def parse_transcript(path):
"""
Parse a Premiere .tsv file, calculate total seconds and write to JSON for use in the app.
"""
data = {
'subtitles': []
}
filename, ext = path.split('/')[-1].split('.')
with codecs.open(path, 'rb', encoding='utf16') as f:
transcript = f.read().encode('utf-8')
tab_reader = csv.reader(BytesIO(transcript), delimiter='\t')
headers = tab_reader.next()
for row in tab_reader:
# Premiere exports kind of suck
if row[0] == '':
words = smartypants(row[1].strip())
time_str = row[2]
else:
words = smartypants(row[0].strip())
time_str = row[1]
hours, minutes, seconds, frame = [int(x) for x in time_str.split(';')]
decimal = (float(frame) / 24)
total_seconds = (hours * 3600) + (minutes * 60) + (seconds + decimal)
segment = {
'time': total_seconds,
'transcript': words,
}
data['subtitles'].append(segment)
with open('www/data/%s.json' % filename, 'w') as wf:
wf.write(json.dumps(data))
def publish_index_pages(self):
file_posts = self.get_all_file_posts_by_date()
posts = [ Post(f, self.s) for f in self.get_all_file_posts_by_date()]
for post in posts:
post.content = smartypants(self.md_to_html(post.content))
post.title = smartypants(post.title)
first_post = 0
last_post = first_post + self.s.POSTS_PER_PAGE
page_number = 0
dest_fname = self.s.INDEX_PAGE
dest_dir = self.s.WWW_DIR
prev_page_url = None
next_page_url = None
while first_post < len(posts):
p = posts[first_post:last_post]
if page_number == 0:
local_fname = dest_fname
else:
local_fname = "%s-%d%s" % (os.path.splitext(dest_fname)[0], page_number, self.s.HTML_EXT )
# Pagination
if len(posts) <= last_post:
# No next page
next_page_url = None
else:
next_page_url = "%s-%d%s" % (os.path.splitext(self.s.INDEX_PAGE)[0], page_number + 1, self.s.HTML_EXT)
if first_post - self.s.POSTS_PER_PAGE < 0:
prev_page_url = None
else:
if page_number == 1:
prev_page_url = self.s.INDEX_PAGE
else:
prev_page_url = "%s-%d.html" % (os.path.splitext(self.s.INDEX_PAGE)[0], page_number - 1)
self.write_posts_to_file(
posts=p,
fname=local_fname,
dir=dest_dir,
template=self.s.INDEX_TEMPLATE,
prev_page_url=prev_page_url,
next_page_url=next_page_url,
)
logging.info("Wrote posts %d-%d to %s." % (first_post, last_post, local_fname))
first_post = last_post
last_post = first_post + self.s.POSTS_PER_PAGE
page_number = page_number + 1
def prepare_post(self, post):
post.title = smartypants(post.title)
post.content = smartypants(markdown.markdown(post.content,
extensions=self.s.MD_EXTENSIONS,
extension_configs=self.s.MD_EXTENSION_CONFIGS,
output_format=self.s.MD_OUTPUT_FORMAT
))
def __init__(self, **kwargs):
self.date = kwargs['date']
self.source_file = kwargs['source_file']
self.summary = smartypants.smartypants(kwargs['summary'])
self.title = smartypants.smartypants(kwargs['title'])
self.route = kwargs['route']
self.url = kwargs['url']
# Having the posts enables a blog post to find its relationships.
self._posts = kwargs['posts']
def __init__(self, **kwargs):
self.date = kwargs["date"]
self.source_file = kwargs["source_file"]
self.summary = smartypants.smartypants(kwargs["summary"])
self.title = smartypants.smartypants(kwargs["title"])
self.route = kwargs["route"]
self.url = kwargs["url"]
# Having the posts enables a blog post to find its relationships.
self._posts = kwargs["posts"]
def __init__(self, **kwargs):
self.date = kwargs['date']
self.source_file = kwargs['source_file']
self.summary = smartypants.smartypants(kwargs['summary'])
self.title = smartypants.smartypants(kwargs['title'])
self.route = kwargs['route']
self.url = kwargs['url']
# Having the posts enables a blog post to find its relationships.
self._posts = kwargs['posts']
def smartquotes(self, text: str) -> str:
"""If enabled, apply 'smart quotes' to the text; replaces quotes and dashes by nicer looking symbols"""
if self.supports_smartquotes and self.do_smartquotes:
if hasattr(smartypants.Attr, "u"):
return smartypants.smartypants(text, smartypants.Attr.q | smartypants.Attr.B |
smartypants.Attr.D | smartypants.Attr.e | smartypants.Attr.u)
else:
# older smartypants lack attribute 'u' for avoiding html entity creation
txt = smartypants.smartypants(text, smartypants.Attr.q | smartypants.Attr.B |
smartypants.Attr.D | smartypants.Attr.e)
import html.parser
return html.parser.unescape(txt) # type: ignore
return text
def make_item_content_encoded(self, text1, text2, url, comment_name):
"""
Called from item_content_encoded() in children.
text1 and text2 are chunks of HTML text (or empty strings).
url is the URL of the item (no domain needed, eg '/diary/1666/10/31/').
comment_name is one of 'comment' or 'annotation'.
"""
return '%s %s <p><strong><a href="%s#%ss">Read the %ss</a></strong></p>' % (
force_unicode(smartypants.smartypants(text1)),
force_unicode(smartypants.smartypants(text2)),
add_domain(Site.objects.get_current().domain, url),
comment_name,
comment_name
)
def render_gist(request, id, raw):
"""Render a raw gist and store it"""
gist = {
'id': raw['id'],
'html_url': raw['html_url'],
'public': raw['public'],
'description': raw['description'],
'created_at': iso8601.parse_date(raw['created_at']),
'updated_at': iso8601.parse_date(raw['updated_at']),
'fetched_at': rdb.now(),
'author_id': raw['user']['id'],
'author_login': raw['user']['login'],
'files': [],
}
for gistfile in raw['files'].values():
format = RENDERABLE.get(gistfile['language'], None)
if format is None:
continue
output = None
if format is FORMAT_MD:
payload = {
'mode': 'gfm',
'text': gistfile['content'],
}
req_render = requests.post('https://api.github.com/markdown',
params=GITHUB_AUTH_PARAMS,
data=unicode(json.dumps(payload)))
if req_render.status_code != 200:
logger.warn('Render {} file {} failed: {}'.format(
id, gistfile['filename'], req_render.status_code))
continue
else:
output = smartypants.smartypants(req_render.text)
if format is FORMAT_RST:
rendered = render_rst(gistfile['content'],
writer_name='html')['fragment']
output = smartypants.smartypants(rendered)
if output is not None:
gistfile['rendered'] = output
gist['files'].append(gistfile)
rdb.table('gists').insert(gist, upsert=True).run(request.rdbconn)
return gist
def render_gist(request, id, raw):
"""Render a raw gist and store it"""
gist = {
'id': raw['id'],
'html_url': raw['html_url'],
'public': raw['public'],
'description': raw['description'],
'created_at': iso8601.parse_date(raw['created_at']),
'updated_at': iso8601.parse_date(raw['updated_at']),
'fetched_at': rdb.now(),
'author_id': raw['user']['id'],
'author_login': raw['user']['login'],
'files': [],
}
for gistfile in raw['files'].values():
format = RENDERABLE.get(gistfile['language'], None)
if format is None:
continue
output = None
if format is FORMAT_MD:
payload = {
'mode': 'gfm',
'text': gistfile['content'],
}
req_render = requests.post('https://api.github.com/markdown',
params=GITHUB_AUTH_PARAMS,
data=unicode(json.dumps(payload)))
if req_render.status_code != 200:
logger.warn('Render {} file {} failed: {}'.format(id, gistfile['filename'], req_render.status_code))
continue
else:
output = smartypants.smartypants(req_render.text)
if format is FORMAT_RST:
rendered = render_rst(gistfile['content'], writer_name='html')['fragment']
output = smartypants.smartypants(rendered)
if output is not None:
gistfile['rendered'] = output
gist['files'].append(gistfile)
rdb.table('gists').insert(gist, upsert=True).run(request.rdbconn)
return gist
def convert_entities(obj):
for key in obj.iterkeys():
if isinstance(obj[key], unicode):
obj[key] = smartypants(obj[key])
obj[key] = obj[key].encode('ascii', 'xmlcharrefreplace')
else:
convert_entities(obj[key])
def prettify_title(title):
"""Sanitizes all (ALL) HTML elements in titles while prettifying the quotes
and dashes used in the titles of threads and categories.
"""
return bleach.clean(\
smartypants(title, "2"),\
tags=[], attributes={})
def convert_entities(obj):
for key in obj.iterkeys():
if isinstance(obj[key], unicode):
obj[key] = smartypants(obj[key])
obj[key] = obj[key].encode('ascii', 'xmlcharrefreplace')
else:
convert_entities(obj[key])
def webfix_unicode(possible_string):
"""
This is ugly but it will create Times-approved HTML
out of terrible cut-and-paste from decision text.
"""
CHAR_MAP = [
(u'\xa7', u'§'),
(u'\u2014', u'—'),
(u'\u2013', u'–'),
(u'\x97', u'—'),
(u'\xa4', u'€'),
(u'\u201c', u'"'),
(u'\u201d', u'"'),
(u'\x96', u'–'),
]
if isinstance(possible_string, basestring):
string = possible_string
string = string.strip()
for char, replace_char in CHAR_MAP:
string = string.replace(char, replace_char)
string = string.decode('utf-8')
string = unicode(string)
string = ftfy.fix_text(string)
string = smartypants.smartypants(string)
return string
return possible_string
def generate(self, markup=None):
"""Generate HTML
>>> class Handler(BaseHandler):
... def _generate(self, markup=None): return markup
>>> handler = Handler(None)
>>> print(handler.generate('foo "bar"'))
foo "bar"
>>> handler.options['smartypants'] = True
>>> print(handler.generate('foo "bar"'))
foo “bar”
"""
if markup is None:
markup = self.markup
html = self._generate(markup)
if self.options.get('smartypants', False):
if not HAS_SMARTYPANTS:
warnings.warn(
"smartypants option is set, "
"but the library isn't installed.", RuntimeWarning)
return html
Attr = smartypants.Attr
html = smartypants.smartypants(html, Attr.set1 | Attr.w)
if self.SUPPORT_EMBED_IMAGES and self.options.get(
'embed_images', False):
html = self.embed_images(html)
return html
def apply_smartypants(self, text, smarty, node):
# Try to be clever about when to use smartypants
if node.__class__ in (docutils.nodes.paragraph,
docutils.nodes.block_quote,
docutils.nodes.title):
return smartypants(text, smarty)
return text
def webfix_unicode(possible_string):
"""
This is ugly but it will create Times-approved HTML
out of terrible cut-and-paste from decision text.
"""
character_map = [
('\xa7', '§'),
('\u2014', '—'),
('\u2013', '–'),
('\x97', '—'),
('\xa4', '€'),
('\u201c', '"'),
('\u201d', '"'),
('\x96', '–'),
]
if isinstance(possible_string, str):
string = possible_string
string = string.strip()
for char, replace_char in character_map:
string = string.replace(char, replace_char)
string = string.decode('utf-8')
string = ftfy.fix_text(string)
string = smartypants.smartypants(string)
return string
return possible_string
def prettify_title(title):
"""Sanitizes all (ALL) HTML elements in titles while prettifying the quotes
and dashes used in the titles of threads and categories.
"""
return bleach.clean(\
smartypants(title, "2"),\
tags=[], attributes={})
def content_transform(name, xmltext, child=None, params={}):
#print("XSLT: %.80s(%s) %r" % (xmltext.replace('\n', ' '), child or '-', params.get('blogmode', '')))
f = BytesIO(xmltext.encode('utf-8'))
try:
doc = etree.parse(f)
except:
print("Text was {!r}".format(xmltext))
raise
if child:
doc = doc.find(child)
params = dict(params)
params.update({
'base': string_param(settings.BASE),
})
html = str(xslt_xform(doc, **params))
# smartypants doesn't handle </a>' properly.
html = re.sub(r"(</\w+>)'", r"\1’", html)
html = smartypants.smartypants(html,
smartypants.Attr.q | smartypants.Attr.n)
#print("Transformed {!r} into {!r}".format(xmltext[:80], html[:80]))
for entry in xslt_xform.error_log:
if entry.filename == '<string>':
fname = name
else:
fname = entry.filename
print("Message, %s @ %d: %s" % (fname, entry.line, entry.message))
return html
def prettify_text(self, text):
"""
Make text more nicerer. Run it through SmartyPants and Widont.
"""
text = self.widont(text)
text = smartypants.smartypants(text)
return text
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
u'The “Green” man'
"""
return _smartypants.smartypants(text)
def publish_pages(self, posts=None, force_publish=False):
if force_publish:
posts = [Post(os.path.join(self.s.PAGES_DIR, f), self.s) for f in os.listdir(self.s.PAGES_DIR) if f.endswith(self.s.MD_EXT)]
for post in posts:
post.content = smartypants(self.md_to_html(post.content))
post.title = smartypants(post.title)
html_fname = "%s%s" % (post.slug, self.s.HTML_EXT)
html_dir = os.path.join(self.s.WWW_DIR, self.s.WWW_PAGES_URL)
html_full_path = os.path.join(html_dir, html_fname)
tools.mkdirp(html_dir)
# TODO: check dir owner/permission
self.write_single_post_to_file(post=post, fname=html_full_path, template=self.s.PAGES_TEMPLATE)
def _smartypants(text):
attrs = (smartypants.Attr.b | smartypants.Attr.D | smartypants.Attr.e)
return (smartypants.smartypants(text, attrs).replace(
"–",
"–").replace("—",
"—").replace("…",
"…").replace("“",
"“").replace("”", "”"))
def _to_smart(verse):
verse = verse.replace(",`",", '")
verse = verse.replace("`","'")
out = smartypants(verse)
parser = HTMLParser()
out = parser.unescape(out)
return out
def clean_typography(self, text):
return smartypants.smartypants(text).\
replace(" ", "").\
replace(" ", " ").\
replace(u'’', u'’').\
replace(u'“', u'“').\
replace(u'”', u'”').\
replace(u'\xa0 ', u' ').replace(u' \xa0', u' ')
def smartypants_wrapper(text):
try:
import smartypants
except ImportError:
from typogrify.filters import TypogrifyError
raise TypogrifyError(
"Error in {% smartypants %} filter: The Python smartypants "
"library isn't installed."
)
else:
attr = smartypants.default_smartypants_attr | smartypants.Attr.w
content = smartypants.smartypants(text, attr=attr)
if isinstance(text, AMPString):
amp_data = text.amp_data
content = AMPString(content)
content.amp_data = smartypants.smartypants(amp_data, attr=attr)
return content
def smartypants_filter(text):
"""
Smarty pants
"""
if text:
return smartypants(text)
else:
return ''
def smartypants_wrapper(text):
try:
import smartypants
except ImportError:
from typogrify.filters import TypogrifyError
raise TypogrifyError(
"Error in {% smartypants %} filter: The Python smartypants "
"library isn't installed.")
else:
attr = smartypants.default_smartypants_attr | smartypants.Attr.w
content = smartypants.smartypants(text, attr=attr)
if isinstance(text, AMPString):
amp_data = text.amp_data
content = AMPString(content)
content.amp_data = smartypants.smartypants(amp_data, attr=attr)
return content
def apply_smartypants(self, text, smarty, node):
# Try to be clever about when to use smartypants
if isinstance(node,
(docutils.nodes.paragraph, docutils.nodes.block_quote,
docutils.nodes.title)):
smarty = _str_attr_to_int(smarty)
return smartypants(text, smarty)
return text
def smartypants_filter(text):
"""
Smarty pants
"""
if text:
return smartypants(text)
else:
return ''
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
u'The “Green” man'
"""
return _smartypants.smartypants(text)
def _to_smart(verse):
verse = verse.replace(",`", ", '")
verse = verse.replace("`", "'")
out = smartypants(verse)
parser = HTMLParser()
out = parser.unescape(out)
return out
def clean_typography(text):
return smartypants.smartypants(text).\
replace(" ", " ").\
replace(" ", " ").\
replace(u'’', u'’').\
replace(u'“', u'“').\
replace(u'”', u'”').\
replace(u'\xa0 ', u' ').replace(u' \xa0', u' ')
def get(self):
urls = self.get_query_arguments('url')
if urls and len(urls) == 1:
url = urls[0]
doc = Document(requests.get(url).text)
self.write(smartypants(doc.summary()))
self.write(STYLE)
else:
self.write("Please provide ?url=[your-url]")
def about():
context = make_context()
f = codecs.open("posts/intro.md", mode="r", encoding="utf-8")
contents = f.read()
html = markdown.markdown(smartypants(contents))
context['markdown'] = Markup(html)
return render_template('post.html', **context)
def about():
context = make_context()
f = codecs.open("posts/intro.md", mode="r", encoding="utf-8")
contents = f.read()
html = markdown.markdown(smartypants(contents))
context['markdown'] = Markup(html)
return render_template('post.html', **context)
def smartquotes(text):
"""Applies smarty pants to curl quotes.
>>> smartquotes('The "Green" man')
u'The “Green” man'
"""
text = unicode(text)
output = smartypants.smartypants(text)
return output
def get(self):
urls = self.get_query_arguments('url')
if urls and len(urls) == 1:
url = urls[0]
doc = Document(requests.get(url).text)
self.write(smartypants(doc.summary()))
self.write(STYLE)
else:
self.write("Please provide ?url=[your-url]")
def smartquotes(text):
"""Applies smarty pants to curl quotes.
>>> smartquotes('The "Green" man')
u'The “Green” man'
"""
text = unicode(text)
output = smartypants.smartypants(text)
return output
def fileToSentenceList(pathToTextFile):
# Import string from file
file = io.open(pathToTextFile, mode='r', buffering=-1, encoding=None, errors=None, newline=None, closefd=True)
rawString = file.read().strip()
# Basic cleaning: Replace line breaks with spaces
def removeLineBreaks(string):
cleanString = re.sub("[\n\r]+", " ", string) # Linke breaks to spaces
cleanString = re.sub("\s{2,}", " ", cleanString) # Remove double spaces
return cleanString
cleanString = removeLineBreaks(rawString);
# Use nltk to tokenize sentences
# See http://www.nltk.org/api/nltk.tokenize.html#module-nltk.tokenize
sentences = sent_tokenize(cleanString)
# Look at all the sentences and throw out things that we don't like
buffer = sentences
sentences = []
for sentence in buffer:
# 1.
# Throw out words that don't begin w/ capital letter (happens often after direct speech).
# These are correct sentences but I prefer not to have them in the pool because they make little
# sense without context.
regex = '^[\s({\["\'“‘\-«»‹›]*[A-ZÄ-Ü0-9]'
match = re.match(regex, sentence)
if match is None:
#print "thrown out b/c sentence doesn't start w/ capital letter: ", sentence
continue
# 2.
# Throw out one-word or two-word sentences that contain numbers
# They are probably headlines: 'Chapter 2.' or '1.F.1.'
if sentence.count(" ") < 2 and re.search("\d", sentence) is not None:
#print "thrown out b/c it seems like a nonsensical headline:", sentence
continue
# Remove white-space at the beginning and end
sentence = sentence.strip()
# Use typographically correct quotation marks, apostrophes and dashes
sentence = HTMLParser().unescape(smartypants.smartypants(sentence))
# Avoid unclosed (or unopened) quotation marks, parentheses, brackets, braces
sentence = complete_pairs(sentence)
sentences.append({
'sentence': sentence,
'numberOfWords': sentence.count(' ') + 1,
'file': pathToTextFile,
'randomPoint': [random.random(), 0] # For efficient random entry retrieval. See http://stackoverflow.com/a/9499484/836005
})
return sentences
def download_story(story_id):
# TODO: probably use {'drafts': 0, 'include_deleted': 0}
storyinfo = session.get(API_STORYINFO + story_id, params={'drafts': 1, 'include_deleted': 1}).json()
story_title = storyinfo['title']
story_description = storyinfo['description']
story_createDate = dateutil.parser.parse(storyinfo['createDate'])
story_modifyDate = dateutil.parser.parse(storyinfo['modifyDate'])
story_author = storyinfo['user']['name']
story_categories = [categories[c] for c in storyinfo['categories'] if c in categories] # category can be 0
story_rating = storyinfo['rating'] # TODO: I think 4 is adult?
story_cover = io.BytesIO(session.get(storyinfo['cover']).content)
story_url = storyinfo['url']
print('Story "{story_title}": {story_id}'.format(story_title=story_title, story_id=story_id))
# Setup epub
book = ez_epub.Book()
book.title = story_title
book.authors = [story_author]
book.sections = []
book.impl.addCover(fileobj=story_cover)
book.impl.description = HTML(story_description, encoding='utf-8') # TODO: not sure if this is HTML or text
book.impl.url = story_url
book.impl.addMeta('publisher', 'Wattpad - scraped')
book.impl.addMeta('source', story_url)
for part in storyinfo['parts']:
chapter_title = part['title']
if part['draft']:
print('Skipping "{chapter_title}": {chapter_id}, part is draft'.format(chapter_title=chapter_title, chapter_id=chapter_id))
continue
if 'deleted' in part and part['deleted']:
print('Skipping "{chapter_title}": {chapter_id}, part is deleted'.format(chapter_title=chapter_title, chapter_id=chapter_id))
continue
chapter_id = part['id']
# TODO: could intelligently only redownload modified parts
chapter_modifyDate = dateutil.parser.parse(part['modifyDate'])
print('Downloading "{chapter_title}": {chapter_id}'.format(chapter_title=chapter_title, chapter_id=chapter_id))
chapter_html = session.get(API_STORYTEXT, params={'id': chapter_id, 'output': 'json'}).json()['text']
chapter_html = smartypants.smartypants(chapter_html)
section = ez_epub.Section()
section.html = HTML(chapter_html, encoding='utf-8')
section.title = chapter_title
book.sections.append(section)
print('Saving epub')
book.make('./{title}'.format(title=book.title.translate(ILLEAGAL_FILENAME_CHARACTERS)))
def apply_smartypants(self, text, smarty, node):
# Try to be clever about when to use smartypants
if isinstance(node, (
docutils.nodes.paragraph,
docutils.nodes.block_quote,
docutils.nodes.title
)):
smarty = _str_attr_to_int(smarty)
return smartypants(text, smarty)
return text
def smarty_pants(self, text):
"""Applies smarty pants to html text"""
# Try to load smartypants
try:
import smartypants
return smartypants.smartypants(text)
except ImportError:
pass # this should be logged maybe??? Right now, silently ignored
return text
def add(self, posts):
"""Add the posts and generate a blog list."""
li_html = []
for post in posts:
# Put the smartified title back into the post.
post.title = title = smartypants.smartypants(post.title)
li_html.append(
u'<li><a href="{route}">{title}</a></li>'.format(
route=post.route, title=title))
self._blog_list = u'\n'.join(li_html)
self._posts = posts
def parse_post(src_path):
_post = {}
with codecs.open(src_path, 'r', 'utf-8') as src:
raw_title = src.readline().strip()
smarty_title = smartypants.smartypants(raw_title)
_post['title'] = flask.Markup(smarty_title)
_post['title_first_char'] = first_alpha_char(raw_title).upper()
_ = src.readline().strip()
while True:
meta = src.readline().strip()
if not meta:
break
if ':' in meta:
meta_key, sep, meta_val = meta.partition(':')
_post[meta_key.lower().strip()] = meta_val.strip()
mkd = markdown.markdown(src.read())
_post['body'] = smartypants.smartypants(mkd)
_post['slug'] = get_slug_from_path(src_path)
_post['published'] = parse_datetime(_post['published'])
_post['year'] = _post.get('published').year
_post['url'] = get_post_url(_post)
return _post
def smart_dict(self):
payload = self.dict()
for key,value in payload.items():
if value:
try:
payload[key] = smartypants.smartypants(value.strip())
except TypeError:
pass
except UnicodeError:
pass
except AttributeError:
pass
return payload
def parse_post(src_path):
_post = {}
with codecs.open(src_path, 'r', 'utf-8') as src:
raw_title = src.readline().strip()
smarty_title = smartypants.smartypants(raw_title)
_post['title'] = flask.Markup(smarty_title)
_post['title_first_char'] = first_alpha_char(raw_title).upper()
_ = src.readline().strip()
while True:
meta = src.readline().strip()
if not meta:
break
if ':' in meta:
meta_key, sep, meta_val = meta.partition(':')
_post[meta_key.lower().strip()] = meta_val.strip()
mkd = markdown.markdown(src.read())
_post['body'] = smartypants.smartypants(mkd)
_post['slug'] = get_slug_from_path(src_path)
_post['published'] = parse_datetime(_post['published'])
_post['year'] = _post.get('published').year
_post['url'] = get_post_url(_post)
return _post
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
u'The “Green” man'
"""
try:
import smartypants
except ImportError:
logger.error("Error in {% smartypants %} filter: The Python smartypants library isn't installed.")
return text
output = smartypants.smartypants(text)
return jinja2.Markup(output)
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
'The “Green” man'
"""
try:
import smartypants
except ImportError:
raise TypogrifyError("Error in {% smartypants %} filter: The Python smartypants library isn't installed.")
else:
output = smartypants.smartypants(text)
return output
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
u'The “Green” man'
"""
try:
import smartypants
except ImportError:
return text
else:
output = smartypants.smartypants(text)
return output
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
'The “Green” man'
"""
try:
import smartypants
except ImportError:
raise TypogrifyError("Error in {% smartypants %} filter: The Python smartypants library isn't installed.")
else:
output = smartypants.smartypants(text)
return output
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
u'The “Green” man'
"""
try:
import smartypants
except ImportError:
logger.error("Error in {% smartypants %} filter: The Python smartypants library isn't installed.")
return text
output = smartypants.smartypants(text)
return jinja2.Markup(output)
def munge_feed(items):
print("++++++++++\nIn munge_feed module ...")
for post in items:
post['title_api'] = smartypants.smartypants(post['title_api'].strip())
post['caption_api'] = smartypants.smartypants(
post['caption_api'].strip())
regex = re.compile(r"^.*\|\|", re.IGNORECASE)
post['categories_api'] = list(
set([regex.sub('', x.lower()) for x in post['categories_api']]))
if post['sections_api']:
post['sections_api'] = [x.lower() for x in post['sections_api']]
post['desc_api'] = smartypants.smartypants(post['desc_api'].strip())
post['desc_api'] = " ".join(post['desc_api'].split())
date_object = datetime.datetime.strptime(post['pubdate_api'],
'%Y-%m-%dT%H:%M:%S')
post['timestamp'] = date_object.strftime('%b %d %I:%M %p')
post['timestamp'] = post['timestamp'].replace(' 0', ' ').replace(
'Jul', 'July').replace('Apr',
'April').replace('Mar', 'March').replace(
'Jun', "June").replace(':00', '')
post['timestamp_epoch'] = int(
(date_object - datetime.datetime(1970, 1, 1)).total_seconds())
if "opinion" in post['sections_api']:
label_start = "OPINION"
else:
label_start = ""
if post['source_api']:
label_end = post['source_api']
else:
label_end = post['site_api']
post['label_api'] = (
(label_start + ' | ') * str_len_check(label_start)) + (
(post['author_api'] + ' | ') *
str_len_check(post['author_api'])) + label_end
post['label_api'] = post['label_api'].replace(
"The Hamilton Spectator",
"The Spec").replace("Hamilton Spectator",
"The Spec").replace("Toronto Star", "The Star")
return items
def parse_form(form_data, kind="list"):
db = TinyDB(cfg.config['db_name'])
Record = Query()
print("incoming form data:")
# print(form_data)
# print("converted to a dict")
print(dict(form_data))
# form data will have keys, values that may be lists or a single string.
form_data_dict = dict(form_data)
if kind == 'list':
# form data is coming from the 'lineup' page,
# which can have multiple changes on multiple assets
for k, v in form_data_dict.items():
if k != "action":
if isinstance(v, list):
# it's a list of strings.
for item in v:
# check if empty string
if item:
asset_id, field, new_value = item.split('__')
print(
f"++++++++\nSetting this item: {asset_id} to {field}: {new_value}\n++++++++"
)
db.update({field: int(new_value)},
Record.asset_id == asset_id)
else:
# check if empty string
if v:
asset_id, field, new_value = v.split('__')
print(
f"++++++++\nSetting this item: {asset_id} to {field}: {new_value}\n++++++++"
)
db.update({field: int(new_value)},
Record.asset_id == asset_id)
else:
# form data is coming from the 'item' page instead,
# mutiple changes possible but only 1 asset affected
post_update = {}
asset_id = form_data_dict['asset_id'][0]
for x in ['draft_user', 'rank', 'rank_time']:
if form_data_dict[x][0] != '':
post_update[x] = int(form_data_dict[x][0])
for x in ['label_user', 'title_user', 'desc_user']:
if form_data_dict[x][0] != '':
post_update[x] = smartypants.smartypants(
form_data_dict[x][0].strip())
print("Data to update:")
print(post_update)
db.update(post_update, Record.asset_id == asset_id)
db.close()
return
def smarty_filter(s):
"""
Filter to smartypants strings.
"""
if type(s) == 'Markup':
s = s.unescape()
# Evaulate COPY elements
if type(s) is not unicode:
s = unicode(s)
s = smartypants(s)
return Markup(s)
def smartypants(text):
"""Applies smarty pants to curl quotes.
>>> smartypants('The "Green" man')
'The “Green” man'
"""
try:
import smartypants
except ImportError:
raise TypogrifyError("Error in {% smartypants %} filter: ")
else:
from smartypants import Attr
attr = Attr.set1 & (~(Attr.mask_d))
output = smartypants.smartypants(text, attr)
return output
def parse_transcript(path):
"""
Parse a Premiere .tsv file, calculate total seconds and write to JSON for use in the app.
"""
data = {'subtitles': []}
filename, ext = path.split('/')[-1].split('.')
with codecs.open(path, 'rb', encoding='utf16') as f:
transcript = f.read().encode('utf-8')
tab_reader = csv.reader(BytesIO(transcript), delimiter='\t')
headers = tab_reader.next()
for row in tab_reader:
# Premiere exports kind of suck
if row[0] == '':
words = smartypants(row[1].strip())
time_str = row[2]
else:
words = smartypants(row[0].strip())
time_str = row[1]
hours, minutes, seconds, frame = [
int(x) for x in time_str.split(':')
]
decimal = (float(frame) / 24)
total_seconds = (hours * 3600) + (minutes * 60) + (seconds +
decimal)
segment = {
'time': total_seconds,
'transcript': words,
}
data['subtitles'].append(segment)
with open('www/data/%s.json' % filename, 'w') as wf:
wf.write(json.dumps(data))