qiita-elasticsearch 0.2.2 → 0.3.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.rubocop.yml +4 -1
- data/CHANGELOG.md +3 -0
- data/README.md +13 -7
- data/lib/qiita/elasticsearch/filterable_token.rb +16 -0
- data/lib/qiita/elasticsearch/hierarchal_token.rb +27 -0
- data/lib/qiita/elasticsearch/matchable_token.rb +29 -0
- data/lib/qiita/elasticsearch/nodes/filter_node.rb +6 -27
- data/lib/qiita/elasticsearch/nodes/filterable_node.rb +4 -21
- data/lib/qiita/elasticsearch/nodes/match_node.rb +3 -19
- data/lib/qiita/elasticsearch/nodes/multi_should_node.rb +2 -7
- data/lib/qiita/elasticsearch/nodes/or_separatable_node.rb +3 -19
- data/lib/qiita/elasticsearch/nodes/query_node.rb +3 -12
- data/lib/qiita/elasticsearch/nodes/term_node.rb +2 -59
- data/lib/qiita/elasticsearch/query_builder.rb +11 -8
- data/lib/qiita/elasticsearch/range_token.rb +57 -0
- data/lib/qiita/elasticsearch/token.rb +21 -32
- data/lib/qiita/elasticsearch/tokenizer.rb +45 -3
- data/lib/qiita/elasticsearch/version.rb +1 -1
- metadata +7 -3
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: cc0094615455695aa7148b67052af7b154c5e31e
|
4
|
+
data.tar.gz: 64562e5f4139f14eb727062d7caf39cf706f92eb
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 183fabc234d8480455a5804eb80a8ccca40cc0b0e39fc5f6ee2f75800958875170203063f76303c88ea68e45388f347aa000f14c13d204eb6f2f8317c213b5d7
|
7
|
+
data.tar.gz: 857a39fb54f660ed9a3edf6c74b45212589ba558586bd0966f2b388454099ce5aad322a7b77949d1a4129b10ae6e755eacf377b8c2eecf3e6a0f43a042637ba6
|
data/.rubocop.yml
CHANGED
@@ -11,11 +11,14 @@ Metrics/CyclomaticComplexity:
|
|
11
11
|
Enabled: false
|
12
12
|
|
13
13
|
Metrics/LineLength:
|
14
|
-
|
14
|
+
Enabled: false
|
15
15
|
|
16
16
|
Metrics/MethodLength:
|
17
17
|
Enabled: false
|
18
18
|
|
19
|
+
Metrics/ParameterLists:
|
20
|
+
Enabled: false
|
21
|
+
|
19
22
|
Metrics/PerceivedComplexity:
|
20
23
|
Enabled: false
|
21
24
|
|
data/CHANGELOG.md
CHANGED
data/README.md
CHANGED
@@ -6,7 +6,6 @@ Elasticsearch client helper for Qiita.
|
|
6
6
|
|
7
7
|
```rb
|
8
8
|
query_builder = Qiita::Elasticsearch::QueryBuilder.new
|
9
|
-
#=> #<Qiita::Elasticsearch::QueryBuilder:0x007ff81e67c4d8 @filterable_fields=nil, @hierarchal_fields=nil, @matchable_fields=nil>
|
10
9
|
|
11
10
|
query_builder.build("a")
|
12
11
|
#=> {"match"=>{"_all"=>"a"}}
|
@@ -29,7 +28,6 @@ Pass `:matchable_fields` option to tell matchable field names (default: `_all`).
|
|
29
28
|
|
30
29
|
```rb
|
31
30
|
query_builder = Qiita::Elasticsearch::QueryBuilder.new(matchable_fields: ["body", "title"])
|
32
|
-
#=> #<Qiita::Elasticsearch::QueryBuilder:0x007ff81fa59168 @filterable_fields=nil, @hierarchal_fields=nil, @matchable_fields=["body", "title"], @range_fields=nil>
|
33
31
|
|
34
32
|
query_builder.build("a")
|
35
33
|
#=> {"multi_match"=>{"fields"=>["body", "title"], "query"=>"a"}}
|
@@ -40,7 +38,6 @@ Pass `:filterable_fields` option to enable filtered queries like `tag:Ruby`.
|
|
40
38
|
|
41
39
|
```rb
|
42
40
|
query_builder = Qiita::Elasticsearch::QueryBuilder.new(filterable_fields: ["tag", "title"])
|
43
|
-
#=> #<Qiita::Elasticsearch::QueryBuilder:0x007ff81e2de1f0 @filterable_fields=["tag", "title"], @hierarchal_fields=nil, @matchable_fields=nil, @range_fields=nil>
|
44
41
|
|
45
42
|
query_builder.build("tag:a")
|
46
43
|
#=> {"filtered"=>{"filter"=>{"term"=>{"tag"=>"a"}}}}
|
@@ -53,25 +50,34 @@ query_builder.build("user:a b")
|
|
53
50
|
```
|
54
51
|
|
55
52
|
### hierarchal_fields
|
56
|
-
Pass `:hierarchal_fields`
|
53
|
+
Pass `:hierarchal_fields` option with `:filterable_fields` to enable prefixed filtered queries.
|
57
54
|
With this option, `tag:foo` will hit documents tagged with `foo`, or `foo/...`.
|
58
55
|
|
59
56
|
```rb
|
60
57
|
query_builder = Qiita::Elasticsearch::QueryBuilder.new(filterable_fields: ["tag"], hierarchal_fields: ["tag"])
|
61
|
-
#=> #<Qiita::Elasticsearch::QueryBuilder:0x007fe96d6d5ed0 @filterable_fields=["tag"], @hierarchal_fields=["tag"], @matchable_fields=nil, @range_fields=nil>
|
62
58
|
|
63
59
|
query_builder.build("tag:ruby")
|
64
60
|
#=> {"filtered"=>{"filter"=>{"bool"=>{"should"=>[{"prefix"=>{"tag"=>"ruby/"}}, {"term"=>{"tag"=>"ruby"}}]}}}}
|
65
61
|
```
|
66
62
|
|
67
63
|
### range_fields
|
68
|
-
Pass `:range_fields`
|
64
|
+
Pass `:range_fields` option with `:filterable_fields` to enable range filtered queries.
|
69
65
|
With this option, `created_at:<2015-04-16` will hit documents created before 2015-04-16.
|
70
66
|
|
71
67
|
```rb
|
72
68
|
query_builder = Qiita::Elasticsearch::QueryBuilder.new(filterable_fields: ["created_at"], range_fields: ["created_at"])
|
73
|
-
#=> #<Qiita::Elasticsearch::QueryBuilder:0x007fe96d6d5ed0 @filterable_fields=["created_at"], @hierarchal_fields=nil, @matchable_fields=nil, @range_fields=["created_at"]>
|
74
69
|
|
75
70
|
query_builder.build("created_at:<2015-04-16")
|
76
71
|
#=> {"filtered"=>{"filter"=>{"range"=>{"created_at"=>{"lt"=>"2015-04-16"}}}}}
|
77
72
|
```
|
73
|
+
|
74
|
+
### downcased_fields
|
75
|
+
Pass `:downcased_fields` option with `:filterable_fields` to downcase any terms in the fields.
|
76
|
+
This option is useful when some fields are stored within downcased format on Elasticsearch.
|
77
|
+
|
78
|
+
```rb
|
79
|
+
query_builder = Qiita::Elasticsearch::QueryBuilder.new(downcased_fields: ["tag"], filterable_fields: ["tag"])
|
80
|
+
|
81
|
+
query_builder.build("tag:Ruby")
|
82
|
+
#=> {"filtered"=>{"filter"=>{"term"=>{"tag"=>"ruby"}}}}
|
83
|
+
```
|
@@ -0,0 +1,27 @@
|
|
1
|
+
require "qiita/elasticsearch/token"
|
2
|
+
|
3
|
+
module Qiita
|
4
|
+
module Elasticsearch
|
5
|
+
class HierarchalToken < Token
|
6
|
+
# @return [Hash]
|
7
|
+
def to_hash
|
8
|
+
{
|
9
|
+
"bool" => {
|
10
|
+
"should" => [
|
11
|
+
{
|
12
|
+
"prefix" => {
|
13
|
+
@field_name => proper_cased_term + "/",
|
14
|
+
},
|
15
|
+
},
|
16
|
+
{
|
17
|
+
"term" => {
|
18
|
+
@field_name => proper_cased_term,
|
19
|
+
},
|
20
|
+
},
|
21
|
+
],
|
22
|
+
},
|
23
|
+
}
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
27
|
+
end
|
@@ -0,0 +1,29 @@
|
|
1
|
+
require "qiita/elasticsearch/token"
|
2
|
+
|
3
|
+
module Qiita
|
4
|
+
module Elasticsearch
|
5
|
+
class MatchableToken < Token
|
6
|
+
attr_writer :matchable_fields
|
7
|
+
|
8
|
+
# @return [Hash]
|
9
|
+
def to_hash
|
10
|
+
if @matchable_fields.nil?
|
11
|
+
{
|
12
|
+
quoted? ? "match_phrase" : "match" => {
|
13
|
+
"_all" => @term,
|
14
|
+
}
|
15
|
+
}
|
16
|
+
else
|
17
|
+
hash = {
|
18
|
+
"multi_match" => {
|
19
|
+
"fields" => @matchable_fields,
|
20
|
+
"query" => @term,
|
21
|
+
},
|
22
|
+
}
|
23
|
+
hash["multi_match"]["type"] = "phrase" if quoted?
|
24
|
+
hash
|
25
|
+
end
|
26
|
+
end
|
27
|
+
end
|
28
|
+
end
|
29
|
+
end
|
@@ -6,23 +6,13 @@ module Qiita
|
|
6
6
|
module Nodes
|
7
7
|
class FilterNode
|
8
8
|
# @param [Array<Qiita::Elasticsearch::Tokens>] tokens
|
9
|
-
|
10
|
-
# @param [Array<String>, nil] matchable_fields
|
11
|
-
# @param [Array<String>, nil] range_fields
|
12
|
-
def initialize(tokens, hierarchal_fields: nil, matchable_fields: nil, range_fields: nil)
|
13
|
-
@hierarchal_fields = hierarchal_fields
|
14
|
-
@matchable_fields = matchable_fields
|
15
|
-
@range_fields = range_fields
|
9
|
+
def initialize(tokens)
|
16
10
|
@tokens = tokens
|
17
11
|
end
|
18
12
|
|
19
13
|
def to_hash
|
20
14
|
if must_not_tokens.empty? && must_tokens.length == 1
|
21
|
-
TermNode.new(
|
22
|
-
must_tokens.first,
|
23
|
-
hierarchal_fields: @hierarchal_fields,
|
24
|
-
range_fields: @range_fields,
|
25
|
-
).to_hash
|
15
|
+
TermNode.new(must_tokens.first).to_hash
|
26
16
|
else
|
27
17
|
{
|
28
18
|
"bool" => {
|
@@ -41,17 +31,10 @@ module Qiita
|
|
41
31
|
def must_not_queries
|
42
32
|
must_not_tokens.map do |token|
|
43
33
|
if token.field_name.nil?
|
44
|
-
MatchNode.new(
|
45
|
-
token,
|
46
|
-
matchable_fields: @matchable_fields,
|
47
|
-
)
|
34
|
+
MatchNode.new(token).to_hash
|
48
35
|
else
|
49
|
-
TermNode.new(
|
50
|
-
|
51
|
-
hierarchal_fields: @hierarchal_fields,
|
52
|
-
range_fields: @range_fields,
|
53
|
-
)
|
54
|
-
end.to_hash
|
36
|
+
TermNode.new(token).to_hash
|
37
|
+
end
|
55
38
|
end
|
56
39
|
end
|
57
40
|
|
@@ -61,11 +44,7 @@ module Qiita
|
|
61
44
|
|
62
45
|
def must_queries
|
63
46
|
must_tokens.map do |token|
|
64
|
-
TermNode.new(
|
65
|
-
token,
|
66
|
-
hierarchal_fields: @hierarchal_fields,
|
67
|
-
range_fields: @range_fields,
|
68
|
-
).to_hash
|
47
|
+
TermNode.new(token).to_hash
|
69
48
|
end
|
70
49
|
end
|
71
50
|
|
@@ -6,35 +6,18 @@ module Qiita
|
|
6
6
|
module Nodes
|
7
7
|
class FilterableNode
|
8
8
|
# @param [Array<Qiita::Elasticsearch::Tokens>] tokens
|
9
|
-
|
10
|
-
# @param [Array<String>, nil] matchable_fields
|
11
|
-
# @param [Array<String>, nil] range_fields
|
12
|
-
def initialize(tokens, hierarchal_fields: nil, matchable_fields: nil, range_fields: nil)
|
13
|
-
@hierarchal_fields = hierarchal_fields
|
14
|
-
@matchable_fields = matchable_fields
|
15
|
-
@range_fields = range_fields
|
9
|
+
def initialize(tokens)
|
16
10
|
@tokens = tokens
|
17
11
|
end
|
18
12
|
|
19
13
|
def to_hash
|
20
14
|
if filter_tokens.empty?
|
21
|
-
QueryNode.new(
|
22
|
-
not_filter_tokens,
|
23
|
-
matchable_fields: @matchable_fields,
|
24
|
-
).to_hash
|
15
|
+
QueryNode.new(not_filter_tokens).to_hash
|
25
16
|
else
|
26
17
|
{
|
27
18
|
"filtered" => {
|
28
|
-
"filter" => FilterNode.new(
|
29
|
-
|
30
|
-
hierarchal_fields: @hierarchal_fields,
|
31
|
-
matchable_fields: @matchable_fields,
|
32
|
-
range_fields: @range_fields,
|
33
|
-
).to_hash,
|
34
|
-
"query" => QueryNode.new(
|
35
|
-
not_filter_tokens,
|
36
|
-
matchable_fields: @matchable_fields,
|
37
|
-
).to_hash,
|
19
|
+
"filter" => FilterNode.new(filter_tokens).to_hash,
|
20
|
+
"query" => QueryNode.new(not_filter_tokens).to_hash,
|
38
21
|
}.reject do |key, value|
|
39
22
|
value.empty?
|
40
23
|
end,
|
@@ -3,29 +3,13 @@ module Qiita
|
|
3
3
|
module Nodes
|
4
4
|
class MatchNode
|
5
5
|
# @param [Qiita::Elasticsearch::Token] token
|
6
|
-
|
7
|
-
def initialize(token, matchable_fields: nil)
|
8
|
-
@matchable_fields = matchable_fields
|
6
|
+
def initialize(token)
|
9
7
|
@token = token
|
10
8
|
end
|
11
9
|
|
10
|
+
# @return [Hash]
|
12
11
|
def to_hash
|
13
|
-
|
14
|
-
{
|
15
|
-
@token.quoted? ? "match_phrase" : "match" => {
|
16
|
-
"_all" => @token.term,
|
17
|
-
}
|
18
|
-
}
|
19
|
-
else
|
20
|
-
hash = {
|
21
|
-
"multi_match" => {
|
22
|
-
"fields" => @matchable_fields,
|
23
|
-
"query" => @token.term,
|
24
|
-
},
|
25
|
-
}
|
26
|
-
hash["multi_match"]["type"] = "phrase" if @token.quoted?
|
27
|
-
hash
|
28
|
-
end
|
12
|
+
@token.to_hash
|
29
13
|
end
|
30
14
|
end
|
31
15
|
end
|
@@ -5,9 +5,7 @@ module Qiita
|
|
5
5
|
module Nodes
|
6
6
|
class MultiShouldNode
|
7
7
|
# @param [Array<Qiita::Elasticsearch::Tokens>] tokens
|
8
|
-
|
9
|
-
def initialize(tokens, matchable_fields: nil)
|
10
|
-
@matchable_fields = matchable_fields
|
8
|
+
def initialize(tokens)
|
11
9
|
@tokens = tokens
|
12
10
|
end
|
13
11
|
|
@@ -24,10 +22,7 @@ module Qiita
|
|
24
22
|
# @return [Array<Hash>] Queries to be used as a value of `should` property of bool query.
|
25
23
|
def should_queries
|
26
24
|
@tokens.map do |token|
|
27
|
-
MatchNode.new(
|
28
|
-
token,
|
29
|
-
matchable_fields: @matchable_fields,
|
30
|
-
).to_hash
|
25
|
+
MatchNode.new(token).to_hash
|
31
26
|
end
|
32
27
|
end
|
33
28
|
end
|
@@ -6,13 +6,7 @@ module Qiita
|
|
6
6
|
module Nodes
|
7
7
|
class OrSeparatableNode
|
8
8
|
# @param [Array<Qiita::Elasticsearch::Tokens>] tokens
|
9
|
-
|
10
|
-
# @param [Array<String>, nil] matchable_fields
|
11
|
-
# @param [Array<String>, nil] range_fields
|
12
|
-
def initialize(tokens, hierarchal_fields: nil, matchable_fields: nil, range_fields: nil)
|
13
|
-
@hierarchal_fields = hierarchal_fields
|
14
|
-
@matchable_fields = matchable_fields
|
15
|
-
@range_fields = range_fields
|
9
|
+
def initialize(tokens)
|
16
10
|
@tokens = tokens
|
17
11
|
end
|
18
12
|
|
@@ -21,22 +15,12 @@ module Qiita
|
|
21
15
|
when 0
|
22
16
|
Nodes::NullNode.new.to_hash
|
23
17
|
when 1
|
24
|
-
Nodes::FilterableNode.new(
|
25
|
-
tokens_grouped_by_or_token.first,
|
26
|
-
hierarchal_fields: @hierarchal_fields,
|
27
|
-
matchable_fields: @matchable_fields,
|
28
|
-
range_fields: @range_fields,
|
29
|
-
).to_hash
|
18
|
+
Nodes::FilterableNode.new(tokens_grouped_by_or_token.first).to_hash
|
30
19
|
else
|
31
20
|
{
|
32
21
|
"bool" => {
|
33
22
|
"should" => tokens_grouped_by_or_token.map do |tokens|
|
34
|
-
Nodes::FilterableNode.new(
|
35
|
-
tokens,
|
36
|
-
hierarchal_fields: @hierarchal_fields,
|
37
|
-
matchable_fields: @matchable_fields,
|
38
|
-
range_fields: @range_fields,
|
39
|
-
).to_hash
|
23
|
+
Nodes::FilterableNode.new(tokens).to_hash
|
40
24
|
end,
|
41
25
|
},
|
42
26
|
}
|
@@ -6,10 +6,7 @@ module Qiita
|
|
6
6
|
module Nodes
|
7
7
|
class QueryNode
|
8
8
|
# @param [Array<Qiita::Elasticsearch::Tokens>] tokens
|
9
|
-
|
10
|
-
def initialize(tokens, hierarchal_fields: nil, matchable_fields: nil)
|
11
|
-
@hierarchal_fields = hierarchal_fields
|
12
|
-
@matchable_fields = matchable_fields
|
9
|
+
def initialize(tokens)
|
13
10
|
@tokens = tokens
|
14
11
|
end
|
15
12
|
|
@@ -18,15 +15,9 @@ module Qiita
|
|
18
15
|
when 0
|
19
16
|
{}
|
20
17
|
when 1
|
21
|
-
MatchNode.new(
|
22
|
-
@tokens.first,
|
23
|
-
matchable_fields: @matchable_fields,
|
24
|
-
).to_hash
|
18
|
+
MatchNode.new(@tokens.first).to_hash
|
25
19
|
else
|
26
|
-
MultiShouldNode.new(
|
27
|
-
@tokens,
|
28
|
-
matchable_fields: @matchable_fields,
|
29
|
-
).to_hash
|
20
|
+
MultiShouldNode.new(@tokens).to_hash
|
30
21
|
end
|
31
22
|
end
|
32
23
|
end
|
@@ -2,71 +2,14 @@ module Qiita
|
|
2
2
|
module Elasticsearch
|
3
3
|
module Nodes
|
4
4
|
class TermNode
|
5
|
-
DEFAULT_HIERARCHAL_FIELDS = []
|
6
|
-
DEFAULT_RANGE_FIELDS = []
|
7
|
-
|
8
5
|
# @param [Qiita::Elasticsearch::Token] token
|
9
|
-
|
10
|
-
# @param [Array<String>, nil] range_fields
|
11
|
-
def initialize(token, hierarchal_fields: nil, range_fields: nil)
|
12
|
-
@hierarchal_fields = hierarchal_fields
|
13
|
-
@range_fields = range_fields
|
6
|
+
def initialize(token)
|
14
7
|
@token = token
|
15
8
|
end
|
16
9
|
|
17
10
|
# @return [Hash]
|
18
11
|
def to_hash
|
19
|
-
|
20
|
-
when has_hierarchal_token?
|
21
|
-
{
|
22
|
-
"bool" => {
|
23
|
-
"should" => [
|
24
|
-
{
|
25
|
-
"prefix" => {
|
26
|
-
@token.field_name => @token.downcased_term + "/",
|
27
|
-
},
|
28
|
-
},
|
29
|
-
{
|
30
|
-
"term" => {
|
31
|
-
@token.field_name => @token.downcased_term,
|
32
|
-
},
|
33
|
-
},
|
34
|
-
],
|
35
|
-
},
|
36
|
-
}
|
37
|
-
when has_range_token?
|
38
|
-
{
|
39
|
-
"range" => {
|
40
|
-
@token.field_name => {
|
41
|
-
@token.range_parameter => @token.range_query,
|
42
|
-
},
|
43
|
-
},
|
44
|
-
}
|
45
|
-
else
|
46
|
-
{
|
47
|
-
"term" => {
|
48
|
-
@token.field_name => @token.downcased_term,
|
49
|
-
},
|
50
|
-
}
|
51
|
-
end
|
52
|
-
end
|
53
|
-
|
54
|
-
private
|
55
|
-
|
56
|
-
def has_hierarchal_token?
|
57
|
-
hierarchal_fields.include?(@token.field_name)
|
58
|
-
end
|
59
|
-
|
60
|
-
def hierarchal_fields
|
61
|
-
@hierarchal_fields || DEFAULT_HIERARCHAL_FIELDS
|
62
|
-
end
|
63
|
-
|
64
|
-
def has_range_token?
|
65
|
-
range_fields.include?(@token.field_name) && @token.range_parameter
|
66
|
-
end
|
67
|
-
|
68
|
-
def range_fields
|
69
|
-
@range_fields || DEFAULT_RANGE_FIELDS
|
12
|
+
@token.to_hash
|
70
13
|
end
|
71
14
|
end
|
72
15
|
end
|
@@ -5,11 +5,13 @@ require "qiita/elasticsearch/tokenizer"
|
|
5
5
|
module Qiita
|
6
6
|
module Elasticsearch
|
7
7
|
class QueryBuilder
|
8
|
+
# @param [Array<String>, nil] downcased_fields
|
8
9
|
# @param [Array<String>, nil] filterable_fields
|
9
10
|
# @param [Array<String>, nil] hierarchal_fields
|
10
11
|
# @param [Array<String>, nil] matchable_fields
|
11
12
|
# @param [Array<String>, nil] range_fields
|
12
|
-
def initialize(hierarchal_fields: nil, filterable_fields: nil, matchable_fields: nil, range_fields: nil)
|
13
|
+
def initialize(downcased_fields: nil, hierarchal_fields: nil, filterable_fields: nil, matchable_fields: nil, range_fields: nil)
|
14
|
+
@downcased_fields = downcased_fields
|
13
15
|
@hierarchal_fields = hierarchal_fields
|
14
16
|
@filterable_fields = filterable_fields
|
15
17
|
@matchable_fields = matchable_fields
|
@@ -23,19 +25,20 @@ module Qiita
|
|
23
25
|
if tokens.size.zero?
|
24
26
|
Nodes::NullNode.new.to_hash
|
25
27
|
else
|
26
|
-
Nodes::OrSeparatableNode.new(
|
27
|
-
tokens,
|
28
|
-
hierarchal_fields: @hierarchal_fields,
|
29
|
-
matchable_fields: @matchable_fields,
|
30
|
-
range_fields: @range_fields,
|
31
|
-
).to_hash
|
28
|
+
Nodes::OrSeparatableNode.new(tokens).to_hash
|
32
29
|
end
|
33
30
|
end
|
34
31
|
|
35
32
|
private
|
36
33
|
|
37
34
|
def tokenizer
|
38
|
-
@tokenizer ||= Tokenizer.new(
|
35
|
+
@tokenizer ||= Tokenizer.new(
|
36
|
+
downcased_fields: @downcased_fields,
|
37
|
+
hierarchal_fields: @hierarchal_fields,
|
38
|
+
filterable_fields: @filterable_fields,
|
39
|
+
matchable_fields: @matchable_fields,
|
40
|
+
range_fields: @range_fields,
|
41
|
+
)
|
39
42
|
end
|
40
43
|
end
|
41
44
|
end
|
@@ -0,0 +1,57 @@
|
|
1
|
+
require "qiita/elasticsearch/token"
|
2
|
+
|
3
|
+
module Qiita
|
4
|
+
module Elasticsearch
|
5
|
+
class RangeToken < Token
|
6
|
+
RANGE_TERM_REGEXP = /\A(?<operand>\<|\<=|\>|\>=)(?<query>.*)\z/
|
7
|
+
|
8
|
+
# @return [Hash]
|
9
|
+
def to_hash
|
10
|
+
if range_parameter
|
11
|
+
{
|
12
|
+
"range" => {
|
13
|
+
@field_name => {
|
14
|
+
range_parameter => range_query,
|
15
|
+
},
|
16
|
+
},
|
17
|
+
}
|
18
|
+
else
|
19
|
+
{
|
20
|
+
"term" => {
|
21
|
+
@field_name => proper_cased_term,
|
22
|
+
},
|
23
|
+
}
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
private
|
28
|
+
|
29
|
+
# @return [String, nil]
|
30
|
+
# @example Suppose @term is "created_at:>=2015-04-16"
|
31
|
+
# range_parameter #=> "gte"
|
32
|
+
def range_parameter
|
33
|
+
range_match[:operand] ? operand_map[range_match[:operand]] : nil
|
34
|
+
end
|
35
|
+
|
36
|
+
# @return [String, nil]
|
37
|
+
# @example Suppose @term is "created_at:>=2015-04-16"
|
38
|
+
# range_query #=> "2015-04-16"
|
39
|
+
def range_query
|
40
|
+
range_match[:query]
|
41
|
+
end
|
42
|
+
|
43
|
+
def range_match
|
44
|
+
@range_match ||= RANGE_TERM_REGEXP.match(@term) || {}
|
45
|
+
end
|
46
|
+
|
47
|
+
def operand_map
|
48
|
+
{
|
49
|
+
">" => "gt",
|
50
|
+
">=" => "gte",
|
51
|
+
"<" => "lt",
|
52
|
+
"<=" => "lte",
|
53
|
+
}
|
54
|
+
end
|
55
|
+
end
|
56
|
+
end
|
57
|
+
end
|
@@ -1,11 +1,11 @@
|
|
1
1
|
module Qiita
|
2
2
|
module Elasticsearch
|
3
3
|
class Token
|
4
|
-
RANGE_TERM_REGEXP = /\A(?<operand>\<|\<=|\>|\>=)(?<query>.*)\z/
|
5
|
-
|
6
4
|
attr_reader :field_name, :term
|
7
5
|
|
8
|
-
|
6
|
+
# @param [true, false] downcased
|
7
|
+
def initialize(downcased: nil, field_name: nil, minus: nil, quoted: nil, term: nil, token_string: nil)
|
8
|
+
@downcased = downcased
|
9
9
|
@field_name = field_name
|
10
10
|
@minus = minus
|
11
11
|
@quoted = quoted
|
@@ -13,10 +13,19 @@ module Qiita
|
|
13
13
|
@token_string = token_string
|
14
14
|
end
|
15
15
|
|
16
|
+
# @return [true, false] True if its term must be treated with downcased
|
17
|
+
def downcased?
|
18
|
+
!!@downcased
|
19
|
+
end
|
20
|
+
|
16
21
|
def downcased_term
|
17
22
|
@downcased_term ||= term.downcase
|
18
23
|
end
|
19
24
|
|
25
|
+
def to_hash
|
26
|
+
fail NotImplementedError
|
27
|
+
end
|
28
|
+
|
20
29
|
def filter?
|
21
30
|
!field_name.nil? || negative?
|
22
31
|
end
|
@@ -50,6 +59,15 @@ module Qiita
|
|
50
59
|
@minus.nil?
|
51
60
|
end
|
52
61
|
|
62
|
+
# @return [String] Downcased or not-downcased term
|
63
|
+
def proper_cased_term
|
64
|
+
if downcased?
|
65
|
+
downcased_term
|
66
|
+
else
|
67
|
+
term
|
68
|
+
end
|
69
|
+
end
|
70
|
+
|
53
71
|
# @return [true, false] True if this token is for phrase matching
|
54
72
|
# @note `Express OR "Ruby on Rails"`
|
55
73
|
# ^^^^^^^^^^^^^^^
|
@@ -57,35 +75,6 @@ module Qiita
|
|
57
75
|
def quoted?
|
58
76
|
!!@quoted
|
59
77
|
end
|
60
|
-
|
61
|
-
# @return [String, nil]
|
62
|
-
# @example Suppose @term is "created_at:>=2015-04-16"
|
63
|
-
# range_parameter #=> "gte"
|
64
|
-
def range_parameter
|
65
|
-
range_match[:operand] ? operand_map[range_match[:operand]] : nil
|
66
|
-
end
|
67
|
-
|
68
|
-
# @return [String, nil]
|
69
|
-
# @example Suppose @term is "created_at:>=2015-04-16"
|
70
|
-
# range_query #=> "2015-04-16"
|
71
|
-
def range_query
|
72
|
-
range_match[:query]
|
73
|
-
end
|
74
|
-
|
75
|
-
private
|
76
|
-
|
77
|
-
def range_match
|
78
|
-
@range_match ||= RANGE_TERM_REGEXP.match(@term) || {}
|
79
|
-
end
|
80
|
-
|
81
|
-
def operand_map
|
82
|
-
{
|
83
|
-
">" => "gt",
|
84
|
-
">=" => "gte",
|
85
|
-
"<" => "lt",
|
86
|
-
"<=" => "lte",
|
87
|
-
}
|
88
|
-
end
|
89
78
|
end
|
90
79
|
end
|
91
80
|
end
|
@@ -1,9 +1,15 @@
|
|
1
|
-
require "qiita/elasticsearch/
|
1
|
+
require "qiita/elasticsearch/filterable_token"
|
2
|
+
require "qiita/elasticsearch/hierarchal_token"
|
3
|
+
require "qiita/elasticsearch/matchable_token"
|
4
|
+
require "qiita/elasticsearch/range_token"
|
2
5
|
|
3
6
|
module Qiita
|
4
7
|
module Elasticsearch
|
5
8
|
class Tokenizer
|
9
|
+
DEFAULT_DOWNCASED_FIELDS = []
|
6
10
|
DEFAULT_FILTERABLE_FIELDS = []
|
11
|
+
DEFAULT_HIERARCHAL_FIELDS = []
|
12
|
+
DEFAULT_RANGE_FIELDS = []
|
7
13
|
|
8
14
|
TOKEN_PATTERN = /
|
9
15
|
(?<token_string>
|
@@ -17,9 +23,17 @@ module Qiita
|
|
17
23
|
)
|
18
24
|
/x
|
19
25
|
|
26
|
+
# @param [Array<String>, nil] downcased_fields
|
20
27
|
# @param [Array<String>, nil] filterable_fields
|
21
|
-
|
28
|
+
# @param [Array<String>, nil] hierarchal_fields
|
29
|
+
# @param [Array<String>, nil] matchable_fields
|
30
|
+
# @param [Array<String>, nil] range_fields
|
31
|
+
def initialize(downcased_fields: nil, filterable_fields: nil, hierarchal_fields: nil, matchable_fields: nil, range_fields: nil)
|
32
|
+
@downcased_fields = downcased_fields
|
22
33
|
@filterable_fields = filterable_fields
|
34
|
+
@hierarchal_fields = hierarchal_fields
|
35
|
+
@matchable_fields = matchable_fields
|
36
|
+
@range_fields = range_fields
|
23
37
|
end
|
24
38
|
|
25
39
|
# @param [String] query_string Raw query string
|
@@ -31,21 +45,49 @@ module Qiita
|
|
31
45
|
term = "#{field_name}:#{term}"
|
32
46
|
field_name = nil
|
33
47
|
end
|
34
|
-
|
48
|
+
token = token_class(field_name).new(
|
49
|
+
downcased: downcased_fields.include?(field_name),
|
35
50
|
field_name: field_name,
|
36
51
|
minus: minus,
|
37
52
|
quoted: !quoted_term.nil?,
|
38
53
|
term: term,
|
39
54
|
token_string: token_string,
|
40
55
|
)
|
56
|
+
token.matchable_fields = @matchable_fields if token.is_a?(MatchableToken)
|
57
|
+
token
|
41
58
|
end
|
42
59
|
end
|
43
60
|
|
44
61
|
private
|
45
62
|
|
63
|
+
def downcased_fields
|
64
|
+
@downcased_fields || DEFAULT_DOWNCASED_FIELDS
|
65
|
+
end
|
66
|
+
|
46
67
|
def filterable_fields
|
47
68
|
@filterable_fields || DEFAULT_FILTERABLE_FIELDS
|
48
69
|
end
|
70
|
+
|
71
|
+
def hierarchal_fields
|
72
|
+
@hierarchal_fields || DEFAULT_HIERARCHAL_FIELDS
|
73
|
+
end
|
74
|
+
|
75
|
+
def range_fields
|
76
|
+
@range_fields || DEFAULT_RANGE_FIELDS
|
77
|
+
end
|
78
|
+
|
79
|
+
def token_class(field_name)
|
80
|
+
case
|
81
|
+
when range_fields.include?(field_name)
|
82
|
+
RangeToken
|
83
|
+
when hierarchal_fields.include?(field_name)
|
84
|
+
HierarchalToken
|
85
|
+
when filterable_fields.include?(field_name)
|
86
|
+
FilterableToken
|
87
|
+
else
|
88
|
+
MatchableToken
|
89
|
+
end
|
90
|
+
end
|
49
91
|
end
|
50
92
|
end
|
51
93
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: qiita-elasticsearch
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.3.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Ryo Nakamura
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2015-04-
|
11
|
+
date: 2015-04-22 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -97,6 +97,9 @@ files:
|
|
97
97
|
- README.md
|
98
98
|
- Rakefile
|
99
99
|
- lib/qiita/elasticsearch.rb
|
100
|
+
- lib/qiita/elasticsearch/filterable_token.rb
|
101
|
+
- lib/qiita/elasticsearch/hierarchal_token.rb
|
102
|
+
- lib/qiita/elasticsearch/matchable_token.rb
|
100
103
|
- lib/qiita/elasticsearch/nodes/filter_node.rb
|
101
104
|
- lib/qiita/elasticsearch/nodes/filterable_node.rb
|
102
105
|
- lib/qiita/elasticsearch/nodes/match_node.rb
|
@@ -106,6 +109,7 @@ files:
|
|
106
109
|
- lib/qiita/elasticsearch/nodes/query_node.rb
|
107
110
|
- lib/qiita/elasticsearch/nodes/term_node.rb
|
108
111
|
- lib/qiita/elasticsearch/query_builder.rb
|
112
|
+
- lib/qiita/elasticsearch/range_token.rb
|
109
113
|
- lib/qiita/elasticsearch/token.rb
|
110
114
|
- lib/qiita/elasticsearch/tokenizer.rb
|
111
115
|
- lib/qiita/elasticsearch/version.rb
|
@@ -130,7 +134,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
130
134
|
version: '0'
|
131
135
|
requirements: []
|
132
136
|
rubyforge_project:
|
133
|
-
rubygems_version: 2.
|
137
|
+
rubygems_version: 2.4.5
|
134
138
|
signing_key:
|
135
139
|
specification_version: 4
|
136
140
|
summary: Elasticsearch client helper for Qiita.
|