def format(self, article, subscriber):
try:
docs = []
for category in article.get('anpa_category'):
pub_seq_num = superdesk.get_resource_service(
'subscribers').generate_sequence_number(subscriber)
anpa = []
# selector codes are only injected for those subscribers that are defined
# in the mapper
selectors = dict()
SelectorcodeMapper().map(article,
category.get('qcode').upper(),
subscriber=subscriber,
formatted_item=selectors)
if 'selector_codes' in selectors and selectors[
'selector_codes']:
anpa.append(b'\x05')
anpa.append(selectors['selector_codes'].encode('ascii'))
anpa.append(b'\x0D\x0A')
# start of message header (syn syn soh)
anpa.append(b'\x16\x16\x01')
anpa.append(
article.get('service_level', 'a').lower().encode('ascii'))
# story number
anpa.append(str(pub_seq_num).zfill(4).encode('ascii'))
# field seperator
anpa.append(b'\x0A') # -LF
anpa.append(
map_priority(article.get('priority')).encode('ascii'))
anpa.append(b'\x20')
anpa.append(category['qcode'].encode('ascii'))
anpa.append(b'\x13')
# format identifier
if article[ITEM_TYPE] == CONTENT_TYPE.PREFORMATTED:
anpa.append(b'\x12')
else:
anpa.append(b'\x11')
anpa.append(b'\x20')
# keyword
keyword = 'bc-{}'.format(
self.append_legal(article=article,
truncate=True)).replace(' ', '-')
keyword = keyword[:24] if len(keyword) > 24 else keyword
anpa.append(keyword.encode('ascii'))
anpa.append(b'\x20')
# version field
anpa.append(b'\x20')
# reference field
anpa.append(b'\x20')
# filing date
anpa.append('{}-{}'.format(
article['_updated'].strftime('%m'),
article['_updated'].strftime('%d')).encode('ascii'))
anpa.append(b'\x20')
# add the word count
anpa.append(
str(article.get('word_count',
'0000')).zfill(4).encode('ascii'))
anpa.append(b'\x0D\x0A')
anpa.append(b'\x02') # STX
self._process_headline(anpa, article,
category['qcode'].encode('ascii'))
keyword = self.append_legal(article=article,
truncate=True).encode(
'ascii', 'ignore')
anpa.append(keyword)
take_key = article.get('anpa_take_key',
'').encode('ascii', 'ignore')
anpa.append((b'\x20' + take_key) if len(take_key) > 0 else b'')
anpa.append(b'\x0D\x0A')
if BYLINE in article:
anpa.append(article.get(BYLINE).encode('ascii', 'ignore'))
anpa.append(b'\x0D\x0A')
if article.get('dateline', {}).get('text'):
anpa.append(
article.get('dateline').get('text').encode(
'ascii', 'ignore'))
body = self.append_body_footer(article)
if article.get(EMBARGO):
embargo = '{}{}'.format(
'Embargo Content. Timestamp: ',
get_utc_schedule(article, EMBARGO).isoformat())
body = embargo + body
if article[ITEM_TYPE] == CONTENT_TYPE.PREFORMATTED:
anpa.append(body.encode('ascii', 'replace'))
else:
anpa.append(
BeautifulSoup(body, "html.parser").text.encode(
'ascii', 'replace'))
anpa.append(b'\x0D\x0A')
if article.get('more_coming', False):
anpa.append('MORE'.encode('ascii'))
else:
anpa.append(article.get('source', '').encode('ascii'))
sign_off = article.get('sign_off', '').encode('ascii')
anpa.append((b'\x20' + sign_off) if len(sign_off) > 0 else b'')
anpa.append(b'\x0D\x0A')
anpa.append(b'\x03') # ETX
# time and date
anpa.append(datetime.datetime.now().strftime(
'%d-%m-%y %H-%M-%S').encode('ascii'))
anpa.append(b'\x04') # EOT
anpa.append(
b'\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A'
)
docs.append((pub_seq_num, b''.join(anpa)))
return docs
except Exception as ex:
raise FormatterError.AnpaFormatterError(ex, subscriber)
def format(self, article, subscriber, codes=None):
try:
docs = []
formatted_article = deepcopy(article)
for category in self._get_category_list(
formatted_article.get('anpa_category')):
mapped_source = self._get_mapped_source(formatted_article)
formatted_article[config.ID_FIELD] = formatted_article.get(
'item_id', formatted_article.get(config.ID_FIELD))
pub_seq_num = superdesk.get_resource_service(
'subscribers').generate_sequence_number(subscriber)
anpa = []
if codes:
anpa.append(b'\x05')
anpa.append(' '.join(codes).encode('ascii'))
anpa.append(b'\x0D\x0A')
# start of message header (syn syn soh)
anpa.append(b'\x16\x16\x01')
anpa.append(
get_service_level(category,
formatted_article).encode('ascii'))
# story number
anpa.append(str(pub_seq_num).zfill(4).encode('ascii'))
# field seperator
anpa.append(b'\x0A') # -LF
anpa.append(
map_priority(
formatted_article.get('priority')).encode('ascii'))
anpa.append(b'\x20')
anpa.append(category['qcode'].lower().encode('ascii'))
anpa.append(b'\x13')
# format identifier
if formatted_article.get(FORMAT,
FORMATS.HTML) == FORMATS.PRESERVED:
anpa.append(b'\x12')
else:
anpa.append(b'\x11')
anpa.append(b'\x20')
# keyword
keyword = 'bc-{}'.format(
self.append_legal(article=formatted_article,
truncate=True)).replace(' ', '-')
keyword = keyword[:24] if len(keyword) > 24 else keyword
anpa.append(keyword.encode('ascii'))
anpa.append(b'\x20')
# version field
anpa.append(b'\x20')
# reference field
anpa.append(b'\x20')
# filing date
anpa.append('{}-{}'.format(
formatted_article['_updated'].strftime('%m'),
formatted_article['_updated'].strftime('%d')).encode(
'ascii'))
anpa.append(b'\x20')
# add the word count
anpa.append(
str(formatted_article.get(
'word_count', '0000')).zfill(4).encode('ascii'))
anpa.append(b'\x0D\x0A')
anpa.append(b'\x02') # STX
self._process_headline(anpa, formatted_article,
category['qcode'].encode('ascii'))
keyword = SluglineMapper().map(
article=formatted_article,
category=category['qcode'].upper(),
truncate=True).encode('ascii', 'ignore')
anpa.append(keyword)
take_key = (formatted_article.get('anpa_take_key', '')
or '').encode('ascii', 'ignore')
anpa.append((b'\x20' + take_key) if len(take_key) > 0 else b'')
anpa.append(b'\x0D\x0A')
if formatted_article.get('ednote', '') != '':
ednote = '{}\r\n'.format(
to_ascii(formatted_article.get('ednote')))
anpa.append(ednote.encode('ascii', 'replace'))
if formatted_article.get(BYLINE):
anpa.append(
get_text(formatted_article.get(BYLINE)).encode(
'ascii', 'replace'))
anpa.append(b'\x0D\x0A')
if formatted_article.get(FORMAT) == FORMATS.PRESERVED:
anpa.append(
get_text(self.append_body_footer(formatted_article),
content='html').encode('ascii', 'replace'))
else:
body = to_ascii(formatted_article.get('body_html', ''))
# we need to inject the dateline
if formatted_article.get(
'dateline', {}).get('text') and not article.get(
'auto_publish', False):
body_html_elem = parse_html(
formatted_article.get('body_html'))
ptag = body_html_elem.find('.//p')
if ptag is not None:
ptag.text = formatted_article['dateline'][
'text'] + ' ' + (ptag.text or '')
body = to_string(body_html_elem)
anpa.append(self.get_text_content(body))
if formatted_article.get('body_footer'):
anpa.append(
self.get_text_content(
to_ascii(
formatted_article.get('body_footer', ''))))
anpa.append(b'\x0D\x0A')
anpa.append(mapped_source.encode('ascii'))
sign_off = (formatted_article.get('sign_off', '')
or '').encode('ascii')
anpa.append((b'\x20' + sign_off) if len(sign_off) > 0 else b'')
anpa.append(b'\x0D\x0A')
anpa.append(b'\x03') # ETX
# time and date
anpa.append(datetime.datetime.now().strftime(
'%d-%m-%y %H-%M-%S').encode('ascii'))
anpa.append(b'\x04') # EOT
anpa.append(
b'\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A\x0D\x0A'
)
docs.append({
'published_seq_num': pub_seq_num,
'encoded_item': b''.join(anpa),
'formatted_item': b''.join(anpa).decode('ascii')
})
return docs
except Exception as ex:
raise FormatterError.AnpaFormatterError(ex, subscriber)
