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

Add option for fieldnorm #1215

Merged
merged 3 commits into from
Apr 8, 2022
Merged
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
42 changes: 42 additions & 0 deletions quickwit-doc-mapper/src/default_doc_mapper/field_mapping_entry.rs
Original file line number Diff line number Diff line change
Expand Up @@ -510,6 +510,8 @@ struct FieldMappingEntryForSerialization {
#[serde(skip_serializing_if = "Option::is_none")]
indexed: Option<bool>,
#[serde(skip_serializing_if = "Option::is_none")]
fieldnorms: Option<bool>,
#[serde(skip_serializing_if = "Option::is_none")]
tokenizer: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
record: Option<IndexRecordOption>,
Expand Down Expand Up @@ -551,6 +553,7 @@ impl From<FieldMappingEntry> for FieldMappingEntryForSerialization {
let type_with_cardinality = value.mapping_type.type_with_cardinality();
let mut fast = false;
let mut indexed = None;
let mut fieldnorms = None;
let mut record = None;
let mut stored = false;
let mut tokenizer: Option<String> = None;
Expand All @@ -560,8 +563,11 @@ impl From<FieldMappingEntry> for FieldMappingEntryForSerialization {
if let Some(indexing_options) = text_options.get_indexing_options() {
tokenizer = Some(indexing_options.tokenizer().to_owned());
record = Some(indexing_options.index_option());
indexed = Some(true);
fieldnorms = Some(indexing_options.fieldnorms());
} else {
indexed = Some(false);
fieldnorms = Some(false);
}
}
FieldMappingType::I64(options, _)
Expand All @@ -585,6 +591,7 @@ impl From<FieldMappingEntry> for FieldMappingEntryForSerialization {
type_with_cardinality,
fast,
indexed,
fieldnorms,
record,
stored,
tokenizer,
Expand Down Expand Up @@ -623,8 +630,10 @@ impl FieldMappingEntryForSerialization {
)
}
let mut options = TextOptions::default();

if self.indexed.unwrap_or(true) {
let mut indexing_options = TextFieldIndexing::default();
indexing_options = indexing_options.set_fieldnorms(self.fieldnorms.unwrap_or(false));
if let Some(index_option) = self.record {
indexing_options = indexing_options.set_index_option(index_option);
}
Expand Down Expand Up @@ -673,6 +682,9 @@ impl FieldMappingEntryForSerialization {
}
if self.indexed.unwrap_or(true) {
options = options.set_indexed();
if self.fieldnorms.unwrap_or(false) {
options = options.set_fieldnorms();
}
}
if self.fast {
options = options.set_fast();
Expand Down Expand Up @@ -720,6 +732,9 @@ impl FieldMappingEntryForSerialization {
}
if self.indexed.unwrap_or(true) {
options = options.set_indexed();
if self.fieldnorms.unwrap_or(false) {
options = options.set_fieldnorm();
}
}
Ok(options)
}
Expand Down Expand Up @@ -821,6 +836,32 @@ mod tests {
Ok(())
}

#[test]
fn test_deserialize_valid_fieldnorms() -> anyhow::Result<()> {
let result = serde_json::from_str::<FieldMappingEntry>(
r#"
{
"name": "my_field_name",
"type": "text",
"stored": true,
"indexed": true,
"fieldnorms": true,
"record": "basic",
"tokenizer": "english"
}"#,
);
match result.unwrap().mapping_type {
FieldMappingType::Text(options, _) => {
assert_eq!(options.is_stored(), true);
let index_options = options.get_indexing_options().unwrap();
assert_eq!(index_options.fieldnorms(), true);
}
_ => panic!("wrong property type"),
}

Ok(())
}

#[test]
fn test_error_on_text_with_invalid_options() -> anyhow::Result<()> {
let result = serde_json::from_str::<FieldMappingEntry>(
Expand Down Expand Up @@ -963,6 +1004,7 @@ mod tests {
match result.mapping_type {
FieldMappingType::I64(options, cardinality) => {
assert_eq!(options.is_indexed(), true); // default
assert_eq!(options.fieldnorms(), false); // default
assert_eq!(options.is_fast(), false); // default
assert_eq!(options.is_stored(), true); // default
assert_eq!(cardinality, Cardinality::MultiValues);
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "log_level",
"record": "basic",
"stored": true,
Expand All @@ -32,6 +34,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "message",
"record": "position",
"stored": true,
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "log_level",
"record": "basic",
"stored": true,
Expand All @@ -32,6 +34,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "message",
"record": "position",
"stored": true,
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,116 @@
{
"index": {
"checkpoint": {
"kafka-source": {
"00000000000000000000": "00000000000000000042"
}
},
"create_timestamp": 1789,
"doc_mapping": {
"field_mappings": [
{
"fast": true,
"indexed": true,
"name": "tenant_id",
"stored": true,
"type": "u64"
},
{
"fast": true,
"indexed": true,
"name": "timestamp",
"stored": true,
"type": "i64"
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "log_level",
"record": "basic",
"stored": true,
"tokenizer": "raw",
"type": "text"
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "message",
"record": "position",
"stored": true,
"tokenizer": "default",
"type": "text"
}
],
"store_source": true,
"tag_fields": [
"log_level",
"tenant_id"
]
},
"index_id": "my-index",
"index_uri": "s3://quickwit-indexes/my-index",
"indexing_settings": {
"commit_timeout_secs": 301,
"demux_enabled": true,
"demux_field": "tenant_id",
"merge_enabled": true,
"merge_policy": {
"demux_factor": 7,
"max_merge_factor": 11,
"merge_factor": 9
},
"resources": {
"heap_size": 3,
"num_threads": 3
},
"sort_field": "timestamp",
"sort_order": "asc",
"split_num_docs_target": 10000001,
"timestamp_field": "timestamp"
},
"search_settings": {
"default_search_fields": [
"message"
]
},
"sources": [
{
"params": {
"client_params": {},
"topic": "kafka-topic"
},
"source_id": "kafka-source",
"source_type": "kafka"
}
],
"update_timestamp": 1789,
"version": "1"
},
"splits": [
{
"create_timestamp": 3,
"demux_num_ops": 1,
"footer_offsets": {
"end": 2000,
"start": 1000
},
"num_docs": 12303,
"size_in_bytes": 234234,
"split_id": "split",
"split_state": "Published",
"tags": [
"234",
"aaa"
],
"time_range": {
"end": 130198,
"start": 121000
},
"update_timestamp": 1789,
"version": "1"
}
],
"version": "0"
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,116 @@
{
"index": {
"checkpoint": {
"kafka-source": {
"00000000000000000000": "00000000000000000042"
}
},
"create_timestamp": 1789,
"doc_mapping": {
"field_mappings": [
{
"fast": true,
"indexed": true,
"name": "tenant_id",
"stored": true,
"type": "u64"
},
{
"fast": true,
"indexed": true,
"name": "timestamp",
"stored": true,
"type": "i64"
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "log_level",
"record": "basic",
"stored": true,
"tokenizer": "raw",
"type": "text"
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "message",
"record": "position",
"stored": true,
"tokenizer": "default",
"type": "text"
}
],
"store_source": true,
"tag_fields": [
"log_level",
"tenant_id"
]
},
"index_id": "my-index",
"index_uri": "s3://quickwit-indexes/my-index",
"indexing_settings": {
"commit_timeout_secs": 301,
"demux_enabled": true,
"demux_field": "tenant_id",
"merge_enabled": true,
"merge_policy": {
"demux_factor": 7,
"max_merge_factor": 11,
"merge_factor": 9
},
"resources": {
"heap_size": 3,
"num_threads": 3
},
"sort_field": "timestamp",
"sort_order": "asc",
"split_num_docs_target": 10000001,
"timestamp_field": "timestamp"
},
"search_settings": {
"default_search_fields": [
"message"
]
},
"sources": [
{
"params": {
"client_params": {},
"topic": "kafka-topic"
},
"source_id": "kafka-source",
"source_type": "kafka"
}
],
"update_timestamp": 1789,
"version": "1"
},
"splits": [
{
"create_timestamp": 3,
"demux_num_ops": 1,
"footer_offsets": {
"end": 2000,
"start": 1000
},
"num_docs": 12303,
"size_in_bytes": 234234,
"split_id": "split",
"split_state": "Published",
"tags": [
"234",
"aaa"
],
"time_range": {
"end": 130198,
"start": 121000
},
"update_timestamp": 1789,
"version": "1"
}
],
"version": "0"
}
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "log_level",
"record": "basic",
"stored": true,
Expand All @@ -32,6 +34,8 @@
},
{
"fast": false,
"fieldnorms": false,
"indexed": true,
"name": "message",
"record": "position",
"stored": true,
Expand Down
Loading