forked from geolba/tethys.frontend
Opensearch progress.
- Term search works - Pending faceted search - Ongoing: highlight of fuzzy results
This commit is contained in:
parent
6d1c1b28c3
commit
4f53411d07
5 changed files with 292 additions and 204 deletions
|
|
@ -1,7 +1,7 @@
|
|||
import api from "../api/api";
|
||||
// import { Observable, of } from "rxjs";
|
||||
import { Observable } from "rxjs";
|
||||
import { map } from "rxjs/operators";
|
||||
import { tap, map } from "rxjs/operators";
|
||||
import { Dataset, DbDataset, Suggestion } from "@/models/dataset";
|
||||
import { OpenSearchResponse, SolrResponse } from "@/models/headers";
|
||||
import { ActiveFilterCategories } from "@/models/solr";
|
||||
|
|
@ -16,130 +16,130 @@ class DatasetService {
|
|||
*
|
||||
* @param {string} searchTerm - The search term to query.
|
||||
*/
|
||||
async fetchDataFromOpenSearch(searchTerm: string): Promise<void> {
|
||||
// Define the OpenSearch endpoint URL
|
||||
const url = "http://opensearch.geoinformation.dev/tethys-records/_search";
|
||||
// async fetchDataFromOpenSearch(searchTerm: string): Promise<void> {
|
||||
// // Define the OpenSearch endpoint URL
|
||||
// const url = "http://opensearch.geoinformation.dev/tethys-records/_search";
|
||||
|
||||
// Set the headers for the POST request
|
||||
const headers = {
|
||||
"Content-Type": "application/json",
|
||||
};
|
||||
// // Set the headers for the POST request
|
||||
// const headers = {
|
||||
// "Content-Type": "application/json",
|
||||
// };
|
||||
|
||||
// Construct the body of the POST request
|
||||
const body = {
|
||||
query: {
|
||||
bool: {
|
||||
// The `should` clause specifies that at least one of these conditions must match
|
||||
should: [
|
||||
{
|
||||
// Match the search term in the title field with fuzziness enabled and a boost of 3
|
||||
match: {
|
||||
title: {
|
||||
query: searchTerm,
|
||||
fuzziness: "AUTO", // Enable fuzzy search
|
||||
boost: 3 // Boosting the relevance of title matches
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
// Match the search term in the author field with fuzziness enabled and a boost of 2
|
||||
match: {
|
||||
author: {
|
||||
query: searchTerm,
|
||||
fuzziness: "AUTO", // Enable fuzzy search
|
||||
boost: 2 // Boosting the relevance of author matches
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
// Match the search term in the subject field with fuzziness enabled and a boost of 1
|
||||
match: {
|
||||
subject: {
|
||||
query: searchTerm,
|
||||
fuzziness: "AUTO", // Enable fuzzy search
|
||||
boost: 1 // Boosting the relevance of subject matches
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
// Match the search term in the title field with a wildcard
|
||||
wildcard: {
|
||||
title: {
|
||||
value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
boost: 3 // Boosting the relevance of title matches
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
// Match the search term in the author field with a wildcard
|
||||
wildcard: {
|
||||
author: {
|
||||
value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
boost: 2 // Boosting the relevance of author matches
|
||||
}
|
||||
}
|
||||
},
|
||||
{
|
||||
// Match the search term in the subject field with a wildcard
|
||||
wildcard: {
|
||||
subject: {
|
||||
value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
boost: 1 // Boosting the relevance of subject matches
|
||||
}
|
||||
}
|
||||
}
|
||||
],
|
||||
// Ensure that at least one of the `should` clauses must match
|
||||
minimum_should_match: 1
|
||||
}
|
||||
},
|
||||
// Limit the number of search results to 10
|
||||
size: 10,
|
||||
// Start from the first result (pagination)
|
||||
from: 0,
|
||||
// Sort the results by the `server_date_published` field in descending order
|
||||
sort: [
|
||||
{ server_date_published: { order: "desc" } }
|
||||
],
|
||||
// Aggregations to provide facets for the `language` and `subject` fields
|
||||
aggs: {
|
||||
language: {
|
||||
terms: {
|
||||
field: "language.keyword" // Aggregate by the exact values of the `language` field
|
||||
}
|
||||
},
|
||||
subject: {
|
||||
terms: {
|
||||
field: "subjects.keyword", // Aggregate by the exact values of the `subjects` field
|
||||
size: 10 // Limit the number of aggregation buckets to 10
|
||||
}
|
||||
}
|
||||
}
|
||||
};
|
||||
// // Construct the body of the POST request
|
||||
// const body = {
|
||||
// query: {
|
||||
// bool: {
|
||||
// // The `should` clause specifies that at least one of these conditions must match
|
||||
// should: [
|
||||
// {
|
||||
// // Match the search term in the title field with fuzziness enabled and a boost of 3
|
||||
// match: {
|
||||
// title: {
|
||||
// query: searchTerm,
|
||||
// fuzziness: "AUTO", // Enable fuzzy search
|
||||
// boost: 3 // Boosting the relevance of title matches
|
||||
// }
|
||||
// }
|
||||
// },
|
||||
// {
|
||||
// // Match the search term in the author field with fuzziness enabled and a boost of 2
|
||||
// match: {
|
||||
// author: {
|
||||
// query: searchTerm,
|
||||
// fuzziness: "AUTO", // Enable fuzzy search
|
||||
// boost: 2 // Boosting the relevance of author matches
|
||||
// }
|
||||
// }
|
||||
// },
|
||||
// {
|
||||
// // Match the search term in the subject field with fuzziness enabled and a boost of 1
|
||||
// match: {
|
||||
// subject: {
|
||||
// query: searchTerm,
|
||||
// fuzziness: "AUTO", // Enable fuzzy search
|
||||
// boost: 1 // Boosting the relevance of subject matches
|
||||
// }
|
||||
// }
|
||||
// },
|
||||
// {
|
||||
// // Match the search term in the title field with a wildcard
|
||||
// wildcard: {
|
||||
// title: {
|
||||
// value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
// boost: 3 // Boosting the relevance of title matches
|
||||
// }
|
||||
// }
|
||||
// },
|
||||
// {
|
||||
// // Match the search term in the author field with a wildcard
|
||||
// wildcard: {
|
||||
// author: {
|
||||
// value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
// boost: 2 // Boosting the relevance of author matches
|
||||
// }
|
||||
// }
|
||||
// },
|
||||
// {
|
||||
// // Match the search term in the subject field with a wildcard
|
||||
// wildcard: {
|
||||
// subject: {
|
||||
// value: `${searchTerm}*`, // Wildcard search for terms starting with searchTerm
|
||||
// boost: 1 // Boosting the relevance of subject matches
|
||||
// }
|
||||
// }
|
||||
// }
|
||||
// ],
|
||||
// // Ensure that at least one of the `should` clauses must match
|
||||
// minimum_should_match: 1
|
||||
// }
|
||||
// },
|
||||
// // Limit the number of search results to 10
|
||||
// size: 10,
|
||||
// // Start from the first result (pagination)
|
||||
// from: 0,
|
||||
// // Sort the results by the `server_date_published` field in descending order
|
||||
// sort: [
|
||||
// { server_date_published: { order: "desc" } }
|
||||
// ],
|
||||
// // Aggregations to provide facets for the `language` and `subject` fields
|
||||
// aggs: {
|
||||
// language: {
|
||||
// terms: {
|
||||
// field: "language.keyword" // Aggregate by the exact values of the `language` field
|
||||
// }
|
||||
// },
|
||||
// subject: {
|
||||
// terms: {
|
||||
// field: "subjects.keyword", // Aggregate by the exact values of the `subjects` field
|
||||
// size: 10 // Limit the number of aggregation buckets to 10
|
||||
// }
|
||||
// }
|
||||
// }
|
||||
// };
|
||||
|
||||
try {
|
||||
// Send the POST request to the OpenSearch endpoint
|
||||
const response = await fetch(url, {
|
||||
method: "POST",
|
||||
headers: headers,
|
||||
body: JSON.stringify(body),
|
||||
});
|
||||
// try {
|
||||
// // Send the POST request to the OpenSearch endpoint
|
||||
// const response = await fetch(url, {
|
||||
// method: "POST",
|
||||
// headers: headers,
|
||||
// body: JSON.stringify(body),
|
||||
// });
|
||||
|
||||
// Check if the response is not successful
|
||||
if (!response.ok) {
|
||||
throw new Error(`Failed to fetch data from ${url}, status: ${response.status}`);
|
||||
}
|
||||
// // Check if the response is not successful
|
||||
// if (!response.ok) {
|
||||
// throw new Error(`Failed to fetch data from ${url}, status: ${response.status}`);
|
||||
// }
|
||||
|
||||
// Parse the response JSON
|
||||
const data = await response.json();
|
||||
// Log the data from OpenSearch
|
||||
console.log("Data from OpenSearch:", data);
|
||||
console.log("Hits:", data.hits.total.value);
|
||||
} catch (error) {
|
||||
// Log any errors that occur during the fetch process
|
||||
console.error("Error fetching data:", error);
|
||||
}
|
||||
}
|
||||
// // Parse the response JSON
|
||||
// const data = await response.json();
|
||||
// // Log the data from OpenSearch
|
||||
// console.log("Data from OpenSearch:", data);
|
||||
// console.log("Hits:", data.hits.total.value);
|
||||
// } catch (error) {
|
||||
// // Log any errors that occur during the fetch process
|
||||
// console.error("Error fetching data:", error);
|
||||
// }
|
||||
// }
|
||||
|
||||
/* https://tethys.at/solr/rdr_data/select?&0=fl%3Did%2Clicence%2Cserver_date_published%2Cabstract_output%2Cidentifier%2Ctitle_output%2Ctitle_additional%2Cauthor%2Csubject%2Cdoctype&q=%2A
|
||||
&q.op=or&defType=edismax&qf=title%5E3%20author%5E2%20subject%5E1&indent=on&wt=json&rows=10&start=0&sort=server_date_published%20desc&facet=on&json.facet.language=%7B%20type%3A%20%22
|
||||
|
|
@ -156,25 +156,53 @@ class DatasetService {
|
|||
should: [
|
||||
{ match: { title: { query: term, fuzziness: "AUTO", boost: 3 } } },
|
||||
{ match: { author: { query: term, fuzziness: "AUTO", boost: 2 } } },
|
||||
{ match: { subject: { query: term, fuzziness: "AUTO", boost: 1 } } },
|
||||
{ match: { subjects: { query: term, fuzziness: "AUTO", boost: 1 } } }, // In SOLR is "subject"!
|
||||
{ wildcard: { title: { value: `${term}*`, boost: 3 } } },
|
||||
{ wildcard: { author: { value: `${term}*`, boost: 2 } } },
|
||||
{ wildcard: { subject: { value: `${term}*`, boost: 1 } } }
|
||||
{ wildcard: { subjects: { value: `${term}*`, boost: 1 } } } // In SOLR is "subject"!
|
||||
],
|
||||
minimum_should_match: 1
|
||||
}
|
||||
},
|
||||
size: 10,
|
||||
from: 0,
|
||||
sort: [{ server_date_published: { order: "desc" } }],
|
||||
// sort: [{ server_date_published: { order: "desc" } }],
|
||||
sort: [{ _score: { order: "desc" } }], // Sort by _score in descending order
|
||||
track_scores: true, // This ensures "_score" is included even when sorting by other criteria. Otherwise the relevance score is not calculated
|
||||
aggs: {
|
||||
language: { terms: { field: "language.keyword" } },
|
||||
subject: { terms: { field: "subjects.keyword", size: 10 } }
|
||||
subjects: { terms: { field: "subjects.keyword", size: 10 } } // In SOLR is "subject"!
|
||||
},
|
||||
highlight: {
|
||||
fields: {
|
||||
title: {},
|
||||
author: {},
|
||||
subjects: {}
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
// Make API call to OpenSearch and return the result
|
||||
/**
|
||||
* When a POST request is made to the OpenSearch server using the api.post<OpenSearchResponse> method, the response received from OpenSearch is an object that includes various details about the search results.
|
||||
* One of the key properties of this response object is _source, which is an array of documents (datasets) that match the search criteria.
|
||||
* It is used the pipe method to chain RxJS operators to the Observable returned by api.get. The map operator is used to transform the emitted items of the Observable.
|
||||
*/
|
||||
return api.post<OpenSearchResponse>(this.openSearchUrl, body).pipe(
|
||||
// tap(response => console.log("OpenSearchResponse:", response)), // Log the complete response
|
||||
// tap(response => console.log("Aggre:", response.aggregations?.subjects.buckets[0])), // log the first subject of the array of subjects returned
|
||||
// tap(response => console.log("Hits:", response.hits)), // log the first subject of the array of subjects returned
|
||||
|
||||
map(response => response.hits.hits.map(hit => hit._source))
|
||||
|
||||
// map(response => response.hits.hits.map(hit => {
|
||||
// const source = hit._source;
|
||||
// const highlights = hit._highlight || {};
|
||||
// return {
|
||||
// ...source,
|
||||
// highlights
|
||||
// };
|
||||
// }))
|
||||
);
|
||||
}
|
||||
|
||||
|
|
@ -219,7 +247,7 @@ class DatasetService {
|
|||
* It is used the pipe method to chain RxJS operators to the Observable returned by api.get. The map operator is used to transform the emitted items of the Observable.
|
||||
*/
|
||||
const stations = api.get<SolrResponse>(base, q_params).pipe(map((res: SolrResponse) => res.response.docs));
|
||||
|
||||
|
||||
return stations;
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Add table
editor.link_modal.header
Reference in a new issue