Skip to content

armgit / mwmbl

Public Imported from GitHub
0

An open source, non-profit web search engine

Group results by domain

Daoud Clarke3e0240a0e1b1
diff --git a/front-end/assets/css/global.css b/front-end/assets/css/global.css
index 92c3d78..b8441a3 100644
--- a/front-end/assets/css/global.css
+++ b/front-end/assets/css/global.css
@@ -120,7 +120,7 @@ body {
padding: 0;
}
-.result {
+.result.expanded {
min-height: 120px;
}
@@ -257,9 +257,9 @@ a {
background-color: var(--dark-gray-color);
color: white;
padding: 5px 10px;
- margin: 10px 10px 10px 0;
- width: 22px;
- height: 22px;
+ margin: 0 10px 0 0;
+ width: 18px;
+ height: 18px;
display: flex;
align-items: center;
justify-content: center;
diff --git a/mwmbl/rankeval/evaluation/evaluate_remote.py b/mwmbl/rankeval/evaluation/evaluate_remote.py
index af43fe4..1a2b96f 100644
--- a/mwmbl/rankeval/evaluation/evaluate_remote.py
+++ b/mwmbl/rankeval/evaluation/evaluate_remote.py
@@ -33,7 +33,7 @@ def run():
ranker = LTRRanker(RemoteIndex(), DummyCompleter(), model, 1000, True, 5)
# ranker = HeuristicRanker(RemoteIndex(), DummyCompleter())
model = MwmblRankingModel(ranker)
- evaluate(model, fraction=0.01, use_test=True)
+ evaluate(model, fraction=0.01, use_test=False)
def single_query(query: str):
diff --git a/mwmbl/templates/result.html b/mwmbl/templates/result.html
index cec0eb8..22a9723 100644
--- a/mwmbl/templates/result.html
+++ b/mwmbl/templates/result.html
@@ -1,5 +1,5 @@
{% load result_filters %}
-<li class="result" is="mwmbl-result">
+<li is="mwmbl-result" class="result {% if result.compact %}compact{% else %}expanded{% endif %}">
<input type="hidden" name="url" value="{{result.url}}" />
<input type="hidden" name="title" value="{{result.title}}" />
<input type="hidden" name="extract" value="{{result.extract}}" />
@@ -15,10 +15,14 @@
{% endif %}
<div class="result-link">
<a href="{{result.url}}">
- <p><span class='link'>{{result.url}}</span> <span class="via">— found via <span class="source">{{result.state|convert_state_to_source|title}}</span></span></p>
+ {% if not result.compact %}
+ <p><span class='link'>{{result.url}}</span> <span class="via">— found via <span class="source">{{result.state|convert_state_to_source|title}}</span></span></p>
+ {% endif %}
<p class='title'>{{result.title|format_for_query:query}}</p>
</a>
- <p class='extract'>{{result.extract|format_for_query:query}}</p>
+ {% if not result.compact %}
+ <p class='extract'>{{result.extract|format_for_query:query}}</p>
+ {% endif %}
</div>
</div>
</li>
diff --git a/mwmbl/tinysearchengine/rank.py b/mwmbl/tinysearchengine/rank.py
index ad2fe8a..2dba441 100644
--- a/mwmbl/tinysearchengine/rank.py
+++ b/mwmbl/tinysearchengine/rank.py
@@ -346,7 +346,7 @@ def get_wiki_results(s: str, max_wiki_results: int):
wiki_response = session.get(WIKI_SEARCH_API_URL.format(query=escaped_query)).json()
wiki_results = [Document(result['title'], get_wiki_url(result['title']), clean_html(result['snippet']),
max_wiki_results + 1 - i, s, state=DocumentState.FROM_WIKI)
- for i, result in enumerate(wiki_response['query']['search'])]
+ for i, result in enumerate(wiki_response['query']['search'][:max_wiki_results])]
return wiki_results
diff --git a/mwmbl/views.py b/mwmbl/views.py
index 7f81cfb..df84dec 100644
--- a/mwmbl/views.py
+++ b/mwmbl/views.py
@@ -1,4 +1,5 @@
import re
+from collections import defaultdict
from dataclasses import asdict
from datetime import datetime
from logging import getLogger
@@ -64,14 +65,33 @@ def justext_with_dom(html_text, stoplist, length_low=LENGTH_LOW_DEFAULT,
def _prepare_results(results: list[Document]) -> list[dict[str, Any]]:
- return [{
- "title": result.title,
- "url": result.url,
- "extract": result.extract,
- "score": result.score,
- "state": result.state,
- }
- for result in results]
+ grouped_domain_results = defaultdict(list)
+ for result in results:
+ domain = parse_url(result.url).netloc
+ grouped_domain_results[domain].append(result)
+
+ rearranged_results = []
+ for domain, domain_results in grouped_domain_results.items():
+ result = domain_results[0]
+ rearranged_results.append({
+ "title": result.title,
+ "url": result.url,
+ "extract": result.extract,
+ "score": result.score,
+ "state": result.state,
+ "compact": False,
+ })
+
+ for result in domain_results[1:]:
+ rearranged_results.append({
+ "title": result.title,
+ "url": result.url,
+ "extract": result.extract,
+ "score": result.score,
+ "state": result.state,
+ "compact": True,
+ })
+ return rearranged_results
def index(request):