def new_markdown(item_type, title=None, **kwargs):
# Some defaults
now = datetime.now()
# FIXME Timezones are hard. Hard coding for now.
now = pytz.timezone('America/New_York').localize(now)
text = ""
metas = ['itemtype: %s' % item_type, 'guid: %s' % uuid.uuid4()]
for key in kwargs:
metas.append('%s: %s' % (key, kwargs[key]))
if 'created' not in kwargs:
metas.append('updated: %s' % now.isoformat())
if 'updated' not in kwargs:
metas.append('updated: %s' % now.isoformat())
if 'category' not in kwargs:
metas.append('category: uncategorized')
text = text + '''
For SEO, please assign a keyword-rich category like "keyword/seo" above.
It will be used to generate a URL.
'''
if title:
metas.append('slug: %s' % slugify(title))
metas.append('title: %s' % title)
else:
metas.append('slug:')
metas.append('title:')
text = text + '''
You need to set a title and slug. Slug will be used to generate a URL.
'''
return "\n".join(metas) + "\n\n" + text
def to_archetype(archivist, text):
"Given text in markdown format, returns a dict of metadata and body text."
timezone = archivist.siteconfig.get('timezone', pytz.utc)
html = md.convert(text)
metadata = md.Meta
itemmeta = {}
catalogmeta = {}
# Here we implement some special case transforms for data that may need
# cleanup or is hard to encode using markdown's simple format.
for key, value in metadata.items():
if key in ['created', 'date', 'published', 'updated']:
# because humans are sloppy, we parse and normalize date values
dt = parse_date(value[0])
if dt.tzinfo:
dt = dt.astimezone(timezone)
else:
dt = timezone.localize(dt)
if key == 'date': # Legacy DC.date, convert to specific
key == 'published'
itemmeta[key] = dt.isoformat()
elif key == 'itemtype':
itemmeta['itemtype'] = string.capwords(metadata['itemtype'][0], '/')
elif key == 'author':
itemmeta["attribution"] = [{"role": "author", "name":
metadata[key][0]}]
elif key == "copyright": # Typical usage provides only notice
itemmeta["rights"] = {"copyright_notice": metadata[key][0]}
elif key.startswith("rights-"):
if "rights" not in itemmeta:
itemmeta["rights"] = {}
itemmeta["rights"][key[7:]] = metadata[key][0]
elif key == "category": # Typical usage provides only name
itemmeta["category"] = {"name": metadata[key][0]}
elif key.startswith("category-"):
if "category" not in itemmeta:
itemmeta["category"] = {}
itemmeta["category"][key[9:]] = metadata[key][0]
elif key.startswith("query-"):
if "query" not in catalogmeta:
catalogmeta["query"] = {}
catalogmeta["query"][key[6:]] = metadata[key][0]
else:
# reads everything as list, but most values should be scalar
itemmeta[key] = value[0] if len(value) == 1 else value
# itemmeta needs a self-referencing archetype key.
key = archivist.pathstrategy.path_for(resourcetype='archetype', **itemmeta)
self_link = {
"bucket": archivist.bucket,
"key": key,
"href": key
}
itemmeta['archetype'] = self_link
itemmeta['contenttype'] = "text/html; charset=utf-8"
if "itemtype" not in itemmeta:
itemmeta["itemtype"] = "Item/Page/Article"
if "slug" not in itemmeta:
itemmeta["slug"] = slugify(itemmeta["title"])
if "published" not in itemmeta:
itemmeta["published"] = itemmeta["updated"]
if "updated" not in itemmeta:
itemmeta["updated"] = itemmeta["published"]
# Some metadata has fallback defaults from the siteconfig
if "rights" not in itemmeta:
itemmeta["rights"] = archivist.siteconfig.get("rights")
if "attribution" not in itemmeta:
itemmeta["attribution"] = archivist.siteconfig.get("attribution")
archetype = {"Item": itemmeta, "Item_Page_Article": {"body": html}}
if itemmeta['itemtype'].startswith('Item/Page/Catalog'):
catalogmeta['query'] = fixup_query(catalogmeta['query'])
archetype['Item_Page_Catalog'] = catalogmeta
return archetype
