
set map(x) [list "\u1E8C" "\u1E8D"] ; # LATIN LETTER X WITH DIAERESIS
foreach k [array names map] {
lappend mappings [string toupper $k] [lindex $map($k) 0]
lappend mappings $k [lindex $map($k) 1]
}
proc mapdoc {doc} {
set doc [regsub -all {[[:space:]]+} $doc " "]
string map $::mappings [string trim $doc]
}
do_test 2.0 {
execsql { CREATE VIRTUAL TABLE t2 USING fts4(tokenize=unicode61, x); }
foreach doc $docs {
set d [mapdoc $doc]
execsql { INSERT INTO t2 VALUES($d) }
}
} {}
do_test 2.1 {
set q [mapdoc "row"]
execsql { SELECT * FROM t2 WHERE t2 MATCH $q }
} [list [mapdoc {
Queries of the form: "SELECT max(x), y FROM table" returns the value of y on
the same row that contains the maximum x value.
}]]
foreach {tn query snippet} {
2 "row" {
...returns the value of y on the same [row] that contains
the maximum x value.
}
3 "ROW" {
...returns the value of y on the same [row] that contains
the maximum x value.
}
4 "rollback" {
...[ROLLBACK]. Instead, the pending statement
will return SQLITE_ABORT upon next access after the [ROLLBACK].
}
5 "rOllback" {
...[ROLLBACK]. Instead, the pending statement
will return SQLITE_ABORT upon next access after the [ROLLBACK].
}
6 "lang*" {
Added support for the FTS4 [languageid] option.
}
} {
do_test 2.$tn {
set q [mapdoc $query]
execsql { SELECT snippet(t2, '[', ']', '...') FROM t2 WHERE t2 MATCH $q }
} [list [mapdoc $snippet]]
}
#-------------------------------------------------------------------------
# Make sure the unicode61 tokenizer does not crash if it is passed a
# NULL pointer.
reset_db
do_execsql_test 3.1 {
CREATE VIRTUAL TABLE t1 USING fts4(tokenize=unicode61, x, y);
INSERT INTO t1 VALUES(NULL, 'a b c');
}
do_execsql_test 3.2 {
SELECT snippet(t1, '[', ']') FROM t1 WHERE t1 MATCH 'b'
} {{a [b] c}}
do_execsql_test 3.3 {
BEGIN;
DELETE FROM t1;
INSERT INTO t1 VALUES('b b b b b b b b b b b', 'b b b b b b b b b b b b b');
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 SELECT * FROM t1;
INSERT INTO t1 VALUES('a b c', NULL);
INSERT INTO t1 VALUES('a x c', NULL);
COMMIT;
}
do_execsql_test 3.4 {
SELECT * FROM t1 WHERE t1 MATCH 'a b';
} {{a b c} {}}
#-------------------------------------------------------------------------
#
reset_db
do_test 4.1 {
set a "abc\uFFFEdef"
set b "abc\uD800def"
set c "\uFFFEdef"
set d "\uD800def"
execsql {
CREATE VIRTUAL TABLE t1 USING fts4(tokenize=unicode61, x);
INSERT INTO t1 VALUES($a);
INSERT INTO t1 VALUES($b);
INSERT INTO t1 VALUES($c);
INSERT INTO t1 VALUES($d);
}
} {}
do_test 4.2 {
set a [binary format c* {0x61 0xF7 0xBF 0xBF 0xBF 0x62}]
set b [binary format c* {0x61 0xF7 0xBF 0xBF 0xBF 0xBF 0x62}]
set c [binary format c* {0x61 0xF7 0xBF 0xBF 0xBF 0xBF 0xBF 0x62}]
set d [binary format c* {0x61 0xF7 0xBF 0xBF 0xBF 0xBF 0xBF 0xBF 0x62}]
execsql {
INSERT INTO t1 VALUES($a);
INSERT INTO t1 VALUES($b);
INSERT INTO t1 VALUES($c);
INSERT INTO t1 VALUES($d);
}
} {}
do_test 4.3 {
set a [binary format c* {0xF7 0xBF 0xBF 0xBF}]
set b [binary format c* {0xF7 0xBF 0xBF 0xBF 0xBF}]
set c [binary format c* {0xF7 0xBF 0xBF 0xBF 0xBF 0xBF}]
set d [binary format c* {0xF7 0xBF 0xBF 0xBF 0xBF 0xBF 0xBF}]
execsql {
INSERT INTO t1 VALUES($a);
INSERT INTO t1 VALUES($b);
INSERT INTO t1 VALUES($c);
INSERT INTO t1 VALUES($d);
}
} {}
finish_test

View File

@@ -188,7 +188,7 @@ test_suite "fts3" -prefix "" -description {
fts4aa.test fts4content.test
fts3conf.test fts3prefix.test fts3fault2.test fts3corrupt.test
fts3corrupt2.test fts3first.test fts4langid.test fts4merge.test
fts4check.test
fts4check.test fts4unicode.test
}

View File

@@ -1056,7 +1056,96 @@ do_test shared-$av-15.2 {
db close
db2 close
}
# Shared cache on a :memory: database. This only works for URI filenames.
#
do_test shared-$av-16.1 {
sqlite3 db1 file::memory: -uri 1
sqlite3 db2 file::memory: -uri 1
db1 eval {
CREATE TABLE t1(x); INSERT INTO t1 VALUES(1),(2),(3);
}
db2 eval {
SELECT x FROM t1 ORDER BY x;
}
} {1 2 3}
do_test shared-$av-16.2 {
db2 eval {
INSERT INTO t1 VALUES(99);
DELETE FROM t1 WHERE x=2;
}
db1 eval {
SELECT x FROM t1 ORDER BY x;
}
} {1 3 99}
# Verify that there is no cache sharing ordinary (non-URI) filenames are
# used.
#
do_test shared-$av-16.3 {
db1 close
db2 close
sqlite3 db1 :memory:
sqlite3 db2 :memory:
db1 eval {
CREATE TABLE t1(x); INSERT INTO t1 VALUES(4),(5),(6);
}
catchsql {
SELECT * FROM t1;
} db2
} {1 {no such table: t1}}
# Shared cache on named memory databases.
#
do_test shared-$av-16.4 {
db1 close
db2 close
forcedelete test.db test.db-wal test.db-journal
sqlite3 db1 file:test.db?mode=memory -uri 1
sqlite3 db2 file:test.db?mode=memory -uri 1
db1 eval {
CREATE TABLE t1(x); INSERT INTO t1 VALUES(1),(2),(3);
}
db2 eval {
SELECT x FROM t1 ORDER BY x;
}
} {1 2 3}
do_test shared-$av-16.5 {
db2 eval {
INSERT INTO t1 VALUES(99);
DELETE FROM t1 WHERE x=2;
}
db1 eval {
SELECT x FROM t1 ORDER BY x;
}
} {1 3 99}
do_test shared-$av-16.6 {
file exists test.db
} {0} ;# Verify that the database is in-memory
# Shared cache on named memory databases with different names.
#
do_test shared-$av-16.7 {
db1 close
db2 close
forcedelete test1.db test2.db
sqlite3 db1 file:test1.db?mode=memory -uri 1
sqlite3 db2 file:test2.db?mode=memory -uri 1
db1 eval {
CREATE TABLE t1(x); INSERT INTO t1 VALUES(1),(2),(3);
}
catchsql {
SELECT x FROM t1 ORDER BY x;
} db2
} {1 {no such table: t1}}
do_test shared-$av-16.8 {
file exists test1.db
} {0} ;# Verify that the database is in-memory
db1 close
db2 close
} ;# end of autovacuum on/off loop
sqlite3_enable_shared_cache $::enable_shared_cache
finish_test