pg_fulltext 0.2.0 → 0.2.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/README.md +11 -7
- data/lib/pg_fulltext/query.rb +23 -43
- metadata +15 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 44e2749ca060cca87ad4d046c44544150a6ec7d68a4e0f834a2f51284c86107d
|
|
4
|
+
data.tar.gz: fe18a7c9bbe15f6914aa0ae659f3728330bbace4dbae5a68850f80636234090f
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: c239868d0fd3cf6823a5ffc95820fe2388c23365ce18f1da1441a6e78a55d74ce2b9fb94147b2ff4b836c38ddca15a41a845d958adaa292ab09af9fe34e4d29b
|
|
7
|
+
data.tar.gz: edda1972f580fe94582c8256cc17cb701cab36673e415e665e908146070d3f6d35a38075477e21cb8198da1c4512a346e30811d0de3344e162b95951cfbaf1a3
|
data/README.md
CHANGED
|
@@ -41,11 +41,12 @@ configured per the following options:
|
|
|
41
41
|
|
|
42
42
|
| Option | Default | Description |
|
|
43
43
|
| ---------------- | :------ | :----------- |
|
|
44
|
-
| `tsvector_column` | `tsv` | If you have a different column containing your tsvector, specify it here |
|
|
45
|
-
| `search_type` | `nil` | Your PostgreSQL probably defaults to `'english'`, but set this to match the tsvector you've generated |
|
|
46
|
-
| `order` | `true` | Whether or not the `order` method should be applied against the generated `rank` for the fulltext query. If you just care about returning matches and not their respective rank, set this to `false
|
|
47
|
-
| `
|
|
48
|
-
| `
|
|
44
|
+
| `tsvector_column` | `tsv` | If you have a different column containing your tsvector, specify it here. |
|
|
45
|
+
| `search_type` | `nil` | Your PostgreSQL probably defaults to `'english'`, but set this to match the tsvector you've generated. |
|
|
46
|
+
| `order` | `true` | Whether or not the `order` method should be applied against the generated `rank` for the fulltext query. If you just care about returning matches and not their respective rank, set this to `false`. |
|
|
47
|
+
| `prefix` | `true` | Default search will match partial words as well as whole words. Set this to `false` if only whole words should be matched. |
|
|
48
|
+
| `reorder` | `false` | If you already have `order` set on this relation, it will take precedence over the fulltext `rank`. `reorder` will call clear, effectively clearing the existing order and applying `rank`. |
|
|
49
|
+
| `any_word` | `false` | Default search uses the `&` operator, ensuring that all terms are matched in the query. If you want to match _any_ term in the query, set this to `true`. |
|
|
49
50
|
| `ignore_accents` | `false` | By default, search queries with accents will be sent through as-is. Setting this to `true` will `unaccent()` the query, which helps match `tsv` columns that have also been unaccented. Alternatively, you can have your `tsv` column be a combination of both, and this option will be unnecesary. Requires the `unaccent` Postgres extension. |
|
|
50
51
|
|
|
51
52
|
## Standalone Configuration
|
|
@@ -56,10 +57,13 @@ string for you to use as you wish.
|
|
|
56
57
|
Something this should do the trick:
|
|
57
58
|
|
|
58
59
|
```ruby
|
|
60
|
+
db = PG.connect(dbname: 'mydb')
|
|
59
61
|
search_string = 'foo bar "include this phrase" !butnotthis !"and and also not this"'
|
|
60
|
-
|
|
62
|
+
tsv_query = db.escape_string(PgFulltext::Query.to_tsquery_string(search_string))
|
|
63
|
+
sql = <<~SQL
|
|
61
64
|
SELECT *
|
|
62
65
|
FROM my_model
|
|
63
|
-
WHERE tsv @@
|
|
66
|
+
WHERE tsv @@ to_tsquery('portuguese', '#{tsv_query}')
|
|
64
67
|
SQL
|
|
68
|
+
db.exec(sql)
|
|
65
69
|
```
|
data/lib/pg_fulltext/query.rb
CHANGED
|
@@ -1,52 +1,34 @@
|
|
|
1
|
+
require 'rltk/lexer'
|
|
2
|
+
|
|
1
3
|
module PgFulltext
|
|
2
4
|
module Query
|
|
3
|
-
|
|
4
|
-
|
|
5
|
-
|
|
6
|
-
query = normalize_query(query)
|
|
7
|
-
|
|
8
|
-
# Collect terms
|
|
9
|
-
terms = []
|
|
5
|
+
class Lexer < RLTK::Lexer
|
|
6
|
+
rule(/!+/)
|
|
7
|
+
rule(/"+/)
|
|
10
8
|
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
negate_phrase = false
|
|
9
|
+
rule(/(!?)[\p{L}!]+/) { |v| [:WORD, v] }
|
|
10
|
+
rule(/"[\p{L}\s!]+"/) { |v| [:PHRASE, v[1..-2]] }
|
|
11
|
+
rule(/!"[\p{L}\s!]+"/) { |v| [:NOT_PHRASE, v[2..-2]] }
|
|
15
12
|
|
|
16
|
-
|
|
17
|
-
|
|
13
|
+
rule(/\s+/)
|
|
14
|
+
rule(/[^\p{L}^\s^"]+/)
|
|
15
|
+
end
|
|
18
16
|
|
|
19
|
-
|
|
20
|
-
|
|
17
|
+
def self.to_tsquery_string(query, prefix: true, operator: '&')
|
|
18
|
+
query = normalize_query(query)
|
|
21
19
|
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
|
|
31
|
-
elsif phrase_terms.length > 0
|
|
32
|
-
if term.end_with?('"')
|
|
33
|
-
phrase_terms << format_term(term[0..-2], prefix: prefix)
|
|
34
|
-
terms << "#{'!' if negate_phrase}(#{reject_falsy(phrase_terms).join(' <-> ')})"
|
|
35
|
-
phrase_terms = []
|
|
36
|
-
negate_phrase = false
|
|
37
|
-
else
|
|
38
|
-
phrase_terms << format_term(term, prefix: prefix)
|
|
39
|
-
end
|
|
40
|
-
else
|
|
41
|
-
terms << format_term(term, prefix: prefix)
|
|
42
|
-
end
|
|
20
|
+
terms = []
|
|
21
|
+
Lexer.lex(query).each do |token|
|
|
22
|
+
if token.type == :WORD
|
|
23
|
+
terms << format_term(token.value, prefix: prefix)
|
|
24
|
+
elsif %i[PHRASE NOT_PHRASE].include?(token.type)
|
|
25
|
+
phrase_terms = Lexer.lex(token.value).map do |phrase_term|
|
|
26
|
+
phrase_term.value.nil? ? nil : format_term(phrase_term.value, prefix: prefix)
|
|
27
|
+
end.compact
|
|
28
|
+
terms << "#{'!' if token.type == :NOT_PHRASE}(#{phrase_terms.join(' <-> ')})"
|
|
43
29
|
end
|
|
44
|
-
else
|
|
45
|
-
query.gsub! /["]/, ''
|
|
46
|
-
terms = reject_falsy(query.split(' ').map { |v| format_term(v, prefix: prefix) })
|
|
47
30
|
end
|
|
48
31
|
|
|
49
|
-
# Join terms with operator
|
|
50
32
|
terms.join(" #{operator} ")
|
|
51
33
|
end
|
|
52
34
|
|
|
@@ -55,9 +37,7 @@ module PgFulltext
|
|
|
55
37
|
def self.normalize_query(query)
|
|
56
38
|
query
|
|
57
39
|
.gsub(/[.,]/, ' ') # Replace all periods and commas with spaces (reasonable delimiters)
|
|
58
|
-
.gsub(/[^\s\p{L}"!]/, '') # Remove all non-unicode,
|
|
59
|
-
.gsub(/"+/, '"') # Replace Repeat quotes with single double-quote
|
|
60
|
-
.gsub(/!+/, '!') # Replace Repeat bangs with single bang
|
|
40
|
+
.gsub(/[^\s\p{L}"!]/, '') # Remove all non-unicode, whitespace, quotes ("), and bangs (!)
|
|
61
41
|
.gsub(/\s+/, ' ') # Replace repeat whitespace occurrences with single spaces
|
|
62
42
|
.strip # Strip space from beginning and end of line
|
|
63
43
|
end
|
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: pg_fulltext
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.2.
|
|
4
|
+
version: 0.2.1
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Adam Robertson
|
|
@@ -38,6 +38,20 @@ dependencies:
|
|
|
38
38
|
- - ">="
|
|
39
39
|
- !ruby/object:Gem::Version
|
|
40
40
|
version: '5.0'
|
|
41
|
+
- !ruby/object:Gem::Dependency
|
|
42
|
+
name: rltk
|
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
|
44
|
+
requirements:
|
|
45
|
+
- - ">="
|
|
46
|
+
- !ruby/object:Gem::Version
|
|
47
|
+
version: 3.0.1
|
|
48
|
+
type: :runtime
|
|
49
|
+
prerelease: false
|
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
|
51
|
+
requirements:
|
|
52
|
+
- - ">="
|
|
53
|
+
- !ruby/object:Gem::Version
|
|
54
|
+
version: 3.0.1
|
|
41
55
|
- !ruby/object:Gem::Dependency
|
|
42
56
|
name: rspec
|
|
43
57
|
requirement: !ruby/object:Gem::Requirement
|