def _pl_usembassy_gov_path(url):
if not urlpath(url).startswith('/pl/'):
return False
else:
if urlpath(url) == '/pl/':
return False
elif urlpath(url).startswith('/pl/category/'):
return False
elif urlpath(url).startswith('/pl/tag/'):
return False
else:
return True
def crawl_azattyk_org(crawler, out):
sitemap = crawler.fetch_sitemap('https://www.azattyk.org/sitemap.xml')
for url in sorted(sitemap.keys()):
if not urlpath(url).startswith('/a/'):
continue
doc = crawler.fetch(url)
if doc.status != 200:
continue
html = doc.content.decode('utf-8')
pubdate = re.search(r'"dateModified":"([^"]+)"', html)
if pubdate is not None:
pubdate = cleantext(pubdate.group(1)).replace(' ', 'T')
title = extract('<title>', '</title>', html)
text = extract('content-offset">', '</div>', html)
if not title or not text:
continue
paras = [title] + re.sub(r'<br\s*?/?>', '\n', text).splitlines()
paras = filter(None, [cleantext(p) for p in paras])
paras = [p for p in paras if not p.startswith('http')]
if not paras:
continue
out.write('# Location: %s\n' % url)
out.write('# Genre: News\n')
if pubdate:
out.write('# Publication-Date: %s\n' % pubdate)
for p in paras:
out.write(p + '\n')
def crawl_azattyk_org(crawler, out):
sitemap = crawler.fetch_sitemap('https://www.azattyk.org/sitemap.xml')
for url in sorted(sitemap.keys()):
if not urlpath(url).startswith('/a/'):
continue
doc = crawler.fetch(url)
if doc.status != 200:
continue
html = doc.content.decode('utf-8')
pubdate = re.search(r'"dateModified":"([^"]+)"', html)
if pubdate is not None:
pubdate = cleantext(pubdate.group(1)).replace(' ', 'T')
title = extract('<title>', '</title>', html)
text = extract('content-offset">', '<footer', html)
if not title or not text:
continue
text = text.split('<span class="share')[0]
text = text.split('<div class="region"')[0]
text = text.replace('\n', ' ')
paras = [title] + re.sub(r'<(?:br|p|div)\s*?/?>', '\n', text).splitlines()
paras = filter(None, [cleantext(p.strip()) for p in paras])
paras = [p for p in paras if not p.startswith('http')]
if not paras:
continue
# Filter out English text.
if ord(paras[0][0]) <= 0xFF or ord(paras[-1][-1]) <= 0xFF:
continue
out.write('# Location: %s\n' % url)
out.write('# Genre: News\n')
if pubdate:
out.write('# Publication-Date: %s\n' % pubdate)
for p in paras:
out.write(p + '\n')
def crawl_pl_usembassy_gov(crawler, out):
sitemap = crawler.fetch_sitemap('https://pl.usembassy.gov/sitemap_index.xml')
trans_regex = re.compile(
r'<h3>Tłumaczenie</h3><div class="translations_sidebar"><ul><li><a href ?="([^"]*)"'
)
pubdate_regex = re.compile(
r'<meta property="article:published_time" content="([^"]*)"'
)
links = set()
for key in sorted(sitemap.keys()):
if _pl_usembassy_gov_path(key):
links.add(key)
for link in sorted(links):
result = crawler.fetch(link)
if result.status != 200:
continue
html = result.content.decode('utf-8')
title = extract('<title>', '</title>', html)
title = title if title else ''
title = title.split(' | ')[0] if ' | ' in title else title
pubdate_match = pubdate_regex.search(html)
pubdate = pubdate_match.group(1) if pubdate_match else None
trans_match = trans_regex.search(html)
trans = trans_match.group(1) if trans_match else None
if pubdate is None: pubdate = result.headers.get('Last-Modified')
if pubdate is None: pubdate = sitemap[link]
exstart = '<div class="entry-content">'
exstart2 = '<div class="mo-page-content">'
exend = '<!-- AddThis Advanced Settings above via filter on the_content -->'
exstart = exstart2 if exstart2 in html else exstart
content = extract(exstart, exend, html)
cleanparas = clean_paragraphs(content) if content else None
# Don't repeat the title if it's the only text content
cleantitle = cleantext(title)
if cleanparas:
if len(cleanparas) == 1 and cleanparas[0] == cleantitle:
paras = [cleantitle]
else:
paras = [cleantitle] + cleanparas
else:
paras = [cleantitle]
# There are quite a few media pages whose only text is the filename
# this, conveniently, is typically also the post's name
if len(paras) == 1 and paras[0].lower() in urlpath(link).lower():
continue
if paras:
out.write('# Location: %s\n' % link)
out.write('# Genre: Diplomatic\n')
if trans:
out.write('# Translation: %s\n' % trans)
if pubdate:
out.write('# Publication-Date: %s\n' % pubdate)
out.write('\n'.join(paras) + '\n')
def crawl_dimma_fo(crawler, out):
num_pages = int(re.search(
r'<a href="http://www.dimma.fo/(\d+)" class="to-last"',
crawler.fetch('http://www.dimma.fo/').content).group(1))
urls = set()
for i in range(1, num_pages + 1):
doc = crawler.fetch('http://www.dimma.fo/%d' % i)
html = doc.content.decode('utf-8')
for u in re.findall(r'href="(http://www.dimma.fo/[^"]+?)"', html):
path = urlpath(u)
if re.match(r'/\d+', path) or u'/' in path[1:]:
continue
urls.add(u)
for url in sorted(urls):
doc = crawler.fetch(urlencode(url))
if doc.status != 200:
continue
html = doc.content.decode('utf-8')
content = html.split('class="content">')[1]
pubdate = re.search(
r'<span class="date">\s*'
r'(\d{4}-\d{2}-\d{2})\s+(\d{2}:\d{2})\s*</span>',
content)
if pubdate != None:
pubdate = '%sT%s:00+01:00' % (pubdate.group(1), pubdate.group(2))
paragraphs = []
title = re.search(r'<h1>(.+?)</h1>', html, flags=re.DOTALL)
if title != None:
paragraphs.append(cleantext(title.group(1)))
text = content.split('<p>', 1)[1].split('</div>')[0]
text = text.replace('\n', ' ').replace('</p>', '\n')
text = text.replace('<br />', '\n')
paragraphs.extend([cleantext(p) for p in text.splitlines()])
paragraphs = filter(None, paragraphs)
if paragraphs:
out.write('# Location: %s\n' % url)
out.write('# Genre: News\n')
if pubdate:
out.write('# Publication-Date: %s\n' % pubdate)
for p in paragraphs:
out.write(p + '\n')
def _rtenuacht_path(url):
rtenuacht = urlpath(url).startswith('/news/nuacht/')
rnagnuacht = urlpath(url).startswith('/rnag/nuacht-gaeltachta')
return rtenuacht or rnagnuacht
def _rtenuacht_path(url):
rtenuacht = urlpath(url).startswith('/news/nuacht/')
rnag = '/rnag/nuacht' in url or '/rnag/articles' in url
return rtenuacht or rnag
