Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Use match instead of match_phrase query for autocomplete #1432

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion query/autocomplete.js
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ var views = {
admin_multi_match_first: require('./view/admin_multi_match_first'),
admin_multi_match_last: require('./view/admin_multi_match_last'),
phrase_first_tokens_only: require('./view/phrase_first_tokens_only'),
match_first_tokens_only: require('./view/match_first_tokens_only'),
boost_exact_matches: require('./view/boost_exact_matches'),
max_character_count_layer_filter: require('./view/max_character_count_layer_filter'),
focus_point_filter: require('./view/focus_point_distance_filter')
Expand All @@ -40,7 +41,7 @@ adminFields = adminFields.concat(['add_name_to_multimatch']);
var query = new peliasQuery.layout.FilteredBooleanQuery();

// mandatory matches
query.score( views.phrase_first_tokens_only, 'must' );
query.score( views.match_first_tokens_only, 'must' );
query.score( views.ngrams_last_token_only_multi( adminFields ), 'must' );

// admin components
Expand All @@ -54,6 +55,7 @@ query.score( peliasQuery.view.address('cross_street') );
query.score( peliasQuery.view.address('postcode') );

// scoring boost
query.score( views.phrase_first_tokens_only, 'should' );
query.score( peliasQuery.view.focus( views.ngrams_strict ) );
query.score( peliasQuery.view.popularity( peliasQuery.view.leaf.match_all ) );
query.score( peliasQuery.view.population( peliasQuery.view.leaf.match_all ) );
Expand Down
1 change: 1 addition & 0 deletions query/autocomplete_defaults.js
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ module.exports = _.merge({}, peliasQuery.defaults, {
'ngram:field': 'name.default',
'ngram:boost': 100,
'ngram:cutoff_frequency': 0.01,
'ngram:minimum_should_match': '1<-1 3<-25%',
Copy link
Member Author

@orangejulius orangejulius Aug 13, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This setting was used just because it's what we've used before, we can and should potentially reconsider if its the best option. Also, as we mentioned, it might make sense to use different values in different conditions, for example if the parser believes its a venue vs autocomplete query.

The explanation of what it does and why is contained in #1333


'phrase:analyzer': 'peliasQuery',
'phrase:field': 'phrase.default',
Expand Down
28 changes: 28 additions & 0 deletions query/view/match_first_tokens_only.js
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
const peliasQuery = require('pelias-query');

/**
Phrase view which trims the 'input:name' and uses ALL BUT the last token.
eg. if the input was "100 foo str", then 'input:name' would only be '100 foo'
note: it is assumed that the rest of the input is matched using another view.
**/

module.exports = function( vs ){
const view_name = 'match_first_tokens_only';

// get a copy of the *complete* tokens produced from the input:name
const tokens = vs.var('input:name:tokens_complete').get();

// no valid tokens to use, fail now, don't render this view.
if( !tokens || tokens.length < 1 ){ return null; }

// set the 'input' variable to all but the last token
vs.var(`match:${view_name}:input`).set( tokens.join(' ') );
vs.var(`match:${view_name}:field`).set(vs.var('phrase:field').get());

vs.var(`match:${view_name}:analyzer`).set(vs.var('phrase:analyzer').get());
vs.var(`match:${view_name}:boost`).set(vs.var('phrase:boost').get());
vs.var(`match:${view_name}:minimum_should_match`).set(vs.var('ngram:minimum_should_match').get());

return peliasQuery.view.leaf.match(view_name)( vs );
};
15 changes: 12 additions & 3 deletions test/unit/fixture/autocomplete_custom_boosts.json
Original file line number Diff line number Diff line change
Expand Up @@ -5,18 +5,27 @@
"bool": {
"must": [
{
"match_phrase": {
"match": {
"phrase.default": {
"analyzer": "peliasQuery",
"boost": 1,
"slop": 3,
"query": "foo"
"query": "foo",
"minimum_should_match": "1<-1 3<-25%"
}
}
}
],
"should": [
{
"match_phrase": {
"phrase.default": {
"analyzer": "peliasQuery",
"boost": 1,
"slop": 3,
"query": "foo"
}
}
}, {
"function_score": {
"query": {
"match_all": {}
Expand Down
13 changes: 11 additions & 2 deletions test/unit/fixture/autocomplete_linguistic_final_token.js
Original file line number Diff line number Diff line change
Expand Up @@ -2,16 +2,25 @@ module.exports = {
'query': {
'bool': {
'must': [{
'match_phrase': {
'match': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'minimum_should_match': '1<-1 3<-25%',
'query': 'one'
}
}
}],
'should':[{
'match_phrase': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'query': 'one'
}
}
}, {
'function_score': {
'query': {
'match_all': {}
Expand Down
17 changes: 13 additions & 4 deletions test/unit/fixture/autocomplete_linguistic_multiple_tokens.js
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ module.exports = {
'query': {
'bool': {
'must': [{
'match_phrase': {
'match': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'minimum_should_match': '1<-1 3<-25%',
'query': 'one two'
}
}
Expand Down Expand Up @@ -37,9 +37,18 @@ module.exports = {
}
}
}],
'should':[
'should':[{
'match_phrase': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'query': 'one two'
}
}
},
{
'function_score': {
'function_score': {
'query': {
'match_all': {}
},
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ module.exports = {
'query': {
'bool': {
'must': [{
'match_phrase': {
'match': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'minimum_should_match': '1<-1 3<-25%',
'query': '1 2'
}
}
Expand All @@ -25,7 +25,16 @@ module.exports = {
}
}
}],
'should': [
'should': [{
'match_phrase': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'query': '1 2'
}
}
},
{
'function_score': {
'query': {
Expand Down
14 changes: 12 additions & 2 deletions test/unit/fixture/autocomplete_linguistic_with_admin.js
Original file line number Diff line number Diff line change
Expand Up @@ -3,11 +3,11 @@ module.exports = {
'bool': {
'must': [
{
'match_phrase': {
'match': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'minimum_should_match': '1<-1 3<-25%',
'query': 'one two'
}
}
Expand Down Expand Up @@ -36,6 +36,16 @@ module.exports = {
}
],
'should': [
{
'match_phrase': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'query': 'one two'
}
}
},
{
'function_score': {
'query': {
Expand Down
16 changes: 12 additions & 4 deletions test/unit/fixture/autocomplete_single_character_street.js
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ module.exports = {
'query': {
'bool': {
'must': [{
'match_phrase': {
'match': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'minimum_should_match': '1<-1 3<-25%',
'query': 'k road'
}
}
Expand All @@ -32,8 +32,7 @@ module.exports = {
'type': 'cross_fields'
}
}],
'should':[
{
'should':[{
'match': {
'address_parts.street': {
'query': 'k road',
Expand All @@ -44,6 +43,15 @@ module.exports = {
}
},
{
'match_phrase': {
'phrase.default': {
'analyzer': 'peliasQuery',
'boost': 1,
'slop': 3,
'query': 'k road'
}
}
}, {
'function_score': {
'query': {
'match_all': {}
Expand Down
22 changes: 15 additions & 7 deletions test/unit/query/autocomplete_token_matching_permutations.js
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ const defaults = new peliasQuery.Vars( require('../../../query/autocomplete_defa
const views = {
ngrams_last_token_only: require('../../../query/view/ngrams_last_token_only'),
ngrams_last_token_only_multi: require('../../../query/view/ngrams_last_token_only_multi')(adminFields),
match_first_tokens_only: require('../../../query/view/match_first_tokens_only'),
phrase_first_tokens_only: require('../../../query/view/phrase_first_tokens_only'),
};

Expand Down Expand Up @@ -84,8 +85,9 @@ module.exports.tests.single_token = function(test, common) {
var vs = vars( clean );

assert( t, generate( clean ), {
must: [ views.phrase_first_tokens_only( vs ) ],
must: [ views.match_first_tokens_only( vs ) ],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand Down Expand Up @@ -124,8 +126,9 @@ module.exports.tests.single_token = function(test, common) {
var vs = vars( clean );

assert( t, generate( clean ), {
must: [ views.phrase_first_tokens_only( vs ) ],
must: [ views.match_first_tokens_only( vs ) ],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand Down Expand Up @@ -164,8 +167,9 @@ module.exports.tests.single_token = function(test, common) {
var vs = vars( clean );

assert( t, generate( clean ), {
must: [ views.phrase_first_tokens_only( vs ) ],
must: [ views.match_first_tokens_only( vs ) ],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand All @@ -189,10 +193,11 @@ module.exports.tests.multiple_tokens = function(test, common) {

assert( t, generate( clean ), {
must: [
views.phrase_first_tokens_only( vs ),
views.match_first_tokens_only( vs ),
views.ngrams_last_token_only_multi( vs )
],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand All @@ -212,9 +217,10 @@ module.exports.tests.multiple_tokens = function(test, common) {

assert( t, generate( clean ), {
must: [
views.phrase_first_tokens_only( vs )
views.match_first_tokens_only( vs )
],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand All @@ -236,10 +242,11 @@ module.exports.tests.multiple_tokens = function(test, common) {

assert( t, generate( clean ), {
must: [
views.phrase_first_tokens_only( vs ),
views.match_first_tokens_only( vs ),
views.ngrams_last_token_only_multi( vs )
],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand All @@ -259,9 +266,10 @@ module.exports.tests.multiple_tokens = function(test, common) {

assert( t, generate( clean ), {
must: [
views.phrase_first_tokens_only( vs )
views.match_first_tokens_only( vs )
],
should: [
views.phrase_first_tokens_only( vs ),
peliasQuery.view.popularity( peliasQuery.view.leaf.match_all )( vs ),
peliasQuery.view.population( peliasQuery.view.leaf.match_all )( vs )
]
Expand Down