| 1 |
#!/usr/bin/env python |
| 2 |
# vim:fileencoding=utf-8 |
| 3 |
|
| 4 |
import base64 |
| 5 |
import json |
| 6 |
import re |
| 7 |
from collections import defaultdict |
| 8 |
from datetime import datetime |
| 9 |
from html import escape |
| 10 |
|
| 11 |
from calibre.web.feeds.news import BasicNewsRecipe |
| 12 |
|
| 13 |
keys = [ |
| 14 |
'ZXlKaGJHY2lPaUpTVXpJMU5pSjk=', |
| 15 |
'V0Zac2FITjNNWGQzU213MFYza3dSWEJ6Y2xR', |
| 16 |
'cXd3QmVkQVVOWEhUUWNob3dRWjV6TXdtblhxREtlTWhvUkpsa0I3ZHJqV21iMGt0WkNTY0locTVscElpV2FNeU5KQQ==', |
| 17 |
'T0RZZ0hBZklvaTdES1drUzhnOEd1bkZOQVhwSkRVT0xkSTJydFFrVEVp', |
| 18 |
'RTNvOTByZFpIdW5QUjdwMFVMalJtSENuRG9mQWRwVFFkSnRUWGpROWVFRFpUMnhvb29WR2RCcG9WS2hF', |
| 19 |
] |
| 20 |
|
| 21 |
|
| 22 |
def _content_to_html(content): |
| 23 |
html = '' |
| 24 |
for node in content or []: |
| 25 |
if not isinstance(node, dict): |
| 26 |
continue |
| 27 |
text = escape(node.get('text', '')) |
| 28 |
if not text: |
| 29 |
continue |
| 30 |
if node.get('type') == 'link' and node.get('uri'): |
| 31 |
html += f'<a href="{node["uri"]}">{text}</a>' |
| 32 |
else: |
| 33 |
html += text |
| 34 |
return html |
| 35 |
|
| 36 |
|
| 37 |
def _body_to_html(body): |
| 38 |
html = '' |
| 39 |
for item in body or []: |
| 40 |
if not isinstance(item, dict): |
| 41 |
continue |
| 42 |
typ = item.get('type', '') |
| 43 |
if typ == 'paragraph': |
| 44 |
text = _content_to_html(item.get('content')) |
| 45 |
if text: |
| 46 |
html += f'<p>{text}</p>\n' |
| 47 |
elif typ in ('heading', 'subheading'): |
| 48 |
text = _content_to_html(item.get('content')) |
| 49 |
if text: |
| 50 |
html += f'<h3>{text}</h3>\n' |
| 51 |
elif typ == 'image': |
| 52 |
src = item.get('url') or item.get('imageUrl', '') |
| 53 |
cap_raw = item.get('caption', '') |
| 54 |
cap = _content_to_html(cap_raw) if isinstance(cap_raw, list) else escape(str(cap_raw or '')) |
| 55 |
if src: |
| 56 |
html += f'<figure><img src="{src}"/>' |
| 57 |
if cap: |
| 58 |
html += f'<figcaption>{cap}</figcaption>' |
| 59 |
html += '</figure>\n' |
| 60 |
elif typ == 'list': |
| 61 |
tag = 'ol' if item.get('ordered') else 'ul' |
| 62 |
html += f'<{tag}>\n' |
| 63 |
for li in item.get('items') or item.get('content') or []: |
| 64 |
if isinstance(li, dict): |
| 65 |
html += f'<li>{_content_to_html(li.get("content"))}</li>\n' |
| 66 |
html += f'</{tag}>\n' |
| 67 |
return html |
| 68 |
|
| 69 |
|
| 70 |
class Barrons(BasicNewsRecipe): |
| 71 |
title = "Barron's Magazine" |
| 72 |
__author__ = 'unkn0wn' |
| 73 |
description = ( |
| 74 |
"Barron's is an American weekly magazine/newspaper published by Dow Jones & Company. Founded in 1921 as a sister " |
| 75 |
"publication to The Wall Street Journal, Barron's covers U.S. financial information, market developments, and " |
| 76 |
'relevant statistics.' |
| 77 |
) |
| 78 |
language = 'en_US' |
| 79 |
encoding = 'utf-8' |
| 80 |
ignore_duplicate_articles = {'url'} |
| 81 |
masthead_url = 'https://www.barrons.com/asset/barrons/images/barrons-logo.png' |
| 82 |
resolve_internal_links = True |
| 83 |
delay = 1 |
| 84 |
|
| 85 |
extra_css = ''' |
| 86 |
.auth { font-size:small; } |
| 87 |
.caption { font-size:small; text-align:center; } |
| 88 |
img { display:block; margin:0 auto; } |
| 89 |
''' |
| 90 |
|
| 91 |
recipe_specific_options = { |
| 92 |
'date': { |
| 93 |
'short': 'The date of the edition to download (MMM DD, YYYY format)', |
| 94 |
'long': 'For example, Dec 30, 2024', |
| 95 |
}, |
| 96 |
} |
| 97 |
|
| 98 |
def preprocess_raw_html(self, raw, url): |
| 99 |
html = raw if isinstance(raw, str) else raw.decode('utf-8', errors='ignore') |
| 100 |
nd_match = re.search(r'<script id="__NEXT_DATA__"[^>]*>(.*?)</script>', html, re.DOTALL) |
| 101 |
if not nd_match: |
| 102 |
return html |
| 103 |
try: |
| 104 |
j = json.loads(nd_match.group(1)) |
| 105 |
pp = j['props']['pageProps'] |
| 106 |
art = pp.get('articleData', {}) |
| 107 |
|
| 108 |
headline_text = art.get('headline', {}) |
| 109 |
if isinstance(headline_text, dict): |
| 110 |
headline_text = headline_text.get('text', '') |
| 111 |
|
| 112 |
byline = art.get('byline', []) |
| 113 |
byline_text = ''.join(x.get('text', '') for x in byline if isinstance(x, dict)).strip() |
| 114 |
|
| 115 |
# Subscribers get full body; non-subscribers get snippet (first para only) |
| 116 |
body = art.get('body') or pp.get('snippet', []) |
| 117 |
|
| 118 |
out = '<html><body>\n' |
| 119 |
if headline_text: |
| 120 |
out += f'<h1>{escape(headline_text)}</h1>\n' |
| 121 |
if byline_text: |
| 122 |
out += f'<div class="auth">{escape(byline_text)}</div>\n' |
| 123 |
out += _body_to_html(body) |
| 124 |
out += '</body></html>' |
| 125 |
return out |
| 126 |
except Exception: |
| 127 |
return html |
| 128 |
|
| 129 |
def parse_index(self): |
| 130 |
# Step 1: Use mobile API for archive listing (cover image + edition dates) |
| 131 |
index = 'https://barrons.djmedia.djservices.io' |
| 132 |
theatre = '/apps/barrons/theaters/' |
| 133 |
archive = self.index_to_soup( |
| 134 |
index + theatre + 'magazine-archive?screen_ids=magazine-archive', |
| 135 |
raw=True, |
| 136 |
) |
| 137 |
scrn = json.loads(archive)['screens'][0]['frames'] |
| 138 |
self.log( |
| 139 |
'Available Editions: ', |
| 140 |
' | '.join(x['screenIds'][0]['name'][6:] for x in scrn), |
| 141 |
) |
| 142 |
|
| 143 |
selected = None |
| 144 |
for frme in scrn: |
| 145 |
edition_date = self.recipe_specific_options.get('date') |
| 146 |
if edition_date and isinstance(edition_date, str): |
| 147 |
nme = frme['screenIds'][0]['name'] |
| 148 |
if edition_date.lower() not in nme.lower(): |
| 149 |
continue |
| 150 |
selected = frme |
| 151 |
break |
| 152 |
|
| 153 |
if selected is None: |
| 154 |
selected = scrn[0] |
| 155 |
|
| 156 |
nme = selected['screenIds'][0]['name'] |
| 157 |
self.cover_url = selected['image']['url'] |
| 158 |
self.log('Downloading ', nme) |
| 159 |
self.timefmt = ' [' + nme[6:] + ']' |
| 160 |
|
| 161 |
# Convert mobile API date to website URL |
| 162 |
date_text = selected['date']['text'] # e.g. "2026-08-31T04:00:00Z" |
| 163 |
dt = datetime.strptime(date_text[:10], '%Y-%m-%d') |
| 164 |
date_str = dt.strftime('%Y%m%d') |
| 165 |
|
| 166 |
# Step 2: Fetch article list from website __NEXT_DATA__ |
| 167 |
web_url = 'https://www.barrons.com/magazine/' + date_str |
| 168 |
raw_html = self.index_to_soup(web_url, raw=True).decode('utf-8', errors='ignore') |
| 169 |
nd_match = re.search(r'<script id="__NEXT_DATA__"[^>]*>(.*?)</script>', raw_html, re.DOTALL) |
| 170 |
if not nd_match: |
| 171 |
raise ValueError('Could not find article data in Barrons magazine page') |
| 172 |
|
| 173 |
mag_data = json.loads(nd_match.group(1))['props']['pageProps']['magazineData'] |
| 174 |
|
| 175 |
sections_map = [ |
| 176 |
('coverArticle', 'Cover Story'), |
| 177 |
('featureArticles', 'Features'), |
| 178 |
('columnsArticles', 'Columns'), |
| 179 |
('departmentsArticles', 'Departments'), |
| 180 |
('restArticles', 'More'), |
| 181 |
] |
| 182 |
|
| 183 |
feeds_dict = defaultdict(list) |
| 184 |
for key, section_name in sections_map: |
| 185 |
for art in mag_data.get(key, []): |
| 186 |
if not isinstance(art, dict): |
| 187 |
continue |
| 188 |
url = art.get('url', '') |
| 189 |
if 'barrons.com/articles/' not in url: |
| 190 |
continue |
| 191 |
url = url.split('?')[0] |
| 192 |
title = art.get('headline', '') |
| 193 |
if not title or not url: |
| 194 |
continue |
| 195 |
desc = art.get('summary', '') or '' |
| 196 |
self.log(' ', title, '\n\t', desc) |
| 197 |
feeds_dict[section_name].append({'title': title, 'url': url, 'description': desc}) |
| 198 |
|
| 199 |
return list(feeds_dict.items()) |
| 200 |
|
| 201 |
def get_browser(self, *args, **kw): |
| 202 |
kw['user_agent'] = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36' |
| 203 |
br = BasicNewsRecipe.get_browser(self, *args, **kw) |
| 204 |
k2 = '.'.join(base64.b64decode(b).decode() for b in keys[:3]) |
| 205 |
k3 = '_'.join(base64.b64decode(v).decode() for v in keys[3:]) |
| 206 |
br.addheaders += [ |
| 207 |
('Accept-Encoding', 'gzip'), |
| 208 |
('Accept-Language', 'en-US,en;q=0.9'), |
| 209 |
('Sec-Ch-Ua', '"Not_A Brand";v="8", "Chromium";v="120", "Google Chrome";v="120"'), |
| 210 |
('Sec-Ch-Ua-Mobile', '?0'), |
| 211 |
('Sec-Ch-Ua-Platform', '"Windows"'), |
| 212 |
('Sec-Fetch-Dest', 'document'), |
| 213 |
('Sec-Fetch-Mode', 'navigate'), |
| 214 |
('Sec-Fetch-Site', 'none'), |
| 215 |
('Sec-Fetch-User', '?1'), |
| 216 |
# Mobile API token (used only for the archive listing) |
| 217 |
('App-Identifier', 'com.news.screens'), |
| 218 |
('App-Version', '1'), |
| 219 |
('Device-Type', 'phone'), |
| 220 |
('Os-Name', 'Android'), |
| 221 |
('X-Access-Token', k2 + '_' + k3), |
| 222 |
] |
| 223 |
return br |
| 224 |
|
| 225 |
def populate_article_metadata(self, article, soup, first): |
| 226 |
if '/articles/' not in article.url: |
| 227 |
article.url = 'https://www.barrons.com/articles/' + article.url.split('=')[-1] |