Spaces:
Running
Running
File size: 9,531 Bytes
078c1e1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 |
import numpy as np
import pandas as pd
from dateutil import parser
from flask import Flask, render_template
from flask_cors import cross_origin, CORS
from db_operations.db_operations import DBOperations
import logging
import traceback
import redis
from datetime import datetime
from functools import lru_cache
from word_cloud import get_frequent_words_html
from config import NEWS_RETENTION_SECONDS
app = Flask(__name__)
CORS(app)
redis_client = redis.Redis(host='localhost', port=6379, decode_responses=True)
db = DBOperations()
REFRESH_FREQ = 300 # 300 secs = 5 mins
def is_db_fetch_reqd():
try:
env_news_time = redis_client.get('NEWSFETCHTIME')
logging.warning(f'fetch_time_env_var: {env_news_time}')
fetch_flag = 1
if env_news_time is None:
redis_client.set("NEWSFETCHTIME", str(datetime.now()))
fetch_flag = 1
if env_news_time is not None:
fetch_time_lapse_seconds = (datetime.now() - datetime.strptime(env_news_time, '%Y-%m-%d %H:%M:%S.%f')).seconds
if fetch_time_lapse_seconds <= REFRESH_FREQ:
fetch_flag = 0
else:
redis_client.set("NEWSFETCHTIME", str(datetime.now()))
fetch_flag = 1
except Exception as e:
print(e)
fetch_flag = 1
return fetch_flag
def correct_date(x):
if (not isinstance(x, str)) or (str(x).find(":") == -1):
logging.warning(f'correct_date() error: {x} is not the right date format')
return "2020-11-07 00:36:44+05:30"
return x
def date_time_parser(dt):
"""
Computes the minutes elapsed since published time.
:param dt: date
:return: int, minutes elapsed.
"""
try:
return int(np.round((dt.now(dt.tz) - dt).total_seconds() / 60, 0))
except:
logging.warning(f'date_time_parser() error: {dt} is not the right date format')
return 100000
def elapsed_time_str(mins):
"""
Return the time elapsed string from minutes passed as an argument.
:param mins: int, minutes elapsed.
:return: str, time elapsed string
"""
try:
time_str = ''
hours = int(mins / 60)
days = np.round(mins / (60 * 24), 1)
remaining_mins = int(mins - (hours * 60))
if days >= 1:
time_str = f'{str(days)} days ago'
if days == 1:
time_str = 'a day ago'
elif (days < 1) & (hours < 24) & (mins >= 60):
time_str = f'{str(hours)} hours and {str(remaining_mins)} mins ago'
if (hours == 1) & (remaining_mins > 1):
time_str = f'an hour and {str(remaining_mins)} mins ago'
if (hours == 1) & (remaining_mins == 1):
time_str = f'an hour and a min ago'
if (hours > 1) & (remaining_mins == 1):
time_str = f'{str(hours)} hours and a min ago'
if (hours > 1) & (remaining_mins == 0):
time_str = f'{str(hours)} hours ago'
if ((mins / 60) == 1) & (remaining_mins == 0):
time_str = 'an hour ago'
elif (days < 1) & (hours < 24) & (mins == 0):
time_str = 'Just in'
else:
time_str = f'{str(mins)} minutes ago'
if mins == 1:
time_str = 'a minute ago'
return time_str
except:
return "-"
def fetch_from_db(fetch_flag):
try:
logging.warning(f'fetch_flag: {fetch_flag}')
if fetch_flag == 1:
final_df = db.read_news_from_db()
freq_tokens = get_frequent_words_html(final_df)
logging.warning('Fetched From DB\n\n')
final_df['_id'] = final_df['_id'].astype('str')
redis_client.set("NEWSDF", final_df.to_json())
redis_client.set("NEWSWORDCLOUD", freq_tokens)
else:
final_df = pd.read_json(redis_client.get("NEWSDF"))
freq_tokens = redis_client.get("NEWSWORDCLOUD")
logging.warning('Fetched From Cache\n\n')
except Exception as e:
print(e)
final_df = []
freq_tokens = ""
raise
return final_df, freq_tokens
@app.route("/")
@cross_origin()
def index():
"""
Entry point
"""
try:
src_str = ''
final_df, freq_tokens = fetch_from_db(is_db_fetch_reqd())
if len(final_df) > 1:
final_df["parsed_date"] = [correct_date(date_) for date_ in final_df['parsed_date']]
final_df["parsed_date"] = [parser.parse(date_) for date_ in final_df['parsed_date']]
final_df["elapsed_time"] =[date_time_parser(date_) for date_ in final_df['parsed_date']]
final_df = final_df.loc[final_df["elapsed_time"] <= NEWS_RETENTION_SECONDS, :].copy()
final_df["elapsed_time_str"] = final_df["elapsed_time"].apply(elapsed_time_str)
final_df.sort_values(by="elapsed_time", inplace=True)
src_str = ", ".join(sorted([*final_df['src'].unique()]))
final_df['src_time'] = final_df['src'] + (" " * 5) + final_df["elapsed_time_str"]
final_df.drop(columns=['_id', 'parsed_date', 'src', 'elapsed_time', 'elapsed_time_str'], inplace=True)
final_df.drop_duplicates(subset='description', inplace=True)
final_df = final_df.loc[(final_df["title"] != ""), :].copy()
else:
final_df = pd.DataFrame({'title': '', 'url': '',
'description': '', 'src_time': ''}, index=[0])
except Exception as e:
final_df = pd.DataFrame({'title': '', 'url': '',
'description': '', 'src_time': ''}, index=[0])
logging.warning(traceback.print_exc())
result_str = f'''
<div class="box" id="main">
<form>
<div class="banner">
<img src="../static/favicon_new.png" class="logo-img" alt="KSV Muralidhar" />
<h1 style="display:inline-block; vertical-align: middle;">Latest News</h1>
</div>
'''
if len(final_df) <= 1:
result_str += f'''<div><p class="unavailable">This app is temporarily unavailable</p></div>'''
else:
# last_update_utc = datetime.strptime(os.getenv("NEWSFETCHTIME"), '%Y-%m-%d %H:%M:%S.%f')
last_update_utc = datetime.strptime(redis_client.get('NEWSFETCHTIME'), '%Y-%m-%d %H:%M:%S.%f')
last_update_utc = last_update_utc.strftime("%Y-%m-%d %H:%M:%S")
result_str += f'<p class="srctxt">News aggregated from <b>{src_str}</b>.<br><br>Last updated: {last_update_utc} UTC</p>'
result_str += '''
<div class="input-container">
<input type="text" class="keyword-input" id="keywordInput" placeholder="Search" oninput="filterContent(true)">
<div class="clear-btn" id="clearBtn" onclick="clearFilter()">×</div>
</div>
'''
result_str += f"{freq_tokens} "
result_str += '<div class="show-more-word-cloud" onclick=word_cloud_display()><p class="three-dots">...</p></div>'
result_str += '''<div style="padding-bottom: 10px; font-size: 12px; font-family: Arial, Helvetica, sans-serif;">
News categories and similar news are AI-generated</div>'''
for n, i in final_df.iterrows(): # iterating through the search results
href = i["url"]
category = i["category"]
description = i["description"]
url_txt = i["title"]
src_time = i["src_time"]
sim_news = i['similar_news']
result_str += f'''<div class="news-item"><div style="padding-top: 7px;">
<a href="{href}" target="_blank" class="article-category">{category}
</a>
</div>
<div>
<a href="{href}" target="_blank" class="headline">{url_txt}
</a>
</div>
<div>
<a href="{href}" target="_blank" class="description">
{description}
</a>
</div>
<div>
<a href="{href}" target="_blank" class="time">
{src_time}
</a>
</div>
<div class="container">
<div class="content" style="display: none;">
{sim_news}
</div>
<div class="show-similar-button-container">
<button type="button" class="show-more">Show similar news</button>
<button type="button" class="show-less">Hide similar news</button>
</div>
</div>
<div>
<p></p>
</div></div>
'''
result_str += '</form></div>'
return render_template("index.html", body=result_str)
if __name__ == "__main__":
app.run(host="0.0.0.0", port=7860, workers=5, threads=5) # workers=(2*ncores) + 1, threads= (2 to 4*ncores) + 1
|