Update src/streamlit_app.py
Browse files- src/streamlit_app.py +180 -38
src/streamlit_app.py
CHANGED
|
@@ -1,40 +1,182 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
import pandas as pd
|
| 4 |
import streamlit as st
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
|
| 17 |
-
num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
|
| 18 |
-
|
| 19 |
-
indices = np.linspace(0, 1, num_points)
|
| 20 |
-
theta = 2 * np.pi * num_turns * indices
|
| 21 |
-
radius = indices
|
| 22 |
-
|
| 23 |
-
x = radius * np.cos(theta)
|
| 24 |
-
y = radius * np.sin(theta)
|
| 25 |
-
|
| 26 |
-
df = pd.DataFrame({
|
| 27 |
-
"x": x,
|
| 28 |
-
"y": y,
|
| 29 |
-
"idx": indices,
|
| 30 |
-
"rand": np.random.randn(num_points),
|
| 31 |
-
})
|
| 32 |
-
|
| 33 |
-
st.altair_chart(alt.Chart(df, height=700, width=700)
|
| 34 |
-
.mark_point(filled=True)
|
| 35 |
-
.encode(
|
| 36 |
-
x=alt.X("x", axis=None),
|
| 37 |
-
y=alt.Y("y", axis=None),
|
| 38 |
-
color=alt.Color("idx", legend=None, scale=alt.Scale()),
|
| 39 |
-
size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
|
| 40 |
-
))
|
|
|
|
| 1 |
+
# app.py
|
| 2 |
+
|
|
|
|
| 3 |
import streamlit as st
|
| 4 |
+
import random
|
| 5 |
+
from urllib.parse import urlparse
|
| 6 |
+
from concurrent.futures import ThreadPoolExecutor
|
| 7 |
+
from search_engine_parser.core.engines.google import Search as GoogleSearch
|
| 8 |
+
from search_engine_parser.core.engines.yandex import Search as YandexSearch
|
| 9 |
+
from search_engine_parser.core.engines.bing import Search as BingSearch
|
| 10 |
+
from search_engine_parser.core.engines.duckduckgo import Search as DuckDuckGoSearch
|
| 11 |
+
|
| 12 |
+
# --- Конфигурация страницы Streamlit ---
|
| 13 |
+
st.set_page_config(
|
| 14 |
+
page_title="MetaSearch | Поисковик-агрегатор",
|
| 15 |
+
page_icon="🌪️",
|
| 16 |
+
layout="wide",
|
| 17 |
+
initial_sidebar_state="collapsed"
|
| 18 |
+
)
|
| 19 |
+
|
| 20 |
+
# --- Кастомные стили для красоты ---
|
| 21 |
+
st.markdown("""
|
| 22 |
+
<style>
|
| 23 |
+
/* Стиль для контейнера с результатами */
|
| 24 |
+
.result-card {
|
| 25 |
+
border: 1px solid #e1e4e8;
|
| 26 |
+
border-radius: 12px;
|
| 27 |
+
padding: 1.5rem;
|
| 28 |
+
margin-bottom: 1.5rem;
|
| 29 |
+
box-shadow: 0 4px 6px rgba(0,0,0,0.05);
|
| 30 |
+
transition: box-shadow 0.3s ease-in-out;
|
| 31 |
+
background-color: #ffffff;
|
| 32 |
+
}
|
| 33 |
+
.result-card:hover {
|
| 34 |
+
box-shadow: 0 8px 12px rgba(0,0,0,0.1);
|
| 35 |
+
}
|
| 36 |
+
/* Стиль для заголовка-ссылки */
|
| 37 |
+
.result-card h3 a {
|
| 38 |
+
color: #0d6efd !important;
|
| 39 |
+
text-decoration: none !important;
|
| 40 |
+
font-weight: 600;
|
| 41 |
+
}
|
| 42 |
+
.result-card h3 a:hover {
|
| 43 |
+
text-decoration: underline !important;
|
| 44 |
+
}
|
| 45 |
+
/* Стиль для домена и источника */
|
| 46 |
+
.source-info {
|
| 47 |
+
display: flex;
|
| 48 |
+
align-items: center;
|
| 49 |
+
gap: 10px;
|
| 50 |
+
font-size: 0.9em;
|
| 51 |
+
color: #586069;
|
| 52 |
+
}
|
| 53 |
+
.domain {
|
| 54 |
+
color: #28a745; /* Зеленый цвет для домена */
|
| 55 |
+
}
|
| 56 |
+
</style>
|
| 57 |
+
""", unsafe_allow_html=True)
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
# --- Словарь с поисковиками и их иконками ---
|
| 61 |
+
ENGINES = {
|
| 62 |
+
"Google": {"engine": GoogleSearch(), "icon": "🇬"},
|
| 63 |
+
"Yandex": {"engine": YandexSearch(), "icon": "🇷🇺"},
|
| 64 |
+
"Bing": {"engine": BingSearch(), "icon": "🇧"},
|
| 65 |
+
"DuckDuckGo": {"engine": DuckDuckGoSearch(), "icon": "🦆"},
|
| 66 |
+
}
|
| 67 |
+
|
| 68 |
+
# --- Функции для поиска ---
|
| 69 |
+
|
| 70 |
+
def search_one_engine(engine_name: str, engine_details: dict, query: str, page: int = 1):
|
| 71 |
+
"""
|
| 72 |
+
Выполняет поиск в одном поисковике и обрабатывает результаты.
|
| 73 |
+
"""
|
| 74 |
+
try:
|
| 75 |
+
results = engine_details["engine"].search(query, page)
|
| 76 |
+
# Приводим результаты к единому формату
|
| 77 |
+
formatted_results = []
|
| 78 |
+
for res in results['results']: # Большинство парсеров возвращают ключ 'results'
|
| 79 |
+
formatted_results.append({
|
| 80 |
+
"engine": engine_name,
|
| 81 |
+
"icon": engine_details["icon"],
|
| 82 |
+
"title": res.get("titles", "Без заголовка"),
|
| 83 |
+
"link": res.get("links"),
|
| 84 |
+
"description": res.get("descriptions", "")
|
| 85 |
+
})
|
| 86 |
+
return formatted_results
|
| 87 |
+
except Exception as e:
|
| 88 |
+
# st.warning(f"Не удалось получить результаты из {engine_name}: {e}")
|
| 89 |
+
print(f"Ошибка при поиске в {engine_name}: {e}")
|
| 90 |
+
return []
|
| 91 |
+
|
| 92 |
+
def run_parallel_search(query: str):
|
| 93 |
+
"""
|
| 94 |
+
Запускает поиск во всех движках параллельно.
|
| 95 |
+
"""
|
| 96 |
+
all_results = []
|
| 97 |
+
with ThreadPoolExecutor(max_workers=len(ENGINES)) as executor:
|
| 98 |
+
# Создаем задачи для каждого поисковика
|
| 99 |
+
future_to_engine = {
|
| 100 |
+
executor.submit(search_one_engine, name, details, query): name
|
| 101 |
+
for name, details in ENGINES.items()
|
| 102 |
+
}
|
| 103 |
+
for future in future_to_engine:
|
| 104 |
+
try:
|
| 105 |
+
results = future.result()
|
| 106 |
+
if results:
|
| 107 |
+
all_results.extend(results)
|
| 108 |
+
except Exception as e:
|
| 109 |
+
engine_name = future_to_engine[future]
|
| 110 |
+
print(f"В задаче для {engine_name} произошла ошибка: {e}")
|
| 111 |
+
|
| 112 |
+
return all_results
|
| 113 |
+
|
| 114 |
+
|
| 115 |
+
# --- Основной интерфейс приложения ---
|
| 116 |
+
|
| 117 |
+
st.title("🌪️ MetaSearch")
|
| 118 |
+
st.markdown("##### Ваш умный агрегатор поиска. Получайте лучшее из Google, Yandex, Bing и DuckDuckGo.")
|
| 119 |
+
|
| 120 |
+
# Поле для ввода и кнопка
|
| 121 |
+
search_query = st.text_input(
|
| 122 |
+
"Что будем искать?",
|
| 123 |
+
placeholder="Например: лучшие практики Python...",
|
| 124 |
+
label_visibility="collapsed"
|
| 125 |
+
)
|
| 126 |
+
|
| 127 |
+
if st.button("Найти", type="primary", use_container_width=True) or search_query:
|
| 128 |
+
if search_query:
|
| 129 |
+
with st.spinner("Собираем информацию со всего интернета... 🕵️♂️"):
|
| 130 |
+
# Запускаем поиск
|
| 131 |
+
search_results = run_parallel_search(search_query)
|
| 132 |
+
|
| 133 |
+
# Обработка результатов: удаление дубликатов и перемешивание
|
| 134 |
+
unique_results = []
|
| 135 |
+
seen_links = set()
|
| 136 |
+
for result in search_results:
|
| 137 |
+
if result['link'] and result['link'] not in seen_links:
|
| 138 |
+
unique_results.append(result)
|
| 139 |
+
seen_links.add(result['link'])
|
| 140 |
+
|
| 141 |
+
if not unique_results:
|
| 142 |
+
st.error("К сожалению, ничего не найдено. Попробуйте другой запрос.")
|
| 143 |
+
else:
|
| 144 |
+
random.shuffle(unique_results) # Перемешиваем
|
| 145 |
+
st.success(f"Найдено {len(unique_results)} уникальных результатов. Вот лучшие из них:")
|
| 146 |
+
st.markdown("---")
|
| 147 |
+
|
| 148 |
+
# Отображение результатов
|
| 149 |
+
for result in unique_results:
|
| 150 |
+
# Извлекаем домен для красивого отображения
|
| 151 |
+
domain = "Неизвестный домен"
|
| 152 |
+
try:
|
| 153 |
+
parsed_uri = urlparse(result['link'])
|
| 154 |
+
domain = f'{parsed_uri.netloc}'
|
| 155 |
+
except:
|
| 156 |
+
pass
|
| 157 |
+
|
| 158 |
+
# Используем st.markdown с unsafe_allow_html для создания карточки
|
| 159 |
+
st.markdown(f"""
|
| 160 |
+
<div class="result-card">
|
| 161 |
+
<div class="source-info">
|
| 162 |
+
<span>{result['icon']} Найдено в <b>{result['engine']}</b></span>
|
| 163 |
+
</div>
|
| 164 |
+
<h3><a href="{result['link']}" target="_blank">{result['title']}</a></h3>
|
| 165 |
+
<p>{result.get('description', 'Описание недоступно.')}</p>
|
| 166 |
+
<div class="source-info">
|
| 167 |
+
<span class="domain">{domain}</span>
|
| 168 |
+
</div>
|
| 169 |
+
</div>
|
| 170 |
+
""", unsafe_allow_html=True)
|
| 171 |
+
else:
|
| 172 |
+
st.info("Пожалуйста, введите поисковый запрос.")
|
| 173 |
|
| 174 |
+
st.markdown(
|
| 175 |
+
"""
|
| 176 |
+
---
|
| 177 |
+
<p style='text-align: center; color: grey;'>
|
| 178 |
+
Создано с помощью Streamlit. <br>
|
| 179 |
+
<b>Внимание:</b> Это неофициальный инструмент. Поисковые системы могут временно блокировать запросы.
|
| 180 |
+
</p>
|
| 181 |
+
""", unsafe_allow_html=True
|
| 182 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|