#!./env/bin/python # -*- coding: utf-8 -*- from datetime import datetime import collections from feedgen.feed import FeedGenerator import glob import json from lxml import etree import re import os.path I2P_OS = os.getenv("I2P_OS", "") I2P_BRANCH = os.getenv("I2P_BRANCH", "") DATA_DIR = os.path.join('data') RELEASE_DIR = os.path.join(DATA_DIR, I2P_OS, I2P_BRANCH) ENTRIES_FILE = os.path.join(DATA_DIR, 'entries.html') PLATFORM_ENTRIES_FILE = os.path.join(DATA_DIR, I2P_OS, I2P_BRANCH, 'entries.html') TRANSLATED_ENTRIES_FILES = os.path.join(DATA_DIR, 'translations/entries.*.html') TRANSLATED_PLATFORM_ENTRIES_FILES = os.path.join(DATA_DIR, I2P_OS, I2P_BRANCH, 'translations/entries.*.html') RELEASES_FILE = os.path.join(RELEASE_DIR, 'releases.json') CRL_FILES = os.path.join(DATA_DIR, 'crls/*.crl') BLOCKLIST_FILE = os.path.join(DATA_DIR, 'blocklist.xml') BUILD_DIR = os.path.join('build', I2P_OS, I2P_BRANCH) NEWS_FILE = os.path.join(BUILD_DIR, 'news.atom.xml') TRANSLATED_NEWS_FILE = os.path.join(BUILD_DIR, 'news_%s.atom.xml') def load_feed_metadata(fg): fg.id('urn:uuid:60a76c80-d399-11d9-b91C-543213999af6') fg.link( href='http://i2p-projekt.i2p/' ) fg.link( href='http://tc73n4kivdroccekirco7rhgxdg5f3cjvbaapabupeyzrqwv5guq.b32.i2p/news.atom.xml', rel='self' ) fg.link( href='http://dn3tvalnjz432qkqsvpfdqrwpqkw3ye4n4i2uyfr4jexvo3sp5ka.b32.i2p/news/news.atom.xml', rel='alternate' ) def load_entries(fg, entries_file, platform_entries_file=None): metadatas = {} finalentries = {} print('Loading entries from %s' % entries_file) entries = prepare_entries_file(fg, entries_file) # split() creates a junk final element with trailing for entry_str in entries[:-1]: entry_parts = entry_str.split('', 1) md = extract_entry_metadata(entry_parts[0]) metadatas[md['published']] = md finalentries[md['id']] = entry_parts[1] if os.path.exists(platform_entries_file) and platform_entries_file != entries_file and platform_entries_file is not None and platform_entries_file != "data/entries.html": print('Loading platform entries from %s' % platform_entries_file) entries = prepare_entries_file(fg, platform_entries_file) for entry_str in entries[:-1]: entry_parts = entry_str.split('', 1) md = extract_entry_metadata(entry_parts[0]) metadatas[md['updated']] = md finalentries[md['id']] = entry_parts[1] sorted_metadata = collections.OrderedDict(sorted(metadatas.items())) for metadata in sorted_metadata.values(): fe = fg.add_entry() fe.id(metadata['id']) fe.title(metadata['title']) fe.summary(metadata['summary']) fe.link( href=metadata['href'] ) fe.author( name=metadata['author'] ) fe.published(metadata['published']) fe.updated(metadata['updated']) fe.content(finalentries[metadata['id']], type='xhtml') def prepare_entries_file(fg, entries_file=None): with open(entries_file) as f: entries_data = f.read().strip('\n') # Replace HTML non-breaking space with unicode entries_data = entries_data.replace(' ', '\u00a0') # Strip the leading
from translations if entries_data.startswith('
'): entries_data = entries_data[5:] entries_parts = entries_data.split('') fg.title(re.findall(r'title="(.*?)"', entries_parts[0])[0]) fg.subtitle(entries_parts[0].split('>')[1]) entries = entries_parts[1].split('') return entries def extract_entry_metadata(s): m = {k:v.strip('"') for k,v in re.findall(r'(\S+)=(".*?"|\S+)', s)} summary = re.findall(r'(.*)', s, re.DOTALL) if len(summary) > 0: m['summary'] = summary[0] return m def load_releases(fg): with open(RELEASES_FILE) as json_data: d = json.load(json_data) for release in d: r = fg.i2p.add_release() r.date(release['date']) r.version(release['version']) if 'minVersion' in release: r.min_version(release['minVersion']) if 'minJavaVersion' in release: r.min_java_version(release['minJavaVersion']) for update_type, update in release['updates'].items(): u = r.add_update(update_type) if 'clearnet' in update: for url in update['clearnet']: u.clearnet(url) if 'clearnetssl' in update: for url in update['clearnetssl']: u.clearnetssl(url) if 'torrent' in update: u.torrent(update['torrent']) if 'url' in update: for url in update['url']: u.url(url) def load_revocations(fg): # Only add a revocations element if there are CRLs r = None for crl in glob.glob(CRL_FILES): if r is None: r = fg.i2p.add_revocations() crl_id = os.path.splitext(os.path.basename(crl))[0] c = r.add_crl(crl_id) c.updated(datetime.fromtimestamp(os.path.getmtime(crl))) with open(crl) as f: crl_content = f.read().decode('utf8').strip() c.content('\n%s\n' % crl_content) def load_blocklist(fg): # Only add a blocklist element if there is content b = None if os.path.isfile(BLOCKLIST_FILE): with open(BLOCKLIST_FILE) as f: content = '%s' % f.read() root = etree.fromstring(content) b = fg.i2p.add_blocklist() b.from_xml(root.getchildren()[0]) def generate_feed(entries_file=None, platform_entries_file=None): language = entries_file and entries_file.split('.')[1] or 'en' fg = FeedGenerator() fg.load_extension('i2p') fg.language(language) load_feed_metadata(fg) load_entries(fg, entries_file and entries_file or ENTRIES_FILE, platform_entries_file and platform_entries_file or PLATFORM_ENTRIES_FILE) load_releases(fg) load_revocations(fg) load_blocklist(fg) if not os.path.exists(BUILD_DIR): os.makedirs(BUILD_DIR) fg.atom_file(entries_file and TRANSLATED_NEWS_FILE % language or NEWS_FILE, pretty=True) if __name__ == '__main__': # Standard feed generate_feed() # Translated feeds for entries_file in glob.glob(TRANSLATED_ENTRIES_FILES): generate_feed(entries_file)