line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Convert::Pheno::SQLite; |
2
|
|
|
|
|
|
|
|
3
|
6
|
|
|
6
|
|
40
|
use strict; |
|
6
|
|
|
|
|
17
|
|
|
6
|
|
|
|
|
177
|
|
4
|
6
|
|
|
6
|
|
39
|
use warnings; |
|
6
|
|
|
|
|
16
|
|
|
6
|
|
|
|
|
141
|
|
5
|
6
|
|
|
6
|
|
26
|
use autodie; |
|
6
|
|
|
|
|
9
|
|
|
6
|
|
|
|
|
44
|
|
6
|
6
|
|
|
6
|
|
32879
|
use feature qw(say); |
|
6
|
|
|
|
|
17
|
|
|
6
|
|
|
|
|
549
|
|
7
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
#use Carp qw(confess); |
9
|
6
|
|
|
6
|
|
10802
|
use DBI; |
|
6
|
|
|
|
|
112926
|
|
|
6
|
|
|
|
|
452
|
|
10
|
6
|
|
|
6
|
|
62
|
use File::Spec::Functions qw(catdir catfile); |
|
6
|
|
|
|
|
13
|
|
|
6
|
|
|
|
|
355
|
|
11
|
6
|
|
|
6
|
|
37
|
use Data::Dumper; |
|
6
|
|
|
|
|
15
|
|
|
6
|
|
|
|
|
256
|
|
12
|
6
|
|
|
6
|
|
3401
|
use Text::Similarity::Overlaps; |
|
6
|
|
|
|
|
37134
|
|
|
6
|
|
|
|
|
317
|
|
13
|
6
|
|
|
6
|
|
49
|
use Exporter 'import'; |
|
6
|
|
|
|
|
20
|
|
|
6
|
|
|
|
|
348
|
|
14
|
|
|
|
|
|
|
our @EXPORT = |
15
|
|
|
|
|
|
|
qw( $VERSION open_connections_SQLite close_connections_SQLite get_ontology); |
16
|
|
|
|
|
|
|
|
17
|
|
|
|
|
|
|
my @sqlites = qw(ncit icd10 ohdsi cdisc omim hpo); |
18
|
|
|
|
|
|
|
my @matches = qw(exact_match full_text_search contains); |
19
|
6
|
|
|
6
|
|
35
|
use constant DEVEL_MODE => 0; |
|
6
|
|
|
|
|
17
|
|
|
6
|
|
|
|
|
11283
|
|
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
######################## |
22
|
|
|
|
|
|
|
######################## |
23
|
|
|
|
|
|
|
# SUBROUTINES FOR DB # |
24
|
|
|
|
|
|
|
######################## |
25
|
|
|
|
|
|
|
######################## |
26
|
|
|
|
|
|
|
|
27
|
|
|
|
|
|
|
sub open_connections_SQLite { |
28
|
|
|
|
|
|
|
|
29
|
10
|
|
|
10
|
0
|
24
|
my $self = shift; |
30
|
|
|
|
|
|
|
|
31
|
|
|
|
|
|
|
# ********************** |
32
|
|
|
|
|
|
|
# *** IMPORTANT STEP *** |
33
|
|
|
|
|
|
|
# ********************** |
34
|
|
|
|
|
|
|
# Well open ALL databases once (instead that on each call), regardless if they user has selected them. |
35
|
|
|
|
|
|
|
# It imrpoves speed by 15% |
36
|
|
|
|
|
|
|
# The only exception is for <ohdsi> that is the larger and may interfere in timings |
37
|
|
|
|
|
|
|
|
38
|
|
|
|
|
|
|
# Only open ohdsi.db if $self->{ohdsi_db} |
39
|
|
|
|
|
|
|
my @databases = |
40
|
10
|
50
|
|
|
|
69
|
$self->{ohdsi_db} ? @sqlites : grep { !m/ohdsi/ } @sqlites; # global |
|
60
|
|
|
|
|
214
|
|
41
|
|
|
|
|
|
|
|
42
|
|
|
|
|
|
|
# Open databases |
43
|
10
|
|
|
|
|
31
|
my $dbh; |
44
|
|
|
|
|
|
|
$dbh->{$_} = open_db_SQLite( $_, $self->{path_to_ohdsi_db} ) |
45
|
10
|
|
|
|
|
65
|
for (@databases); |
46
|
|
|
|
|
|
|
|
47
|
|
|
|
|
|
|
# Add $dbh HANDLE to $self |
48
|
10
|
|
|
|
|
76
|
$self->{dbh} = $dbh; # Dynamically adding attributes (setter) |
49
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
# Prepare the query once |
51
|
10
|
|
|
|
|
56
|
prepare_query_SQLite($self); |
52
|
|
|
|
|
|
|
|
53
|
10
|
|
|
|
|
45
|
return 1; |
54
|
|
|
|
|
|
|
} |
55
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
sub close_connections_SQLite { |
57
|
|
|
|
|
|
|
|
58
|
10
|
|
|
10
|
0
|
38
|
my $self = shift; |
59
|
10
|
|
|
|
|
59
|
my $dbh = $self->{dbh}; |
60
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
# Check flag ohdsi_db |
62
|
|
|
|
|
|
|
my @databases = |
63
|
10
|
50
|
|
|
|
76
|
$self->{ohdsi_db} ? @sqlites : grep { !m/ohdsi/ } @sqlites; # global |
|
60
|
|
|
|
|
194
|
|
64
|
10
|
|
|
|
|
82
|
close_db_SQLite( $dbh->{$_} ) for (@databases); |
65
|
10
|
|
|
|
|
49
|
return 1; |
66
|
|
|
|
|
|
|
} |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
sub open_db_SQLite { |
69
|
|
|
|
|
|
|
|
70
|
50
|
|
|
50
|
0
|
169
|
my ( $ontology, $path_to_ohdsi_db ) = @_; |
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
# Search file in two dirs, except for ($ontology eq 'ohdsi' && defined $path_to_ohdsi_db) |
73
|
50
|
|
|
|
|
116
|
my $filename = qq/$ontology.db/; |
74
|
50
|
50
|
33
|
|
|
418
|
my $path = |
75
|
|
|
|
|
|
|
( $ontology eq 'ohdsi' && defined $path_to_ohdsi_db ) |
76
|
|
|
|
|
|
|
? $path_to_ohdsi_db |
77
|
|
|
|
|
|
|
: catdir( $Convert::Pheno::share_dir, 'db' ); |
78
|
50
|
|
|
|
|
208
|
my $dbfile = catfile( $path, $filename ); |
79
|
50
|
50
|
|
|
|
1262
|
die "Sorry we could not find <$dbfile> file\n" unless -f $dbfile; |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
# Connect to the database |
82
|
50
|
|
|
|
|
160
|
my $user = ''; |
83
|
50
|
|
|
|
|
70
|
my $passwd = ''; |
84
|
50
|
|
|
|
|
171
|
my $dsn = "dbi:SQLite:dbname=$dbfile"; |
85
|
50
|
|
|
|
|
512
|
my $dbh = DBI->connect( |
86
|
|
|
|
|
|
|
$dsn, $user, $passwd, |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
# PRAGMAs |
89
|
|
|
|
|
|
|
{ |
90
|
|
|
|
|
|
|
PrintError => 0, |
91
|
|
|
|
|
|
|
RaiseError => 1, |
92
|
|
|
|
|
|
|
ReadOnly => 1, |
93
|
|
|
|
|
|
|
AutoCommit => 1, |
94
|
|
|
|
|
|
|
FetchHashKeyName => 'NAME_lc', |
95
|
|
|
|
|
|
|
} |
96
|
|
|
|
|
|
|
); |
97
|
|
|
|
|
|
|
|
98
|
|
|
|
|
|
|
# These extra PRAGMAs are supposed to speed-up queries?? |
99
|
50
|
|
|
|
|
59981
|
$dbh->do("PRAGMA synchronous = OFF"); |
100
|
50
|
|
|
|
|
12551
|
$dbh->do("PRAGMA cache_size = 800000"); |
101
|
|
|
|
|
|
|
|
102
|
50
|
|
|
|
|
1434
|
return $dbh; |
103
|
|
|
|
|
|
|
} |
104
|
|
|
|
|
|
|
|
105
|
|
|
|
|
|
|
sub close_db_SQLite { |
106
|
|
|
|
|
|
|
|
107
|
50
|
|
|
50
|
0
|
95
|
my $dbh = shift; |
108
|
50
|
|
|
|
|
8131
|
$dbh->disconnect(); |
109
|
50
|
|
|
|
|
273
|
return 1; |
110
|
|
|
|
|
|
|
} |
111
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
sub prepare_query_SQLite { |
113
|
|
|
|
|
|
|
|
114
|
10
|
|
|
10
|
0
|
25
|
my $self = shift; |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
############### |
117
|
|
|
|
|
|
|
# EXPLANATION # |
118
|
|
|
|
|
|
|
############### |
119
|
|
|
|
|
|
|
# |
120
|
|
|
|
|
|
|
# Even though we did not gain a lot of speed (~15%), we decided to do the "prepare step" once, instead of on each query. |
121
|
|
|
|
|
|
|
# Then, if we want to search in a different column than 'label' we also need to create that $sth |
122
|
|
|
|
|
|
|
# To solve that we have created a nested sth->{ncit}{label}, sth->{icd10}{label}, sth->{ohdsi}{concept_id} and sth->{ohdsi}{label} |
123
|
|
|
|
|
|
|
# On top of that, we add the "match" type, so that we can have other matches in the future if needed |
124
|
|
|
|
|
|
|
# NB: In principle, is is possible to change the "prepare" during queries but we must revert it back to default after using it |
125
|
|
|
|
|
|
|
# We recommend using small db such as ncit/icd10 as they're fast |
126
|
|
|
|
|
|
|
|
127
|
|
|
|
|
|
|
# Check flag ohdsi_db |
128
|
|
|
|
|
|
|
my @databases = |
129
|
10
|
50
|
|
|
|
55
|
$self->{ohdsi_db} ? @sqlites : grep { !m/ohdsi/ } @sqlites; # global |
|
60
|
|
|
|
|
171
|
|
130
|
|
|
|
|
|
|
|
131
|
|
|
|
|
|
|
# NB: |
132
|
|
|
|
|
|
|
# dbh = "Database Handle" |
133
|
|
|
|
|
|
|
# sth = "Statement Handle" |
134
|
|
|
|
|
|
|
|
135
|
10
|
|
|
|
|
65
|
for my $match (@matches) { |
136
|
30
|
|
|
|
|
61
|
for my $ontology (@databases) { #global |
137
|
150
|
|
|
|
|
224
|
for my $column ( 'label', 'concept_id' ) { |
138
|
|
|
|
|
|
|
|
139
|
|
|
|
|
|
|
# We only need to open 'concept_id' in ohdsi |
140
|
300
|
100
|
66
|
|
|
1135
|
next if ( $column eq 'concept_id' && $ontology ne 'ohdsi' ); |
141
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
############################## |
143
|
|
|
|
|
|
|
# Start building the queries # |
144
|
|
|
|
|
|
|
############################## |
145
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
# NCIT_table or NCIT_fts depending on type of match |
147
|
150
|
|
|
|
|
363
|
my $db = uc($ontology) . '_table'; |
148
|
150
|
|
|
|
|
220
|
my $db_fts = uc($ontology) . '_fts'; |
149
|
150
|
|
|
|
|
266
|
my $dbh = $self->{dbh}{$ontology}; |
150
|
150
|
|
|
|
|
957
|
my %query_type = ( |
151
|
|
|
|
|
|
|
|
152
|
|
|
|
|
|
|
# Regular queries |
153
|
|
|
|
|
|
|
contains => |
154
|
|
|
|
|
|
|
qq(SELECT * FROM $db WHERE $column LIKE '%' || ? || '%' COLLATE NOCASE) |
155
|
|
|
|
|
|
|
, # NOT USED |
156
|
|
|
|
|
|
|
|
157
|
|
|
|
|
|
|
#begins_with => qq(SELECT * FROM $db WHERE $column LIKE ? || '%' COLLATE NOCASE), # NOT USED |
158
|
|
|
|
|
|
|
exact_match => |
159
|
|
|
|
|
|
|
qq(SELECT * FROM $db WHERE $column = ? COLLATE NOCASE), |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
# ********************** |
162
|
|
|
|
|
|
|
# *** IMPORTANT STEP *** |
163
|
|
|
|
|
|
|
# ********************** |
164
|
|
|
|
|
|
|
|
165
|
|
|
|
|
|
|
# Full-text-search queries only on column <label> BUT IT CAN BE DONE ALL COLUMNS!!!! |
166
|
|
|
|
|
|
|
# The speed of the FTS in $column == $db_fts |
167
|
|
|
|
|
|
|
# FTS is 2x faster than 'contains' |
168
|
|
|
|
|
|
|
# NOTE (Jan-2023): We don't check for misspelled words |
169
|
|
|
|
|
|
|
# --> TO DO - Tricky --> https://www.sqlite.org/spellfix1.html |
170
|
|
|
|
|
|
|
full_text_search => |
171
|
|
|
|
|
|
|
qq(SELECT * FROM $db_fts WHERE $column MATCH ?) |
172
|
|
|
|
|
|
|
, # SINGLE COLUMN |
173
|
|
|
|
|
|
|
#qq(SELECT * FROM $db_fts WHERE $db_fts MATCH ?), # ALL TABLE |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
# SOUNDEX using TABLE_fts but only on column <label> |
176
|
|
|
|
|
|
|
# soundex => qq(SELECT * FROM $db_fts WHERE SOUNDEX($column) = SOUNDEX(?)) # NOT USED |
177
|
|
|
|
|
|
|
); |
178
|
|
|
|
|
|
|
|
179
|
|
|
|
|
|
|
# Prepare the query |
180
|
150
|
|
|
|
|
797
|
my $sth = $dbh->prepare( $query_type{$match} ); |
181
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
# Autovivification of $self->{sth}{$ontology}{$column}{$match} |
183
|
150
|
|
|
|
|
22863
|
$self->{sth}{$ontology}{$column}{$match} = |
184
|
|
|
|
|
|
|
$sth; # Dynamically adding nested attributes (setter) |
185
|
|
|
|
|
|
|
} |
186
|
|
|
|
|
|
|
} |
187
|
|
|
|
|
|
|
} |
188
|
|
|
|
|
|
|
|
189
|
|
|
|
|
|
|
#print Dumper $self and die; |
190
|
10
|
|
|
|
|
51
|
return 1; |
191
|
|
|
|
|
|
|
} |
192
|
|
|
|
|
|
|
|
193
|
|
|
|
|
|
|
sub get_ontology { |
194
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
############### |
196
|
|
|
|
|
|
|
# START QUERY # |
197
|
|
|
|
|
|
|
############### |
198
|
|
|
|
|
|
|
|
199
|
206
|
|
|
206
|
0
|
322
|
my $arg = shift; |
200
|
206
|
|
|
|
|
354
|
my $ontology = $arg->{ontology}; |
201
|
206
|
|
|
|
|
278
|
my $sth_column_ref = $arg->{sth_column_ref}; #it contains hashref |
202
|
206
|
|
|
|
|
327
|
my $query = $arg->{query}; |
203
|
206
|
|
|
|
|
282
|
my $column = $arg->{column}; |
204
|
206
|
|
|
|
|
246
|
my $search = $arg->{search}; |
205
|
206
|
|
|
|
|
257
|
my $text_similarity_method = $arg->{text_similarity_method}; |
206
|
206
|
|
|
|
|
320
|
my $min_text_similarity_score = $arg->{min_text_similarity_score}; |
207
|
206
|
|
|
|
|
295
|
my $type_of_search = 'full_text_search'; # 'contains' and 'full_text_search' |
208
|
|
|
|
|
|
|
#say $type_of_search; |
209
|
206
|
|
|
|
|
239
|
say "QUERY <$query>" if DEVEL_MODE; |
210
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
# A) 'exact' |
212
|
|
|
|
|
|
|
# - exact_match |
213
|
|
|
|
|
|
|
# B) Mixed queries: |
214
|
|
|
|
|
|
|
# 1 - exact_match |
215
|
|
|
|
|
|
|
# if we don't get results |
216
|
|
|
|
|
|
|
# 2 - contains |
217
|
|
|
|
|
|
|
# for which we rank by similarity w/ Text:Similarity |
218
|
|
|
|
|
|
|
|
219
|
206
|
50
|
|
|
|
611
|
my $default_id = |
220
|
|
|
|
|
|
|
$ontology eq 'hpo' ? 'HP:NA0000' : uc($ontology) . ':NA0000'; |
221
|
206
|
|
|
|
|
278
|
my $default_label = 'NA'; |
222
|
|
|
|
|
|
|
|
223
|
|
|
|
|
|
|
# exact_match (always performed) |
224
|
|
|
|
|
|
|
my ( $id, $label ) = execute_query_SQLite( |
225
|
|
|
|
|
|
|
{ |
226
|
|
|
|
|
|
|
sth => $sth_column_ref->{exact_match}, # IMPORTANT STEP |
227
|
206
|
|
|
|
|
1230
|
query => $query, |
228
|
|
|
|
|
|
|
ontology => $ontology, |
229
|
|
|
|
|
|
|
match => 'exact_match', |
230
|
|
|
|
|
|
|
text_similarity_method => $text_similarity_method, # Not used here |
231
|
|
|
|
|
|
|
min_text_similarity_score => $min_text_similarity_score |
232
|
|
|
|
|
|
|
} |
233
|
|
|
|
|
|
|
); |
234
|
|
|
|
|
|
|
|
235
|
|
|
|
|
|
|
# mixed |
236
|
206
|
50
|
0
|
|
|
1004
|
if ( $search eq 'mixed' && ( !defined $id && !defined $label ) ) { |
|
|
|
33
|
|
|
|
|
237
|
|
|
|
|
|
|
( $id, $label ) = execute_query_SQLite( |
238
|
|
|
|
|
|
|
{ |
239
|
0
|
|
|
|
|
0
|
sth => $sth_column_ref->{$type_of_search}, # IMPORTANT STEP |
240
|
|
|
|
|
|
|
query => $query, |
241
|
|
|
|
|
|
|
ontology => $ontology, |
242
|
|
|
|
|
|
|
match => $type_of_search, |
243
|
|
|
|
|
|
|
text_similarity_method => $text_similarity_method, |
244
|
|
|
|
|
|
|
min_text_similarity_score => $min_text_similarity_score |
245
|
|
|
|
|
|
|
} |
246
|
|
|
|
|
|
|
); |
247
|
|
|
|
|
|
|
} |
248
|
|
|
|
|
|
|
|
249
|
|
|
|
|
|
|
# Set defaults if undef |
250
|
206
|
|
66
|
|
|
566
|
$id = $id // $default_id; |
251
|
206
|
|
66
|
|
|
489
|
$label = $label // $default_label; |
252
|
|
|
|
|
|
|
|
253
|
|
|
|
|
|
|
############# |
254
|
|
|
|
|
|
|
# END QUERY # |
255
|
|
|
|
|
|
|
############# |
256
|
|
|
|
|
|
|
|
257
|
206
|
|
|
|
|
1024
|
return ( $id, $label ); |
258
|
|
|
|
|
|
|
|
259
|
|
|
|
|
|
|
} |
260
|
|
|
|
|
|
|
|
261
|
|
|
|
|
|
|
sub execute_query_SQLite { |
262
|
|
|
|
|
|
|
|
263
|
206
|
|
|
206
|
0
|
342
|
my $arg = shift; |
264
|
206
|
|
|
|
|
310
|
my $sth = $arg->{sth}; |
265
|
206
|
|
|
|
|
369
|
my $query = $arg->{query}; |
266
|
206
|
|
|
|
|
316
|
my $text_similarity_method = $arg->{text_similarity_method}; |
267
|
206
|
|
|
|
|
280
|
my $min_text_similarity_score = $arg->{min_text_similarity_score}; |
268
|
206
|
|
|
|
|
276
|
my $ontology = $arg->{ontology}; |
269
|
206
|
|
|
|
|
321
|
my $match = $arg->{match}; |
270
|
|
|
|
|
|
|
|
271
|
|
|
|
|
|
|
# set $id and $label to undef |
272
|
206
|
|
|
|
|
391
|
my ( $id, $label ) = ( undef, undef ); |
273
|
|
|
|
|
|
|
|
274
|
|
|
|
|
|
|
# Premature return if $query eq '' |
275
|
206
|
100
|
|
|
|
449
|
return ( $id, $label ) if $query eq ''; |
276
|
|
|
|
|
|
|
|
277
|
|
|
|
|
|
|
# Columns in DBs |
278
|
|
|
|
|
|
|
# *<ncit.db>, <icd10.db> and <cdisc.db> were pre-processed to have "id" and "label" columns only |
279
|
|
|
|
|
|
|
# label [0] |
280
|
|
|
|
|
|
|
# id [1] |
281
|
|
|
|
|
|
|
# |
282
|
|
|
|
|
|
|
# * <ohdsi.db> consists of 4 columns: |
283
|
|
|
|
|
|
|
# concept_name => label [0] |
284
|
|
|
|
|
|
|
# concept_code => id [1] |
285
|
|
|
|
|
|
|
# concept_id => concept_id [2] |
286
|
|
|
|
|
|
|
# vocabulary_id => vocabulary_id [3] |
287
|
|
|
|
|
|
|
|
288
|
|
|
|
|
|
|
# Define a hash for column position on databases |
289
|
|
|
|
|
|
|
# We may encounter a situation where order of columns is different |
290
|
204
|
|
|
|
|
349
|
my $position = {}; |
291
|
204
|
|
|
|
|
7712
|
$position->{$_} = { label => 0, id => 1 } for (@sqlites); |
292
|
204
|
|
|
|
|
781
|
my $id_column = $position->{$ontology}{id}; |
293
|
204
|
|
|
|
|
297
|
my $label_column = $position->{$ontology}{label}; |
294
|
|
|
|
|
|
|
|
295
|
|
|
|
|
|
|
# ********************** |
296
|
|
|
|
|
|
|
# *** IMPORTANT STEP *** |
297
|
|
|
|
|
|
|
# ********************** |
298
|
|
|
|
|
|
|
# full_text_search is supposed to be ONLY in text fields, but, for |
299
|
|
|
|
|
|
|
# whatever reaon the binding of parameters e.g, '2 - mild' (starts w/ number) |
300
|
|
|
|
|
|
|
# produce exceptions on SQLite. We'll be parsing them for ALL SEARCHES!!! |
301
|
|
|
|
|
|
|
|
302
|
|
|
|
|
|
|
# NB: Order matters in the changes below |
303
|
204
|
|
|
|
|
709
|
$query =~ s/^\d+\s+\-\s+//; # for ALL SEARCHES!!! |
304
|
204
|
50
|
|
|
|
478
|
$query =~ tr#_,-/# # if $match eq 'full_text_search'; # FTS |
305
|
204
|
|
|
|
|
517
|
$query =~ |
306
|
|
|
|
|
|
|
tr/ //s; # remove duplicated spaces # for ALL SEARCHES!!! |
307
|
|
|
|
|
|
|
|
308
|
|
|
|
|
|
|
# Execute query |
309
|
204
|
|
|
|
|
1494
|
$sth->bind_param( 1, $query ) |
310
|
|
|
|
|
|
|
; # docstore.mik.ua/orelly/linux/dbi/ch05_03.htm |
311
|
204
|
|
|
|
|
2320032
|
$sth->execute(); # eq to $sth->execute($query); |
312
|
|
|
|
|
|
|
|
313
|
|
|
|
|
|
|
# Prune 'hpo' ontology for being printed as HP: |
314
|
204
|
50
|
|
|
|
1348
|
$ontology = 'hp' if $ontology eq 'hpo'; |
315
|
|
|
|
|
|
|
|
316
|
|
|
|
|
|
|
# Process depending on typf of match |
317
|
204
|
50
|
|
|
|
605
|
if ( $match eq 'exact_match' ) { |
318
|
|
|
|
|
|
|
|
319
|
|
|
|
|
|
|
# Parse query |
320
|
204
|
|
|
|
|
2005967
|
while ( my $row = $sth->fetchrow_arrayref ) { |
321
|
168
|
50
|
|
|
|
1667
|
$id = |
322
|
|
|
|
|
|
|
$ontology ne 'ohdsi' |
323
|
|
|
|
|
|
|
? uc($ontology) . ':' . $row->[$id_column] |
324
|
|
|
|
|
|
|
: $row->[3] . ':' . $row->[$id_column]; |
325
|
168
|
|
|
|
|
338
|
$label = $row->[$label_column]; |
326
|
168
|
|
|
|
|
498
|
last; # Note that sometimes we get more than one (they're discarded) |
327
|
|
|
|
|
|
|
} |
328
|
|
|
|
|
|
|
} |
329
|
|
|
|
|
|
|
else { |
330
|
|
|
|
|
|
|
|
331
|
|
|
|
|
|
|
# Parse query w/ sub |
332
|
0
|
|
|
|
|
0
|
( $id, $label ) = text_similarity( |
333
|
|
|
|
|
|
|
{ |
334
|
|
|
|
|
|
|
sth => $sth, |
335
|
|
|
|
|
|
|
query => $query, |
336
|
|
|
|
|
|
|
ontology => $ontology, |
337
|
|
|
|
|
|
|
id_column => $id_column, |
338
|
|
|
|
|
|
|
label_column => $label_column, |
339
|
|
|
|
|
|
|
text_similarity_method => $text_similarity_method, |
340
|
|
|
|
|
|
|
min_text_similarity_score => $min_text_similarity_score |
341
|
|
|
|
|
|
|
} |
342
|
|
|
|
|
|
|
); |
343
|
|
|
|
|
|
|
} |
344
|
|
|
|
|
|
|
|
345
|
|
|
|
|
|
|
# Finish $sth |
346
|
204
|
|
|
|
|
1243
|
$sth->finish(); |
347
|
|
|
|
|
|
|
|
348
|
|
|
|
|
|
|
# We return results |
349
|
204
|
|
|
|
|
2115
|
return ( $id, $label ); |
350
|
|
|
|
|
|
|
} |
351
|
|
|
|
|
|
|
|
352
|
|
|
|
|
|
|
sub text_similarity { |
353
|
|
|
|
|
|
|
|
354
|
0
|
|
|
0
|
0
|
|
my $arg = shift; |
355
|
0
|
|
|
|
|
|
my $sth = $arg->{sth}; |
356
|
0
|
|
|
|
|
|
my $query = $arg->{query}; |
357
|
0
|
|
|
|
|
|
my $ontology = $arg->{ontology}; |
358
|
0
|
|
|
|
|
|
my $id_column = $arg->{id_column}; |
359
|
0
|
|
|
|
|
|
my $label_column = $arg->{label_column}; |
360
|
0
|
|
|
|
|
|
my $min_score = $arg->{min_text_similarity_score}; |
361
|
0
|
|
|
|
|
|
my $text_similarity_method = $arg->{text_similarity_method}; |
362
|
0
|
0
|
0
|
|
|
|
die "--text-similarity-method <$text_similarity_method> not allowed" |
363
|
|
|
|
|
|
|
unless ( $text_similarity_method eq 'dice' |
364
|
|
|
|
|
|
|
|| $text_similarity_method eq 'cosine' ); |
365
|
|
|
|
|
|
|
|
366
|
|
|
|
|
|
|
#say $text_similarity_method; |
367
|
|
|
|
|
|
|
|
368
|
|
|
|
|
|
|
# Create a new Text::Similarity object |
369
|
0
|
|
|
|
|
|
my $ts = Text::Similarity::Overlaps->new(); |
370
|
|
|
|
|
|
|
|
371
|
|
|
|
|
|
|
# Fetch the query results |
372
|
0
|
|
|
|
|
|
my $data; # hashref |
373
|
0
|
|
|
|
|
|
while ( my $row = $sth->fetchrow_arrayref() ) { |
374
|
|
|
|
|
|
|
|
375
|
0
|
|
|
|
|
|
say "---Checking <$row->[$label_column]>" if DEVEL_MODE; |
376
|
|
|
|
|
|
|
|
377
|
|
|
|
|
|
|
# We have a threshold to assign a result as valid |
378
|
0
|
|
|
|
|
|
my ( $score, %scores ) = |
379
|
|
|
|
|
|
|
$ts->getSimilarityStrings( $query, $row->[$label_column] ); |
380
|
|
|
|
|
|
|
|
381
|
|
|
|
|
|
|
# Only load $data if dice >= $min_score; |
382
|
|
|
|
|
|
|
$data->{ $row->[$label_column] } = { |
383
|
|
|
|
|
|
|
id => $ontology ne 'ohdsi' |
384
|
|
|
|
|
|
|
? uc($ontology) . ':' . $row->[$id_column] |
385
|
|
|
|
|
|
|
: $row->[3] . ':' . $row->[$id_column], |
386
|
|
|
|
|
|
|
label => $row->[$label_column], |
387
|
|
|
|
|
|
|
scores => {%scores}, |
388
|
|
|
|
|
|
|
query => $query |
389
|
|
|
|
|
|
|
} |
390
|
0
|
0
|
|
|
|
|
if $scores{$text_similarity_method} >= $min_score; |
|
|
0
|
|
|
|
|
|
391
|
|
|
|
|
|
|
} |
392
|
|
|
|
|
|
|
|
393
|
|
|
|
|
|
|
# Sort the results by similarity score |
394
|
|
|
|
|
|
|
#$Data::Dumper::Sortkeys = 1 ; |
395
|
|
|
|
|
|
|
my @sorted_keys = |
396
|
|
|
|
|
|
|
sort { |
397
|
|
|
|
|
|
|
$data->{$b}{scores}{$text_similarity_method} |
398
|
0
|
|
|
|
|
|
<=> $data->{$a}{scores}{$text_similarity_method} |
399
|
0
|
|
|
|
|
|
} keys %{$data}; |
|
0
|
|
|
|
|
|
|
400
|
|
|
|
|
|
|
|
401
|
0
|
|
|
|
|
|
print Dumper $data if DEVEL_MODE; |
402
|
0
|
0
|
0
|
|
|
|
say "WINNER <$sorted_keys[0]>" if ( $sorted_keys[0] && DEVEL_MODE ); |
403
|
|
|
|
|
|
|
|
404
|
|
|
|
|
|
|
# Return 1st element if present |
405
|
|
|
|
|
|
|
return $sorted_keys[0] |
406
|
|
|
|
|
|
|
? ( $data->{ $sorted_keys[0] }{id}, $data->{ $sorted_keys[0] }{label} ) |
407
|
0
|
0
|
|
|
|
|
: ( undef, undef ); |
408
|
|
|
|
|
|
|
} |
409
|
|
|
|
|
|
|
1; |