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
|
<?xml version="1.0" encoding="utf-8"?>
<test>
<name>ranker internals (PACKEDFACTORS(), different IDF variants, etc)</name>
<config>
indexer
{
mem_limit = 16M
}
searchd
{
<searchd_settings/>
}
source test
{
type = mysql
<sql_settings/>
sql_query = select * from test_table
sql_attr_uint = gid
}
index test
{
source = test
path = <data_path/>/test
index_field_lengths = 1
}
index test_stop
{
source = test
path = <data_path/>/test_stop
stopwords = stopwords.txt
}
</config>
<db_create>
CREATE TABLE test_table
(
id INTEGER PRIMARY KEY NOT NULL,
gid INTEGER NOT NULL,
title VARCHAR(255) NOT NULL
);
</db_create>
<db_drop>DROP TABLE IF EXISTS test_table;</db_drop>
<db_insert>INSERT INTO test_table VALUES
( 1, 123, 'THE document collection. Aiming for exactly 10 docs.' ),
( 2, 123, 'the idftest2' ),
( 3, 123, 'the idftest2 and idftest1' ),
( 4, 123, 'the' ),
( 5, 123, 'the' ),
( 6, 123, 'the' ),
( 7, 123, 'the' ),
( 8, 123, 'the' ),
( 9, 123, 'the' ),
( 10, 123, 'the' )
</db_insert>
<sphqueries>
<sphinxql>select *, weight(), packedfactors() from test where match('the') option ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('the') option idf=plain, ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1') option idf=normalized, ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1') option idf=plain, ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1 | fnord') option idf=normalized, ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1 | fnord') option idf=plain, ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1 | fnord') option idf='plain,tfidf_unnormalized', ranker=expr('1000000+1000000*sum(sum_idf)')</sphinxql>
<!-- regression heavy factors got enabed along with packedfactors -->
<sphinxql>select *, weight(), packedfactors() from test where match('idftest1 the') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('and idftest1 the') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('the idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors() from test where match('the and idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<!-- packedfactors with json output -->
<sphinxql>select *, weight(), factors({json=1}) from test where match('idftest1 the') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors({json=1}) from test where match('and idftest1 the') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), factors({json=1}) from test where match('the idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors({json=1}) from test where match('the and idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
<sphinxql>select *, weight(), packedfactors({json=1}) from test_stop where match('the and idftest1') option idf='plain,tfidf_unnormalized', ranker=expr('1')</sphinxql>
</sphqueries>
</test>
|
